JP5382601B2 - Information presenting apparatus, information presenting method, and information presenting program - Google Patents

Description

  The present invention relates to an information presentation device, an information presentation method, and an information providing program, and in particular, an information presentation device that provides information while estimating a user's situation from a user's search operation history and the like, an information presentation method, and an information presentation method Regarding the program.

  In an information providing apparatus in which a large amount of information is stored, in order to search for desired content, the user inputs a search query as a search query to narrow down the content, and has a function of generating and presenting a content list. Yes. If the desired content cannot be found by entering a single search query because the search intent could not be accurately made, change the search query multiple times and look at the content list each time. Search for content. A series of search operations until a user starts content search for a certain purpose and obtains content according to the search purpose is called a search session. In using the information providing service, the number of search sessions is not limited to one, and a plurality of search sessions are often combined.

  Non-Patent Document 1 and Patent Document 1 have been reported as examples in which a search session is determined in an information providing apparatus and used for customization of the information providing apparatus.

  Non-Patent Document 1 reports a technique for defining the degree of association between words included in a search query by determining a search session. This is based on the hypothesis that the search query used in the same search session is a phrase that expresses the same search purpose from multiple viewpoints. Is a technique for defining synonyms by associating words included in a phrase.

  Also, in Patent Document 1, by determining a search session, content once displayed during a search session is excluded from the content list or placed at the end of the content list when re-searched in the same search session. Technology has been reported. In general, a search query used in a search session is often input that is semantically similar. For this reason, even if a search query is changed and a content list is presented, the displayed content list includes a large amount of content already presented in the previous content list. The user must browse the content list that contains a large amount of content that has already been determined to be unnecessary, even though the user has changed the search query because he / she does not like the content seen in the previous content list. There were oversight and giving up of browsing the list. Patent Document 1 aims to improve the above-mentioned problem, and records the content displayed while the search session continues, and already generates a content list when a new search query is input. Remove the displayed content from the content list or move to the end of the content list.

  In order to realize an information providing apparatus represented by Non-Patent Document 1 and Patent Document 1, a technique for determining a search session is indispensable, and a method for determining a search session has been reported for each document.

  In Non-Patent Document 1, a search session is determined based on a search operation time interval. This is described in p. The left column of 2252, “Search terms used by the same user are likely to be used to obtain the same information if the use time interval is short and to obtain different information if the use time interval is short.” Based on a hypothesis, a search session is discriminated by inferring the relation between operations by a monotonically decreasing function with time as an input variable.

In Patent Document 1, based on the hypothesis that content that is hit by a search is not greatly changed for the same search purpose, a content set that is searched by the immediately preceding search query and a content set that is searched by a new search query Are compared, it is determined that the search session is continuing when the ratio of the common content is large, and it is determined that the search session is shifted to another search when the ratio of the common content is small.
Okubo et al., “Extracting Information Needs Based on WWW Search Logs”, Transactions of Information Processing Society of Japan, Vol. 39, no. 7, issued in July 1998, pp. 2250-2258 JP 2006-285656 A

  However, the search session determination methods in Non-Patent Document 1 and Patent Document 1 have a problem in that there are many cases where a search session cannot be properly determined.

  The search session determination method of Non-Patent Document 1 is based on the premise that the operation of the information presentation device by the user is continuously performed. However, considering the user's operation scene, it is easy to imagine that the search operation will be interrupted or resumed by taking a break during the operation or causing another interruption work such as telephone reception or mail processing. Is done. In the search session determination method of Non-Patent Document 1, since the search session is determined based on the search operation time interval, the time interval becomes longer before and after the search operation is interrupted. As a result, the search session cannot be correctly determined. This is called the first problem.

  Further, in the search session determination method of Patent Document 1, since the search session is not determined based on the search time, the first problem does not occur, and the search session is determined without being affected by the interruption of the search operation. It is possible. However, when the search query is input three times or more, the identity of the search purpose cannot be correctly determined. A case where the determination cannot be performed correctly will be described with reference to FIG.

  FIG. 35 shows a set of contents searched by a search query when three searches are performed. A user inputs a search query in the order of a search query Q1, a search query Q2, and a search query Q3, and sets a set of contents searched by each search query as a content set 101, a content set 102, and a content set 103. The content set 101 and the content set 102 are largely overlapped, and the content set 101 and the content set 103 are greatly overlapped, but the content set 102 and the content set 103 are hardly overlapped. When the search session is determined at the timing when the search is performed with the search query Q3, the method of Patent Document 1 calculates the ratio of content common to the content set 102 in the search query Q2 and the content set 103 in the search query Q3. Therefore, it is determined that the search sessions are different, and the search session cannot be correctly determined. This is called the second problem.

  An object of the present invention is to provide an information presentation apparatus capable of correctly determining a search session even when a search is repeatedly performed.

In order to achieve the above object, a first information presentation device according to the present invention includes a content set searched by a search query specified by a user search operation and a search query specified by the user's past search operation. The specified search query is specified by a search operation for the same search purpose as the user's past search operation by comparing with one content set synthesized using a plurality of content sets searched in It has the means to discriminate | determine whether it is.

The second information presenting apparatus according to the present invention includes a first content set storage unit that holds a content set searched by a search query specified by a user search operation, and is specified by a past search operation of the user. Content set synthesizing means for synthesizing into one content set using a plurality of content sets searched by the search query, second content set storage means for holding the content set generated by the content set synthesizing means, a content set of one content set storage unit, by comparing the contents set in the second content set storage means, whether the given search query belonging to a search session past search operation the same search intent of the user And a search session determining means for determining

The third information presentation apparatus according to the present invention includes a search query storage unit that stores a search query input by a user's search operation, a content search unit that searches for content according to the search query, and the search query storage unit A first content set storage means for holding a content set searched by the content search means using a search query stored in the search query history storage means for storing a search query input by a past search operation of the user Content set synthesizing means for synthesizing into one content set using a plurality of content sets searched by the content search means using past search queries stored in the search query history storage means, and the content set synthesis Second content set storage that holds content sets generated by means Stage and the content set in the first content set storage unit, identical to the previous search operation of the second content set storage means search query the user stored in the search query storing means by comparing the contents set of Search session determination means for determining whether the search session belongs to a search session for search purposes.

  In the first information presentation method according to the present invention, a content set searched by a search query specified by a user's search operation is stored in a first content set storage unit, and the search is performed by the user's past search query. A content set is stored in the second content set storage means, the content set in the first content set storage means, the content set in the second content set storage means, And determining whether the specified search query belongs to a search session for the same search purpose as the user's past search operation.

  In the second information presentation method according to the present invention, the search query input by the user's search operation is stored in the search query storage means, and the content is searched according to the search query stored in the search query storage means. The retrieved content set is stored in the first content set storage means, the user's past search query is stored in the search query history storage means, and the search is performed using the past search query stored in the search query history storage means. The content set is synthesized using the set of content sets, the synthesized content set is stored in the second content set storage unit, the content set of the first content set storage unit, and the content set of the second content set storage unit And the search query stored in the search query storage means And judging whether belonging to the search sessions of search operations and the same search intent.

  The first information presentation program according to the present invention includes a process of storing a content set searched by a search query specified by a user search operation in a first content set storage unit, and a past search query of the user. A process of combining the content set using the content set searched in the step, a process of storing the combined content set in the second content set storage means, the content set of the first content set storage means, and the second content Comparing the content set of the set storage means and determining whether or not the specified search query belongs to a search session for the same search purpose as the user's past search operation. And

  Further, the second information presentation program according to the present invention includes a process for storing a search query input by a user search operation in a search query storage unit, and a search for content according to the search query stored in the search query storage unit. Storing the searched content set in the first content set storage means, storing the user's past search query in the search query history storage means, and storing the search query history storage means in the search query history storage means Processing for combining a content set using a content set searched using the user's past search query, processing for storing the combined content set in second content set storage means, and the first content set storage means And the content set of the second content set storage means are compared. , Characterized in that to execute a process of determining whether it belongs to search session past search operation the same search object of the search query stored in the search query storage means the user computer.

  According to the present invention, even when the search is repeatedly performed, the search session can be correctly determined. The reason is that the user's search intention can be accurately estimated by synthesizing a set of contents obtained as a result of a plurality of search queries used by the user, and the search session is determined based on the estimated user's search intention. Is to do.

  Hereinafter, embodiments of the present invention will be specifically described with reference to the drawings.

  In the information presentation device according to the first embodiment of the present invention, a first content set storage unit that holds a content set searched by a specified search query and a search by a plurality of past search queries by at least one or more. A content set synthesizing unit that synthesizes a content set using a plurality of content sets at least one, a second content set storage unit that holds the content set generated by the content set synthesizing unit, and a first content set A search session determination unit that compares the content set of the storage unit with the content set of the second content set storage unit to determine whether the specified search query belongs to a past search session, Multiple content sets used for composition by the composition means are included in the same search session. A plurality of content sets retrieved by the search query.

  Adopting such a configuration, when determining whether the specified search query belongs to the previous search session, it does not depend on only the most recent past search query, and the operation time interval The object of the present invention can be achieved by making a search session determination by comprehensively using the search queries included in the search session.

  In the information presentation device according to the second embodiment of the present invention, in addition to the configuration of the first information presentation device, the history of the content set synthesized by the content set synthesis means is stored in association with the history of the search query. And a synthesis history storage means.

  By adopting such a configuration and storing the content set synthesized by the content set synthesizing unit once in the synthesis history storage unit, the first implementation is performed while reducing the content synthesis processing by the content set synthesizing unit. The effect by this form can be obtained, and the object of the present invention can be achieved.

  In the information presentation device according to the third embodiment of the present invention, in addition to the configurations of the first and second information presentation devices, a search result history for storing a content set searched for content in association with a search query Storage means.

  By adopting such a configuration, the effects of the first and second embodiments can be obtained while reducing the content search processing, and the object of the present invention can be achieved.

  In the information presentation device according to the fourth exemplary embodiment of the present invention, in addition to the configurations of the first, second, and third information presentation devices, when the search session determination unit determines a plurality of search sessions. And a comprehensive determination means for comprehensively determining the plurality of determination results and outputting a final search session determination result.

  By adopting such a configuration, it is possible to hold information related to a plurality of search sessions and comprehensively determine which search session includes the search query to be determined. Can be achieved.

  In the information presentation apparatus according to the fifth embodiment of the present invention, in addition to the configurations of the first, second, third, and fourth information presentation apparatuses, the user is based on the determination result in the search session determination unit. Presentation information generating means for generating information to be presented to the computer.

  By adopting such a configuration, it is possible to determine a search session using a search query input from the user, and generate information to be presented to the user according to the determined search session, thereby achieving the object of the present invention. can do.

  In the information presentation device according to the sixth exemplary embodiment of the present invention, in the configuration of the fifth information presentation device, a presentation history management unit that manages a content information presentation history in the presentation information generation unit, and the presentation history Content list generating means for preferentially displaying contents not yet displayed with reference to the presentation history of each content stored in the management means.

  Adopting such a configuration, sorting the content based on the content presentation history, lowering the priority of content that has been presented under different content selection conditions, and increasing the priority of unpresented content In addition, the object of the present invention can be achieved more effectively.

  In the information presentation device according to the seventh embodiment of the present invention, in addition to the configurations of the first, second, third, fourth, and fifth information presentation devices, the number of search sessions used for each user is set. Search session counting means for counting.

  By adopting such a configuration and counting the number of search sessions used for each user, it is possible to perform processing in accordance with the usage status of the user and achieve the object of the present invention.

  FIG. 1 is a block diagram showing the configuration of the first embodiment of the present invention.

  The information presentation apparatus according to the present embodiment includes means for determining a series of search operations of a user with reference to a plurality of search queries. Specifically, the search specified by the user is performed by comparing the content set searched by the search query specified by the user with the content set synthesized using the content set searched by the user's past search query. It is determined whether or not the query has the same purpose as the user's past search operation.

  Specifically, as shown in FIG. 1, search query storage means 1, content search means 2, first content set storage means 3, search query history storage means 4, content set composition means 5, second content set storage means. 6 and search session determination means 7.

  The search query storage unit 1 stores a search query input by a user. The content search unit 2 searches the content according to the search query storage unit 1 and the search query. The first content set storage unit 3 holds the content set searched by the content search unit 2 using the search query stored in the search query storage unit 1. The search query history storage unit 4 stores past search queries. The content set synthesizing unit 5 uses at least one or more content sets searched by the content search unit 2 using at least one or more search queries stored in the search query history storage unit 4 and uses the plurality of content sets. Is synthesized. The second content set storage unit 6 holds the content set generated by the content set combining unit 5. The search session determination unit 7 compares the content set in the first content set storage unit 3 with the content set in the second content set storage unit 6 and stores the search query stored in the search query storage unit 1 in the past search session. Determine if it belongs.

  FIG. 2 is a flowchart showing the operation of the information presentation apparatus according to the present embodiment. The operation of this embodiment will be described with reference to FIG. In this description, the user performs n searches in the same search session, and the search queries Q1, Q2,. . . , Qn (n> = 1) will be described from the state stored in the search query history storage unit 4. Furthermore, for simplification of description, the search query history storage means 4 will be described assuming that only search queries belonging to the same search session are stored.

(Step S1)
In step S1, the search query storage means 1 stores a search query Qn + 1 for which a search session is to be determined. Then, it progresses to step S2.

(Step S2)
In step S2, a search is performed by the content search means 2 using the search query Qn + 1, and a content set Cn + 1 is acquired as a search result. Here, the content set information acquired from the content search means 2 is assumed to be an ID (Identity number) or URI (Uniform Resource Identifier) assigned to uniquely identify the content. The information for identifying the content or the content itself may be acquired and held. The acquired content set Cn + 1 is stored in the first content set storage means 3. Then, it progresses to step S3.

(Step S3)
In step S3, each search query Qi (1 ≦ i ≦ n) stored in the search query history storage unit 4 is searched in the content search unit 2, and as a result, a content set Ci (1 ≦ i ≦ n) is obtained. get. The acquired content set Ci (1 ≦ i ≦ n) is sent to the content set combining means 5. Then, it progresses to step S4.

(Step S4)
In step S4, the content set synthesis means 5 synthesizes the content set using the content set Ci (1 ≦ i ≦ n). Various methods are conceivable, but here, as an example, by calculating the union of Ci (1 ≦ i ≦ n), the combined content set D [C1,. . . , Cn] (= C1∪ ... ∪Cn). The synthesized content set D [C1,. . . , Cn] is stored in the second content set storage means 6. Then, it progresses to step S5.

(Step S5)
In step S5, the content set Cn + 1 stored in the first content set storage unit 3 by the search session determination unit 7 and the content set D [C1,. . . , Cn]. Various methods can be considered as the comparison method. Here, as an example, the product set Cn + 1∩D [C1,. . . , Cn] includes the number of contents | Cn + 1∩D [C1,. . . , Cn] | and the number of contents | Cn + 1 | contained in the first content set Cn + 1, the evaluation value E = | Cn + 1∩D [C1,. . . , Cn] | / | Cn + 1 | Then, it progresses to step S6.

(Step S6)
In step S6, if the evaluation value E is large, the contents of the search query Qn + 1 and the previous search queries Q1,. . . , Qn can be determined to be similar, and if the evaluation value E is small, the content of the search query Qn + 1 and the previous search queries Q1,. . . , Qn can be determined to be different. Based on this, when the evaluation value E is equal to or greater than the threshold value α (0 ≦ α ≦ 1), the search session determination unit 7 determines that the search session is the same. On the other hand, when the evaluation value E is less than the threshold value α, it is determined that the search session is different. If it is determined that the search session is the same, the process proceeds to step S8. If it is determined that the search session is different, the process proceeds to step S7.

(Step S7)
In step S7, since the search session is changed, the search query Qi (1 ≦ i ≦ n) stored in the search query history storage unit 4 is deleted. Then, it progresses to step S8.

(Step S8)
In step S8, the search query history storage means 4 stores the search query Qn + 1.

  In the above description, the content search unit 2 performs content search using each search query stored in the search query history storage unit 4 in step S3, and the content searched by each search query in step S4. It describes that the content sets to be stored in the second content set storage means 6 are combined by combining the sets. Not limited to this, for example, by combining step S3 and step S4, a search query for searching for a content set to be stored in the second content set storage unit 6 is synthesized from a search query stored in the search query storage unit 4 Then, an operation may be performed in which the content search unit 2 searches for the content set using the combined search query and the content set is stored in the second content set storage unit 6.

  Also, in this embodiment, the evaluation value E calculated in step S5 increases the value of the search query Qn + 1 and the previous search queries Q1,. . . , Qn is determined to be similar. Other implementation methods include, for example, a) content set Cn + 1 and content set C1Ｃ. . . Content set (Cn + 1) XOR (C1∪ ... ∪Cn) obtained by exclusive OR with ∪Cn, b) Content set Cn + 1 and content set C1∪. . . The union C1∪ of ∪Cn. . .評 価 Cn∪Cn + 1 and the evaluation value E = | (Cn + 1) XOR (C1∪ ... ∪Cn) | / | C1∪. . . It may be defined as ∪Cn と Cn + 1 |. In this case, as the evaluation value E is smaller, the contents of the search query Qn + 1 and the previous search queries Q1,. . . , Qn is determined to be similar. Furthermore, other implementation methods may be used.

  Thus, in the search session determination in the information presentation apparatus according to the present embodiment, the search query included in the search session does not depend only on the most recent past search query and does not depend only on the operation time interval. Since the search session is determined by comprehensively using, the plurality of search queries so far can be comprehensively determined, and the identity determination for the purpose of search can be performed more accurately.

[Application Example 1-1]
A first application example in this embodiment will be described.

  FIG. 3 shows an example of a search query stored in the search query history storage unit 4 in this embodiment. In the example of the search query illustrated in FIG. 3, the user performs three searches as a search query in the order of search query Q1 = “security”, search query Q2 = “security & vulnerability”, search query Q3 = “vulnerability & diagnosis”. Assume that a query is specified and the search query history information 110 is stored in the search query storage unit 4.

  Note that the search query history information 110 shown in FIG. 3 is an example of search query information stored in the search query storage unit 4, and may be in other formats as long as it stores the search query history. It doesn't matter.

  Here, when a search query Q4 = “security & distribution” is input as the fourth search query by the user, the search session determination is performed, and subsequently, the search query Q5 is set as the fifth search query. = Operation of performing a search session determination when a search query of “broadband” is input will be described with reference to the flowchart of FIG.

  First, a case where the search query Q4 = “security & distribution” is input as the fourth search query will be described.

(Step S1): First Time In step S1, the search query storage means 1 stores a search query Q4 = “security & distribution” for which a search session is to be determined. Then, it progresses to step S2.

(Step S2): First time In step S2, the content search unit 2 performs a search with the search query Q4 = “security & distribution”, and acquires a content set C4 as a search result. The content set C4 is stored in the first content set storage means 3. Then, it progresses to step S3.

(Step S3): First time In Step S3, the search query Q1 = “security”, search query Q2 = “security & vulnerability”, search query Q3 = “vulnerability & diagnosis” stored in the search query history storage means 4 ”In the content search means 2 and the content set C1 as the search result of the first search query Q1 =“ security ”and the search result of the second search query Q2 =“ security & vulnerability ” The content set C2 is acquired as the search result of the third search query Q3 = “vulnerability & diagnosis”, respectively. The acquired content sets C1, C2, and C3 are sent to the content set combining means 5. Then, it progresses to step S4.

(Step S4): First Time In step S4, the content set is synthesized by the content set synthesis means 5 using the content set Ci (1 ≦ i ≦ 3). Various methods are conceivable, but as an example here, by calculating the union of Ci (1 ≦ i ≦ 3), the combined content set D [C1, C2, C3] (= C1∪C2 ∪ C3) is obtained.

  FIG. 4A shows a Venn diagram describing the relationship between the content sets C1, C2, and C3 and the combined content set D [C1, C2, C3]. In the Venn diagram shown in FIG. 4A, the content set 111 of the content set C1, the content set 112 of the content set C2, the content set 113 of the content set C3, and the content set synthesized as the union of the content sets C1, C2, and C3. A content set 114 that is D [C1, C2, C3] is illustrated. The content set D [C1, C2, C3] is stored in the second content set storage means 6, and the process proceeds to step S5.

(Step S5): First time In step S5, the content set C4 stored in the first content set storage unit 3 by the search session determination unit 7 and the content set D [C1, stored in the second content set storage unit 6 are stored. C2, C3].

  In the Venn diagram of FIG. 4A, the content set 114 of the content set C4 is also illustrated, and the relationship between the content set C4 and the content set D [C1, C2, C3] is described.

  Here, various methods can be considered as a comparison method. Here, as an example, the number of contents contained in the product set C4∩D [C1, C2, C3] of the first content set and the second content set | C4∩ The evaluation value E = | C4∩D [C1, C2, C3] | / | C4 | is calculated using D [C1, C2, C3] | and the number of contents | C4 | contained in the first content set C4. To do.

  FIG. 5A shows the number of contents included in the content sets C1, C2, C3, C4, D [C1, C2, C3] (= C1∪C2∪C3), C4∩D [C1, C2, C3]. Show. When the number of contents included in each content set is FIG. 5A, the evaluation value E is 504/900 = 0.56. Then, it progresses to step S6.

(Step S6): First time In Step S6, when the evaluation value E is large, it can be determined that the content of the search query Q4 is similar to the content of the search queries Q1, Q2, and Q3 so far. When the value E is small, it can be determined that the content of the search query Q4 is different from the content of the search queries Q1, Q2, and Q3 so far. Based on this, the search session determination means 7 determines whether or not the search sessions are the same using the evaluation value E = 0.56 calculated in step S5.

  Here, the threshold value α of the evaluation value E is set to 0.4, and when the evaluation value E is equal to or larger than the threshold value α, it is determined that the search session is the same. When the evaluation value E is less than the threshold value α, it is determined that the search session is different. In this case, the evaluation value E = 0.56> threshold α = 0.4, and it is determined that they are the same search session.

  Referring to the Venn diagram in FIG. 4 (a), the content set C4 and the content set D [C1, C2, C3] are largely overlapped, so the meaning of the search query Q4 and the search query so far It can be determined that the meanings of Q1, Q2, and Q3 are similar, and it is appropriate to determine that the search query Q4 = “security & distribution” is the same search session as when the search queries Q1, Q2, and Q3 are input. It is.

  Here, since it is determined that the search session is the same, the process proceeds to step S8.

(Step S8): First Time In step S8, the search query Q4 is stored in the search query history storage means 4. Thus, the search queries Q1, Q2, Q3 and Q4 are stored in the search query history storage means 4.

  Next, a case where the search query Q5 = “broadband” is input as the fifth search query will be described.

(Step S <b> 1): Second time In step S <b> 1, the search query storage unit 1 stores a search query Q <b> 5 = “broadband” for which a search session is to be determined. Then, it progresses to step S2.

(Step S2): Second time In step S2, the content search means 2 performs a search with the search query Q5 = “broadband”, and acquires a content set C5 as a search result. The content set C5 is stored in the first content set storage means 3. Then, it progresses to step S3.

(Step S3): Second time In Step S3, the search query Q1 = “security”, search query Q2 = “security & vulnerability”, search query Q3 = “vulnerability & diagnosis” stored in the search query history storage means 4 ”, The search query Q4 =“ security & distribution ”is searched in the content search means 2, and the content set C1 is set as the search result of the first search query Q1 =“ security ”, and the second search query Q2 = The content set C2 as the search result of “security & vulnerability”, the third search query Q3 = the content set C3 as the search result of “vulnerability & diagnosis”, and the fourth search query Q3 = “security & distribution” The content sets C4 are respectively acquired as search results. The acquired content sets C1, C2, C3, C4 are sent to the content set combining means 5. Then, it progresses to step S4.

(Step S4): Second time In step S4, the content set combining means 5 combines the content set using the content set Ci (1 ≦ i ≦ 4). Various methods are conceivable, but here, as an example, by calculating the union of Ci (1 ≦ i ≦ 4), the combined content set D [C1, C2, C3, C4] (= C1 ∪C2∪C3∪C4) is obtained.

  FIG. 4B shows a Venn diagram describing the relationship between the content sets C1, C2, C3, C4 and the combined content set D [C1, C2, C3, C4]. In the Venn diagram shown in FIG. 4B, the content set 111 of the content set C1, the content set 112 of the content set C2, the content set 113 of the content set C3, the content set 115 of the content set C4, and the content sets C1, C2, and C3. , A content set 116 which is a content set D [C1, C2, C3, C4] synthesized as a union of C4 is illustrated. The content set D [C1, C2, C3, C4] is stored in the second content set storage means 6, and the process proceeds to step S5.

(Step S5): Second time In step S5, the search session determination means 7 causes the content set C5 stored in the first content set storage means 3 and the content set D [C1 stored in the second content set storage means 6 to be stored. , C2, C3, C4].

  In the Venn diagram of FIG. 4B, the content set 117 of the content set C5 is also illustrated, and the relationship between the content set C5 and the content set D [C1, C2, C3, C4] is described.

  Here, various methods can be considered as the comparison method. Here, as an example, the number of contents included in the product set C5∩D [C1, C2, C3, C4] of the first content set and the second content set is | Using C5∩D [C1, C2, C3, C4] | and the number of contents | C5 | contained in the first content set C5, the evaluation value E = | C5∩D [C1, C2, C3, C4] | / | C5 | is calculated.

  FIG. 5B shows content sets C1, C2, C3, C4, C5, D [C1, C2, C3, C4] (= C1∪C2∪C3∪C4), C5∩D [C1, C2, C3, C4] indicates the number of contents included. When the number of contents included in each content set is shown in FIG. 5B, the evaluation value E is 114/950 = 0.12. Then, the process proceeds to step S6.

(Step S6): Second Time In step S6, the search session determination means 7 determines whether or not the search sessions are the same using the evaluation value E = 0.12 calculated in step S5. Since the threshold value α of the evaluation value E is 0.4, the evaluation value E = 0.12 <the threshold value α = 0.4, and it is determined that the search session is different.

  Referring to the Venn diagram in FIG. 4B, the content set C5 and the content set D [C1, C2, C3, C4] do not overlap so much. It can be determined that the meanings of the search queries Q1, Q2, Q3, and Q4 are different, and it is appropriate to determine that the search query Q5 = “broadband” is a different search session from when the search queries Q1, Q2, and Q3 are input. It is.

  Here, since it is determined that the search session is different, the process proceeds to step S7.

(Step S7): Second time In Step S7, since the search session is changed, the search query Qi (1 ≦ i ≦ 4) stored in the search query history storage unit 4 is deleted. Then, it progresses to step S8.

(Step S8): Second Time In step S8, the search query Q5 is stored in the search query history storage means 4. As a result, only the search query Q5 is stored in the search query history storage means 4.

  In the above description, the content search unit 2 performs content search using each search query stored in the search query history storage unit 4 in step S3, and the content searched by each search query in step S4. It describes that the content sets to be stored in the second content set storage means 6 are combined by combining the sets. For example, in order to search for a content set to be stored in the second content set storage unit 6 from the search queries Q1, Q2, and Q3 stored in the search query storage unit 4 by combining step S3 and step S4. The search queries Q1 ク エ リ Q2∪Q3 are combined, the content search means 2 searches the content set using the combined search query Q1∪Q2∪Q3, and the content set is stored in the second content set storage means 6. You can do this.

  Thus, in the search session determination in the information presentation apparatus according to the present embodiment, it is possible to comprehensively determine a plurality of search queries so far, and to more accurately determine the identity of the search purpose. .

[Application Example 1-2]
A second application example in the present embodiment will be described. In this application example, an upper limit of the number of search queries stored in the search query history storage unit 4 is set.

  By holding the upper limit of the number of search queries, the operation described in Application Example 1-1 is different from the processing in step S8. That is, when storing a new search query in step S8, the number of search queries stored in the search query history storage means 4 is confirmed. If the upper limit of the number of search queries is not exceeded, the new search query is stored as it is. To do. On the other hand, when the upper limit of the number of search queries is exceeded, a search query to be deleted without being stored is selected from the search queries already stored in the search query history storage unit 4 and the search queries to be newly added this time. Perform the decision process.

  As a method for determining a search query to be erased without storing, a method for erasing without referring to time information at the time of execution of the search query without storing it in the search query history storage unit 4, or content searched for by the search query A method of determining a search query to be deleted without referring to the number and storing it in the search query history storage means 4 can be considered.

  There are various variations of the method of referring to the time information when executing the search query, such as a method of deleting the oldest search query in order and a method of storing the search query in several skips. Further, in the method of determining a search query to be deleted from the search query history storage unit 4 with reference to the number of contents searched by the search query, a search query having a large number of contents, that is, a large content set is left, and the number of contents is A method of deleting a search query with a small number, that is, a small content set, can be considered.

  Thus, in the search session determination in the information presentation apparatus according to this application example, when the number of search queries increases, the time required for the search may increase and it may take time to obtain the content set. By determining the upper limit, it is possible to determine a search session while preventing an increase in time required for search, that is, time required for determination of a search session.

  In addition, when the search query is gradually changed, in principle, the content set stored in the second content set storage means becomes a content set including all the contents. Although there is a possibility that the search session is determined, it is possible to determine the search session more accurately by determining the upper limit of the search query.

[Application Example 1-3]
A third application example in the present embodiment will be described. In this application example, a variation of a method of synthesizing a content set by using the content set Ci (1 ≦ i ≦ n) in the content set combining unit 5 in step S4 will be described.

  The content set synthesizing unit 5 in this application example receives and receives the content set Ci (1 ≦ i ≦ n) searched by the search query Qi (1 ≦ i ≦ n) stored in the search query history storage unit 4. The union of the obtained content set Ci (1 ≦ i ≦ n) is calculated, and the calculated content set, ie, C1∪C2∪. . . Content set D [C1,. . . Cn].

[Application Example 1-4]
A fourth application example in this embodiment will be described. In this application example, a variation (modification or modification) of a method of synthesizing a content set by using the content set Ci (1 ≦ i ≦ n) in the content set combining unit 5 in step S4 will be described.

  The content set synthesizing unit 5 in this embodiment receives the content set Ci (1 ≦ i ≦ n) searched by the search query Qi (1 ≦ i ≦ n) stored in the search query history storage unit 4. Then, for each content included in each received content set Ci (1 ≦ i ≦ n), the number of content sets including each content is totaled.

  FIG. 6 shows an example of a tabulation table 120 in which the number of content sets belonging to each content is tabulated. The tabulation table 120 shown in FIG. 6 is an example of a tabulation method, and other tabulation methods may be used. The aggregation table 120 indicates that the content with content ID = A1 is included in the two content sets of the content sets C1 and C3.

  In the content set synthesizing means 5 in the present embodiment, a synthesis rule “Generate a content set after synthesis with content included in m or more content sets” is defined. The threshold value m may be any integer greater than or equal to 1, but here, the description will be given assuming that M = 2.

  Therefore, the content set synthesizing unit 5 collects contents included in two or more content sets. In the example of the summary table 120, the content IDs A1, A4, and A5 correspond. Therefore, the content set D [C1, C2, C3, C4, C5] is calculated by synthesizing the content set {A1, A4, A5} composed of the contents A1, A4, and A5.

[Application Example 1-5]
A fifth application example in this embodiment will be described. In this application example, a variation of a method of synthesizing a content set by using the content set Ci (1 ≦ i ≦ n) in the content set combining unit 5 in step S4 will be described.

  The content set synthesizing unit 5 in this application example receives the content set Ci (1 ≦ i ≦ n) searched by the search query Qi (1 ≦ i ≦ n) stored in the search query history storage unit 4. Then, for each content included in each received content set Ci (1 ≦ i ≦ n), the number of content sets including each content is totaled.

  FIG. 6 shows an example of a tabulation table 120 in which the number of content sets belonging to each content is tabulated. The tabulation table 120 shown in FIG. 6 is an example of a tabulation method, and other tabulation methods may be used. The aggregation table 120 indicates that the content with content ID = A1 is included in the two content sets of the content sets C1 and C3.

  In the content set synthesizing means 5 in this application example, a synthesis rule is defined that “a synthesized content set is generated with content included in a content set of p% or more of the content sets to be synthesized”. The threshold value p may be any real integer greater than 0 and less than or equal to 100, but here, p = 50, that is, a content set that is synthesized with content included in a content set of 50% or more As the explanation proceeds. For the sake of simplicity, the description will be made assuming that the number of content sets n = 5.

  The content set synthesizing unit 5 collects content included in 2.5 or more content sets at n = 5 × p = 50%. In the example of the summary table 120, the content IDs A4 and A5 correspond. Therefore, the content set D [C1, C2, C3, C4, C5] is calculated by combining the content set {A4, A5} composed of the contents A4 and A5.

[Application Example 1-6]
A sixth application example in this embodiment will be described. In this application example, a variation of a method of synthesizing a content set by using the content set Ci (1 ≦ i ≦ n) in the content set combining unit 5 in step S4 will be described.

  The content set synthesizing means 5 in this application example holds a content evaluation function based on how many times the search query has been searched in the past. Here, the content evaluation function for the search query m times before is represented by f (m).

  In the content set synthesizing means 5 in this application example, each content included in the content set Ci (1 ≦ i ≦ n) is searched using the content evaluation function f (m) with a search query m times before. For each content included in the content set, the value of f (m) is added to the content score. That is, when a certain content is included in the search query one time before, three times before, and seven times before, a score of f (1) + f (3) + f (7) is given to the content. In the present embodiment, the content score is calculated by adding the value of f (m). However, this is an example of score calculation, such as by multiplying the value of f (m). A method for calculating the score is also conceivable.

  Further, in the application example, the content set synthesizing unit 5 holds a threshold value x for the content score. Then, the content score and the threshold value x are compared, contents whose content score is equal to or greater than the threshold value are collected, and the content set is combined. . . Cn].

  Next, the content evaluation function f (m) will be mentioned. Various variations can be considered for the content evaluation function f (m). For example, f (m) = e ^ (m−1), f (m) = − log (m), or f (m) = 1 for the most recent b search queries, b For a search query more than once before, a setting method of setting f (m) = 0 is also effective.

[Application Example 1-7]
A seventh application example in this embodiment will be described. In this application example, a variation of a method of synthesizing a content set by using the content set Ci (1 ≦ i ≦ n) in the content set combining unit 5 in step S4 will be described.

  In the content search means 2 in this application example, when searching for content with a specified search query, a score based on the degree of match with the search query is assigned to each content included in the search result.

  In the content set synthesizing means 5 in this application example, the scores are totalized for each content included in the content set Ci (1 ≦ i ≦ n). For example, it is assumed that the score Score (Ci, Aj) is given to the content Aj included in the content set Ci. If the content Aj is included in the content sets C1, C3, and C5, Score (C1, Aj) + Score (C3, Aj) + Score (C5, Aj), which is the sum of the scores of the content Aj in each content set, is obtained. Calculated as content score.

  Further, in the application example, the content set synthesizing unit 5 holds a threshold value x for the content score. Then, the content score and the threshold value x are compared, contents whose content score is equal to or greater than the threshold value are collected to synthesize the content set, and the combined content set is set to the content set D [C1,. . . Cn].

  Hereinafter, this application example will be further described with a specific example. In this description, an example in which 15 contents are included in 5 content sets will be described.

  FIG. 7 shows a table 131 showing the scores of the contents Aj (1 ≦ j ≦ 15) in each content set Ci (1 ≦ i ≦ 5) searched by the search query Qi (1 ≦ i ≦ 5). In the “score in the content set” column of the table 131, in the content set C1 obtained by searching with the search query Q1, the score Score (C1, A1) in the search for the content A1 is 1, and the score Score of the content A2 It shows that (C1, A2) is 5. Similarly, in the “Score in content set” column of Table 131, it can also be seen that the content A1 belongs to the content sets C1, C3, and C5.

  FIG. 8A shows a Venn diagram representing the content set shown in the table 131 of FIG.

  The “summed score” column shown in the table 131 of FIG. 7 indicates ΣScore (Ci, Ai) obtained by summing up the content scores in each content set. The content A1 belongs to the content set C1, C3, C5, and the score in each content set is Score (C1, A1) = 1, Score (C3, A1) = 3, Score (C5, A1). ) = 2, and the combined score ΣScore (Ci, Ai) = 6.

  The “synthesized content set” column shown in the table 131 of FIG. 7 indicates whether each content belongs to the content set synthesized by the content set synthesizing unit 5 when the score threshold x = 10. The contents in which this column is ◯ belong to the synthesized content set.

  FIG. 8B shows the content set 130 synthesized by the content set synthesis means 5 using the threshold value x = 10 of the score. According to this synthesizing method, the content A1 included in three of the five content sets has a low score in each content set, that is, the degree of matching with each search query is low. It is not included in the content set 130 to be combined in the above. On the other hand, of the five content sets, the content A9 that belongs to only one content set has a high score in the content set C3, that is, has a high degree of matching with the search query Q3, and is therefore synthesized by the content set synthesizing unit 5. It will be included in the content set 130.

  In this application example, the content set can be combined by the content set combining unit 5 using the content score given by the content search unit 2. In other words, since a content set can be synthesized with highly reliable content as a search result, a search session with higher reliability can be determined.

[Application Example 1-8]
An eighth application example in this embodiment will be described. In this application example, a variation of a method of synthesizing a content set by using the content set Ci (1 ≦ i ≦ n) in the content set combining unit 5 in step S4 will be described.

  In the content set synthesizing means 5 in this application example, it is also effective to synthesize a content set by combining a plurality of content set synthesis variations described above.

  For example, the content set composition method is switched according to the number n of search queries stored in the search query history storage unit 4. For example, when the number n of search queries stored in the search query history storage unit 4 is less than 4, the content set D [C1,. . , Cn] and the number of search queries n is 4 or more, as described in Example 1-2, the content sets D [C1,. . , Cn].

  Note that the comparison reference number 4 for the number of search queries is merely an example, and the number of other search queries may be used. Also, the method of combining the content set synthesis method is effective other than the above-described combinations, and the method of combination is arbitrary.

[Application Example 1-9]
A ninth application example in this embodiment will be described. In this application example, a content search method from a search query in the content search unit 2 will be described.

  The search target content stored in the content search means 2 includes various methods such as a method of storing all the content actually presented to the user and a method of storing a reduced number of content in order to perform a search session determination process at high speed. A method can be considered.

  In addition, a method in which content obtained by removing content searched by an extremely large number of search queries is stored in the content search set 2 is also conceivable. This is because content searched with an extremely large number of search queries may be searched regardless of the search purpose of the user, and there is a possibility that the accuracy of search session determination based on the content set in this application example may be lowered. Because.

  As another application example, there may be a method of storing in the content search set 2 content obtained by removing content that can be searched with only one search query or an extremely small number of search queries. This is because content that can be searched only with an extremely small number of search queries is extremely unlikely to match search content sets, and is therefore considered not likely to contribute to search session determination. .

  You may use the content search method in the content search means 2 other than having demonstrated above.

[Application Example 1-10]
A tenth application example in this embodiment will be described. In this application example, a method of comparing the content set stored in the first content set storage unit 3 with the content set stored in the second content storage unit 6 in the search session determination unit 7 in step S5 will be described.

  In this application example, the comparison unit between the content set stored in the first content set storage unit 3 (referred to as a first content set) and the content set stored in the second content storage unit 6 (referred to as a second content set). Assume that the evaluation value E is calculated.

  The evaluation value E is a product set Cn + 1∩D [C1,. . . , Cn] includes the number of contents | Cn + 1∩D [C1,. . . , Cn] | and the number of contents | Cn + 1 | contained in the first content set Cn + 1, the evaluation value E = | Cn + 1∩D [C1,. . . , Cn] | / | Cn + 1 | This evaluation value E means the ratio of content included in common with the second content set in the content included in the first content set. Using the calculated evaluation value E, a search session is determined according to the calculated evaluation value E in step S6.

  The search session determination means according to this application example determines whether the search sessions are the same based on how many content sets searched for the search query to be determined are included in the content set to be determined. It becomes possible to do.

[Application Example 1-11]
An eleventh application example in this embodiment will be described. In this application example, a method of comparing the content set stored in the first content set storage unit 3 with the content set stored in the second content storage unit 6 in the search session determination unit 7 in step S5 will be described.

  In this application example, the comparison unit between the content set stored in the first content set storage unit 3 (referred to as a first content set) and the content set stored in the second content storage unit 6 (referred to as a second content set). Assume that the evaluation value E is calculated.

  The evaluation value E is a product set Cn + 1∩D [C1,. . . , Cn] includes the number of contents | Cn + 1∩D [C1,. . . , Cn] | and the number of contents | Cn + 1 | contained in the first content set Cn + 1, the evaluation value candidate E1 = | Cn + 1∩D [C1,. . . , Cn] | / | Cn + 1 |, and the product set Cn + 1∩D [C1,. . . , Cn] includes the number of contents | Cn + 1∩D [C1,. . . , Cn] | and the second content set D [C1,. . . , Cn] contains the number of contents | D [C1,. . . , Cn] | and the evaluation value candidate E2 = | Cn + 1∩D [C1,. . . , Cn] | / | D [C1,. . . , Cn] |. Then, the evaluation value candidate E1 and the evaluation value candidate E2 are compared, and the larger value max (E1, E2) is set as the evaluation value E.

  That is, the evaluation value E is the ratio of the content included in common with the second content set in the content included in the first content set, and the first content set occupied in the content included in the second content set. The value of the content ratio that is included in common is the larger value. Using the calculated evaluation value E, a search session is determined according to the calculated evaluation value E in step S6.

  In the search session determination means according to this application example, the content set searched by the search query to be determined is included in the content set to be determined and the content set searched by the search query to be determined. It is possible to determine whether or not the search sessions are the same from the two viewpoints of how much the content set to be determined is included.

  In this application example, even when the user performs a search in a small range and then performs a search in a large range by expanding the search range, the search session can be correctly determined.

[Application Example 1-12]
A twelfth application example in this embodiment will be described. In this application example, a method of comparing the content set stored in the first content set storage unit 3 with the content set stored in the second content storage unit 6 in the search session determination unit 7 in step S5 will be described.

  In this application example, the comparison unit between the content set stored in the first content set storage unit 3 (referred to as a first content set) and the content set stored in the second content storage unit 6 (referred to as a second content set). Assume that the evaluation value E is calculated.

  The evaluation value E is a product set Cn + 1∩D [C1,. . . , Cn] includes the number of contents | Cn + 1∩D [C1,. . . , Cn] | and the number of contents | Cn + 1 | contained in the first content set Cn + 1, the evaluation value candidate E1 = | Cn + 1∩D [C1,. . . , Cn] | / | Cn + 1 |, and the product set Cn + 1∩D [C1,. . . , Cn] includes the number of contents | Cn + 1∩D [C1,. . . , Cn] | and the second content set D [C1,. . . , Cn] contains the number of contents | D [C1,. . . , Cn] | and the evaluation value candidate E2 = | Cn + 1∩D [C1,. . . , Cn] | / | D [C1,. . . , Cn] |. A value E1 + E2 obtained by adding the evaluation value candidate E1 and the evaluation value candidate E2 is defined as an evaluation value E.

  That is, the evaluation value E is the ratio of the content included in common with the second content set in the content included in the first content set, and the first content set occupied in the content included in the second content set. This is the sum of the proportions of content that are included in common. Using the calculated evaluation value E, a search session is determined according to the calculated evaluation value E in step S6.

  In the search session determination means according to this application example, the content set searched by the search query to be determined is included in the content set to be determined and the content set searched by the search query to be determined. It is possible to determine whether or not the search sessions are the same from the two viewpoints of how much the content set to be determined is included.

  In this application example, even when the user performs a search in a small range and then performs a search in a large range by expanding the search range, the search session can be correctly determined.

[Application Example 1-13]
A thirteenth application example in this embodiment will be described. In this application example, a method of comparing the content set stored in the first content set storage unit 3 with the content set stored in the second content storage unit 6 in the search session determination unit 7 in step S5 will be described.

  In this application example, when the content search unit 2 searches for content using a search query, it is assumed that a search score indicating the degree of match with the search query is given to each content as a search result.

  When the content set is stored in the first content set storage unit 3 and the second content set storage unit 6, it is assumed that the search score for each content is also stored.

  When the search session determination means 7 compares the contents of the first content set storage means 3 and the second content set storage means 6, only the search score of the stored contents is greater than a certain threshold value. Perform a search session.

  This application example can eliminate the influence of content having a low search score and a low matching degree with the search query, and can determine a search session using content having a high search score and a high matching degree with the search query.

[Application Example 1-14]
A fourteenth application example in this embodiment will be described. In this application example, the machine configuration of the information providing apparatus according to the present embodiment, particularly the machine configuration when including a network such as the Internet or an intranet will be described.

  First, as a first configuration example, the search query storage unit 1 is configured by another machine, that is, a configuration example in which the search query storage unit 1 and the content search unit 2 are connected by a network such as the Internet or an intranet. Can be considered. In this case, a case in which a terminal that receives a search query input from a user and a server that includes an information providing device that performs search session determination are connected via a network may be considered.

  As a second configuration example, the search query storage unit 1 and the search query history storage unit 4 are configured by different machines, that is, the search query storage unit 1, the search query history storage unit 4, the content search unit 2, A configuration example in which the network is connected by a network such as the Internet or an intranet can be considered. In this configuration example, search query history information is stored in the user terminal, and search query history is sent to the server side when a search session determination is required, so that the search query history is permanently stored on the server side. There is no need to keep it. Therefore, it is effective in protecting the privacy information of the user.

  As a third configuration example, a configuration example in which the content search unit 2 is configured by another machine and communicates with other modules via a network can be considered. In this configuration example, there are effects such as sharing the content search unit 2 with other services, or using the content search unit 2 being provided by another service provider.

  As a fourth configuration example, the server-side configuration for determining a search session includes a first content set storage unit 3, a content set combining unit 5, a second content set storage unit 6, and a search session determination. A configuration example in which only the means 7 is configured and information is stored in the first content set storage means 3 and the content set combining means 5 via a network is conceivable. In this configuration example, it is possible to provide a function for determining a search session via a network, and when it is necessary to determine a search session in another apparatus, a search query is transmitted via the network according to this embodiment. By transmitting to the information providing apparatus, it is possible to receive the determination result of the search session.

  The machine configuration described above is merely an example, and other than the above, it is possible to implement in a form in which information is transmitted and received between each means via a network such as the Internet or an intranet.

  FIG. 9 is a block diagram showing the configuration of the second embodiment of the present invention. In the information presentation device according to the present embodiment, in addition to the configuration of the information providing device according to the first embodiment, the content set history combined by the content set combining means 5 is stored in association with the search query history. A history storage means 8 is further provided.

  FIG. 10 is a flowchart showing the operation of the information presentation apparatus according to the present embodiment.

  The operation of this embodiment will be described with reference to FIG. In this description, the user performs n searches in the same search session, and the search queries Q1, Q2,. . . , Qn (n> = 1) will be described from the state stored in the search query history storage unit 4. Further, the composition history storage means 8 stores each content set Ci (1 ≦ 1 ≦ 1) searched by the content search means 2 for each search query Qi (1 ≦ 1 ≦ n) stored in the search query history history storage means 4. n) as an input, the content set D [C1,. . . , Cn] are stored. Furthermore, for simplification of description, the search query history storage means 4 will be described assuming that only search queries belonging to the same search session are stored.

(Step S1)
In step S1, the search query storage means 1 stores a search query Qn + 1 for which a search session is to be determined. Then, it progresses to step S2.

(Step S2)
In step S2, the content search means 2 performs a search using the search query Qn + 1, and acquires a content set Cn + 1 as a search result. The content set Cn + 1 is stored in the first content set storage means 3. Further, the content set D [C1,. . . , Cn] is stored in the second content set storage means 6. Then, it progresses to step S5.

(Step S5)
In step S5, the search session determination means 7 causes the content set Cn + 1 stored in the first content set storage means 3 and the content set D [C1,. . . , Cn]. Various methods can be considered as the comparison method. Here, as an example, the product set Cn + 1∩D [C1,. . . , Cn] includes the number of contents | Cn + 1∩D [C1,. . . , Cn] | and the number of contents | Cn + 1 | contained in the first content set Cn + 1, the evaluation value E = | Cn + 1∩D [C1,. . . , Cn] | / | Cn + 1 | Then, it progresses to step S6.

(Step S6)
In step S6, if the evaluation value E is large, it can be determined that the content of the search query Q4 is similar to the content of the search queries Q1, Q2, and Q3 so far. If the evaluation value E is small, the search is performed. It can be determined that the content meaning the query Q4 is different from the content meaning the search queries Q1, Q2, Q3 so far. Based on this, when the evaluation value E is equal to or greater than the threshold value α (0 ≦ α ≦ 1), the search session determination unit 7 determines that the search session is the same. When the evaluation value E is less than the threshold value α, it is determined that the search session is different. If it is determined that the search session is the same, the process proceeds to step S8. If it is determined that the search session is different, the process proceeds to step S7.

(Step S7)
In step S7, since the search session is changed, the search query Qi (1 ≦ i ≦ n) stored in the search query history storage unit 4 is deleted. Then, it progresses to step S8.

(Step S8)
In step S8, the search query history storage means 4 stores the search query Qn + 1. Then, it progresses to step S9.

(Step S9)
In step S9, the search query Qn + 1 stored in the search query history storage unit 4 is searched in the content search unit 2, and as a result, a content set Cn + 1 is acquired. The acquired content set Ci (1 ≦ i ≦ n + 1) is sent to the content set combining means 5. Then, it progresses to step S10.

(Step S10)
In step S10, the content set D [C1,. . . , Cn] and the content set Cn + 1 generated in step S3 ′, the content set D [C1,. . . , Cn + 1]. Various methods are conceivable, but here, as an example, the content set D [C1,. . . , Cn] and content set Cn + 1 are calculated to calculate the combined content set D [C1,. . . , Cn + 1]. The synthesized content set D [C1,. . . , Cn + 1] is stored in the second content set storage means 6.

  In the information presentation apparatus according to the present embodiment, in addition to the information presentation function according to the first embodiment, the content set D [C1,. . . , Cn] in the combination history storage unit 8 has the effect of reducing the content combination processing in the content set combining unit 5 in addition to the effect of the first embodiment.

  FIG. 11 is a block diagram showing the configuration of the third embodiment of the present invention. In the information presentation device according to the present embodiment, in addition to the configuration of the information providing device according to the first embodiment, the search result history storage unit 9 stores the content set searched by the content search unit 2 in association with the search query. And further comprising.

  FIG. 12 is a flowchart showing the operation of the information presentation apparatus according to this embodiment. The operation of this embodiment will be described with reference to FIG. In this description, the user performs n searches in the same search session, and the search queries Q1, Q2,. . . , Qn (n> = 1) will be described as being stored in the search query history storage unit 4. The search result history storage unit 9 stores each content set Ci (1 ≦ 1 ≦ n) searched by the content search unit 2 for each search query Qi (1 ≦ 1 ≦ n) stored in the search query history storage unit 4. In the following description, n) is stored. Furthermore, for simplification of description, the search query history storage means 4 will be described assuming that only search queries belonging to the same search session are stored.

(Step S1)
In step S1, the search query storage means 1 stores a search query Qn + 1 for which a search session is to be determined. Then, it progresses to step S2.

(Step S2)
In step S2, the content search means 2 performs a search using the search query Qn + 1, and acquires a content set Cn + 1 as a search result. The content set Cn + 1 is stored in the first content set storage means 3. Further, the content set Cn + 1 is stored in the search result history storage unit 9. Then, it progresses to step S11.

(Step S11)
In step S 11, a content set Ci (1 ≦ i ≦ n) corresponding to each search query Qi (1 ≦ i ≦ n) stored in the search query history storage unit 4 is acquired from the search result history storage unit 9. . The acquired content set Ci (1 ≦ i ≦ n) is sent to the content set combining means 5. Then, it progresses to step S4.

(Step S4)
In step S4, the content set synthesizing unit 5 synthesizes the content set using the content set Ci (1 ≦ i ≦ n). Various methods are conceivable, but here, as an example, by calculating the union of Ci (1 ≦ i ≦ n), the combined content set D [C1,. . . , Cn]. The synthesized content set D [C1,. . . , Cn] is stored in the second content set storage means 6. Then, it progresses to step S5.

(Step S5)
In step S5, the search session determination means 7 causes the content set Cn + 1 stored in the first content set storage means 3 and the content set D [C1,. . . , Cn]. Various methods can be considered as the comparison method. Here, as an example, the product set Cn + 1∩D [C1,. . . , Cn] includes the number of contents | Cn + 1∩D [C1,. . . , Cn] | and the number of contents | Cn + 1 | contained in the first content set Cn + 1, the evaluation value E = | Cn + 1∩D [C1,. . . , Cn] | / | Cn + 1 | Then, it progresses to step S6.

(Step S6)
In step S6, if the evaluation value E is large, it can be determined that the content of the search query Q4 is similar to the content of the search queries Q1, Q2, and Q3 so far. If the evaluation value E is small, the search is performed. It can be determined that the content meaning the query Q4 is different from the content meaning the search queries Q1, Q2, Q3 so far. Based on this, when the evaluation value E is equal to or greater than the threshold value α (0 ≦ α ≦ 1), the search session determination unit 7 determines that the search session is the same. When the evaluation value E is less than the threshold value α, it is determined that the search session is different. If it is determined that the search session is the same, the process proceeds to step S8. If it is determined that the search session is different, the process proceeds to step S7.

(Step S7)
In step S7, since the search session is changed, the search query Qi (1 ≦ i ≦ n) stored in the search query history storage unit 4 is deleted. Then, it progresses to step S8.

(Step S8)
In step S8, the search query history storage means 4 stores the search query Qn + 1.

  In the information presentation apparatus according to the present embodiment, in addition to the information presentation function according to the first embodiment, the content set retrieved by the content retrieval unit 2 is stored in the search result history storage unit 9 once. In addition to the effect of the first embodiment, it is possible to obtain the effect of reducing the content search process in the content search unit 2.

  Further, the search result history storage means 9 described in the present embodiment also has the same effect that the content search processing in the content search means 2 can be reduced also in the embodiment provided in addition to the configuration of the second embodiment. be able to.

  FIG. 13 is a block diagram showing the configuration of the fourth embodiment of the present invention. In the information presentation apparatus according to the present embodiment, in addition to the configuration of the information presentation apparatus according to the first embodiment, when the search session determination unit 7 determines a plurality of search sessions, the plurality of determination results are comprehensively determined. And a comprehensive determination means 10 for determining the final search session and outputting the determination result of the final search session.

  FIG. 14 is a flowchart showing the operation of the information presentation apparatus according to the present embodiment. The operation of this embodiment will be described with reference to FIG. In the present description, it is assumed that the user has executed m types of search sessions so far, and search queries related to the m types of search sessions are stored in the search query history storage unit 4. Here, for description, the name of each search session stored in the search query history storage unit 4 is referred to as Rj (1 ≦ j ≦ m).

(Step S1)
In step S1, the search query storage unit 1 stores a search query Qx for which a search session is to be determined. Then, it progresses to step S2.

(Step S2)
In step S2, the content search unit 2 performs a search using the search query Qx, and acquires a content set Cx as a search result. The content set Cx is stored in the first content set storage means 3. Then, it progresses to step S12.

(Step S12)
In step S12, the processing from step S3 to step S13 is performed for each search session Rj (1 ≦ j ≦ m) stored in the search query history storage unit 4.

(Step S3)
In step S3, the content search unit 2 performs a search for each search query Qji (1 ≦ i ≦ n) belonging to the search session Rj stored in the search query history storage unit 4, and as a result, the content set Cji (1 ≦ i) i ≦ n) is acquired.

  In this description, the number of search queries included in each search session is assumed to be n, but in practice, the number of search queries included in each search session is different for each search session.

  The acquired content set Cji (1 ≦ i ≦ n) is sent to the content set combining means 5. Then, it progresses to step S4.

(Step S4)
In step S4, the content set synthesizing means 5 synthesizes the content set using the content set Cji (1 ≦ i ≦ n). Various methods are conceivable, but here, as an example, by calculating the union of Cji (1 ≦ i ≦ n), the combined content set D [Cj1,. . . , Cjn]. The synthesized content set D [Cj1,. . . , Cjn] is stored in the second content set storage means 6. Then, it progresses to step S5.

(Step S5)
In step S5, the search session determination unit 7 causes the content set Cx stored in the first content set storage unit 3 and the content set D [Cj1,. . . , Cjn]. Various methods can be considered as the comparison method. Here, as an example, the product set Cx∩D [Cj1,. . . , Cjn] includes the number of contents | CxＣD [Cj1,. . . , Cjn] | and the number of contents | Cx | included in the first content set Cx, the evaluation value Ej = | Cx∩D [Cj1,. . . , Cjn] | / | Cx |. Then, it progresses to step S13.

(Step S13)
In step S <b> 13, the evaluation value Ej calculated by the search session determination unit 7 is sent to the comprehensive determination unit 10. If there is a search session that still needs processing, the process proceeds to step S3 to process the next search session. When the processing for all the search sessions is completed, the process proceeds to step S14.

(Step S14)
In step S14, the overall determination means 10 performs a final search session determination according to the evaluation value E calculated in steps S12 to S13. Then, it progresses to step S15.

(Step S15)
In step S15, the search query Qx for which the search session stored in the search query storage unit 1 is determined is stored in the search query history storage unit 4. When storing, the search session determined by the comprehensive determination means 10 is stored in association with each other. If it is determined that there is no corresponding search session, it is stored as a new search session.

  Note that step S1, step S2, step S3 to step S5 in the flowchart of FIG. 14 are the same operations as those in the flowchart of the first embodiment shown in FIG.

  As described above, in the search session determination in the information presentation apparatus according to the present embodiment, information regarding a plurality of search sessions is held, and it is comprehensively determined which search session includes the search query to be determined. It becomes possible.

[Application Example 4-1]
A first embodiment in this embodiment will be described. In this application example, a variation of search session determination in the comprehensive determination unit 10 will be described.

  In the search session determination in the comprehensive determination unit 10 in this application example, the evaluation values for all the search sessions are calculated, and the search session with the maximum evaluation value is determined as the search session to which the search query Qx belongs.

  Further, as another variation of this application example, as in application example 1-1, a threshold value α of the evaluation value is provided, and when the maximum evaluation value is less than the threshold value α, there is no corresponding search session. The implementation method which determines that is mentioned.

  Further, as a further variation of this application example, there is an implementation method in which an evaluation value threshold value α is provided and all search sessions having an evaluation value of α or more are determined as corresponding search sessions.

  As described above, in the search session determination according to this application example, it is possible to select the most suitable search session after comparing a plurality of search sessions.

[Application Example 4-2]
A second application example in the present embodiment will be described. In this application example, a variation of search session determination in the comprehensive determination unit 10 will be described.

  FIG. 15 is a flowchart showing the operation of this application example. The operation of this application example will be described with reference to FIG. In the present description, it is assumed that the user has executed m types of search sessions so far, and search queries related to the m types of search sessions are stored in the search query history storage unit 4. Here, for the sake of explanation, the name of each search session stored in the search query history storage means 4 is referred to as Ri (1 ≦ i ≦ m).

(Step S1)
In step S1, the search query storage unit 1 stores a search query Qx for which a search session is to be determined. Then, it progresses to step S2.

(Step S2)
In step S2, the content search unit 2 performs a search using the search query Qx, and acquires a content set Cx as a search result. The content set Cx is stored in the first content set storage means 3. Then, it progresses to step S12.

(Step S12)
In step S12, the processing from step S3 to step S17 is performed on each search session Rj (1 ≦ j ≦ m) stored in the search query history storage unit 4. Note that the processing from step S3 is performed in order from the latest search session. Here, the newness of a search session is defined as a new search session to which a recent search query belongs.

(Step S3)
In step S3, the content search unit 2 performs a search for each search query Qji (1 ≦ i ≦ n) belonging to the search session Rj stored in the search query history storage unit 4, and as a result, the content set Cji (1 ≦ i) i ≦ n) is acquired. The acquired content set Cji (1 ≦ i ≦ n) is sent to the content set combining means 5. Then, it progresses to step S4.

(Step S4)
In step S4, the content set synthesizing means 5 synthesizes the content set using the content set Cji (1 ≦ i ≦ n). Various methods are conceivable, but here, as an example, by calculating the union of Cji (1 ≦ i ≦ n), the combined content set D [Cj1,. . . , Cjn]. The synthesized content set D [Cj1,. . . , Cjn] is stored in the second content set storage means 6. Then, it progresses to step S5.

(Step S5)
In step S5, the search session determination unit 7 causes the content set Cx stored in the first content set storage unit 3 and the content set D [Cj1,. . . , Cjn]. Various methods can be considered as the comparison method. Here, as an example, the product set Cx∩D [Cj1,. . . , Cjn] includes the number of contents | CxＣD [Cj1,. . . , Cjn] | and the number of contents | Cx | included in the first content set Cx, the evaluation value Ej = | Cx∩D [Cj1,. . . , Cjn] | / | Cx |. Then, it progresses to step S16.

(Step S16)
In step S <b> 16, the evaluation value Ej calculated by the search session determination unit 7 is sent to the comprehensive determination unit 10. Then, it progresses to step S17.

(Step S17)
In step S <b> 17, the overall determination means 10 compares the evaluation value Ej with the evaluation value threshold value α. If the evaluation value Ej is less than α, it is determined that it does not correspond to the search session Rj, and the process proceeds to step S12. If the evaluation value Ej is greater than or equal to α, it is determined that the evaluation session Ej corresponds to the search session Rj and the process proceeds to step S15.

(Step S15)
In step S15, the search query Qx for which the search session stored in the search query storage unit 1 is determined is stored in the search query history storage unit 4. At the time of storage, the search session Rj determined by the comprehensive determination unit 10 is stored in association with each other. If it is determined that there is no corresponding search session, it is stored as a new search session.

  As a variation of this application example, when the search session to be determined is selected in step S3, the determination is not performed for all the search sessions stored in the search query history storage unit 4, but the latest five, for example, It is also conceivable to perform a search session determination only for a search session. The number of search sessions for determining a search session is not limited to five, and is arbitrary and can be changed when the service is implemented.

  As described above, in the search session determination according to this application example, it is possible to determine the search session while determining the degree of matching with the search query to be compared in order from the latest search session.

[Application Example 4-3]
A third application example in the present embodiment will be described. In this application example, a variation of search session determination in the comprehensive determination unit 10 will be described.

  FIG. 16 is a flowchart showing the operation of this application example.

  The operation of this embodiment will be described with reference to FIG. In the present description, it is assumed that the user has executed m types of search sessions so far, and search queries related to the m types of search sessions are stored in the search query history storage unit 4. Here, for the sake of explanation, the name of each search session stored in the search query history storage means 4 is referred to as Ri (1 ≦ i ≦ m). Further, as the newness of the search session, it is defined that the search session to which the latest search query belongs is newer, and R1, R2,. . . , Rn.

(Step S1)
In step S1, the search query storage unit 1 stores a search query Qx for which a search session is to be determined. Then, it progresses to step S2.

(Step S2)
In step S2, the content search unit 2 performs a search using the search query Qx, and acquires a content set Cx as a search result. The content set Cx is stored in the first content set storage means 3. Then, it progresses to step S18.

(Step S18)
In step S18, a search session determination is performed for the latest search session R1 stored in the search query storage unit 4. Then, it progresses to step S19.

(Step S19)
In step S19, when it is determined that the search query Qx belongs to the search query R1 (YES), the process ends. On the other hand, if it is determined that the search query Qx does not belong to the search query R1 (NO), the process proceeds to step S20.

(Step S20)
In step S20, a search session determination is performed for the search session Rj (2 ≦ j ≦ m) stored in the search query storage unit 4.

  As described above, in the search session determination according to the present embodiment, first, a search session is performed for the latest search session, and if it is determined that the search session does not correspond to the latest search session, the remaining search sessions are determined. As a result, it is possible to determine a search session corresponding to a large number of search sessions while giving priority to the latest search session that the user will most memorize.

[Application Example 4-4]
A fourth application example in this embodiment will be described. In this application example, an application example in which the number of search sessions stored in the search query history storage unit 4 is limited will be described.

When the search query Qx stored in the search query storage unit 1 is a new search session in step S15 of the flowchart of the present embodiment shown in FIG. 14, when the search query Qx is stored in the search query history storage unit 4 When the upper limit of the number of search sessions that can be stored in the search query history storage unit 4 is exceeded, one or more search sessions are deleted, and a new search session to which the search query Qx belongs is stored in the search query history storage unit 4. To do.
As a method of determining the search session to be deleted when the upper limit of the number of search sessions is exceeded, the implementation method of deleting the oldest search session, the implementation method of deleting the least frequently used search session, all search sessions Variations such as a search method that deletes can be considered.

  As described above, in the search session determination by the information presentation device of this application example, the number of search sessions stored in the search query history storage unit 4 can be limited, and the search session determination in a finite time Can be performed.

[Application Example 4-5]
A fifth application example in this embodiment will be described. In this application example, a variation of the search session stored in the search query history storage unit 4 will be described.

  The search query history storage means 4 includes not only search session information generated from the user's search query, but also search session information defined in advance by a service builder or service operator, and search queries of other users. An implementation method for storing information on a search session generated from other than the search query of the user, such as information on a search session generated by referring to FIG.

  When the service builder or service operator predefines search session information, search queries to be determined as the same search session are listed and stored in the search query history storage unit 4. When defining search session information with reference to another user's search query, the method of using all search session information generated by the search query of another user, or the frequency of use of other user search queries Various methods such as a method for generating search session information using a high-priority query can be considered.

  Further, as another variation of implementation, it is conceivable to have a function of editing search session information stored in the search query history storage unit 4.

  As described above, in the search session determination by the information presentation device of this application example, since the search session is stored in the search query history storage unit 4 in advance, it is stored in advance for a user with a small search history. The search session can be appropriately determined using the information of the search session that has been set.

  FIG. 17 is a block diagram showing the configuration of the fifth embodiment of the present invention. In the information presentation device according to the present embodiment, the information presentation device presents to the user on the basis of the determination results of the search session calculation unit 11 by the information presentation device according to the first embodiment and the search session determination unit 7 of the search session calculation unit 11. Presentation information generating means 12 for generating information to be generated.

  FIG. 18 is a flowchart showing the operation of the information presentation apparatus according to the present embodiment. The operation of this embodiment will be described with reference to FIG.

(Step S21)
In step S21, the search session calculation unit 11 receives a search query input by the user. The search query can be input by the user using a keyboard or the like, or can be selected by the user from candidates presented on the screen, but is not particularly limited. Then, it progresses to step S22.

(Step S22)
In step S22, the search session calculation unit 11 determines a search session for the search query input by the user. As a search session determination method, various methods including the methods described in the first to fourth embodiments of the present invention can be considered. Then, it progresses to step S23.

(Step S23)
In step S23, the presentation information generation means 12 generates information to be presented to the user according to the search session obtained in step S22.

  In the information presentation apparatus shown in FIG. 17, the search session calculation means 11 describes the calculation of the search session according to the first embodiment, but the methods described in the first to fourth embodiments of the present invention are also included. The object of the present invention can also be realized by using various methods.

  With the operation described above, the information presentation apparatus according to the present embodiment can determine a search session using a search query input from the user, and generate information to be presented to the user according to the determined search session.

[Application Example 5-1]
A first application example in this embodiment will be described.

  FIG. 19 is a block diagram illustrating a configuration of an information presentation device according to this application example. In the information presentation apparatus shown in FIG. 19, the presentation information generation unit 12 includes a content list generation unit 1211 and a recommended content generation unit 1212.

  Next, the operation of this application example will be described using the flowchart of FIG.

(Step S21)
In step S21, the search session calculation unit 11 receives a search query input by the user. The search query can be input by the user using a keyboard or the like, or can be selected by the user from candidates presented on the screen, but is not particularly limited. In this embodiment, it is assumed that the user inputs a search query “security & distribution”. Then, it progresses to step S22.

(Step S22)
In step S22, the search session calculation unit 11 determines a search session for the search query input by the user. As a search session determination method, various methods including the methods described in the first to fourth embodiments of the present invention can be considered.

  In this embodiment, the search query “security & distribution” input by the user according to the procedure shown in the embodiment 1-1 is the search query “security” and “security & vulnerability” input by the user so far. , Suppose that it belongs to a search session consisting of “Vulnerability & Diagnosis”. Then, it progresses to step S23.

(Step S23)
In step S23, the presentation information generation means 12 generates information to be presented to the user according to the search session obtained in step S22.

  In this embodiment, first, the content list generation unit 1211 searches for content corresponding to the search query “security & distribution”, and generates a content list by arranging the content obtained as a search result.

  Next, the recommended content generation unit 1212 determines content to be recommended to the user with reference to the search session. Various methods are conceivable as a method for determining recommended content. For example, there is a method of manually defining the content to be recommended for each search session on the service operator side or the like. Here, the content to be recommended is determined by the operation of determining the recommended content shown in FIG. I will do it.

  FIG. 20 is a flowchart showing the operation of an example of a method for determining content recommended to the user in accordance with the search session in the recommended content generation means 1212 in this embodiment. The operation of the recommended content determination method will be described with reference to the flowchart of FIG.

  In the description of the operation, a search query included in the search session is assumed to be a search query Qi (1 ≦ i ≦ n). The current search query input by the user is set as a search query Qn + 1. A content set searched by the search query Qi is set as a content set Ci (1 ≦ i ≦ n + 1).

(Step S23-1)
In step S23-1, the search queries Q1,. . . , Qn, the content set C1∪C2∪. . . ∪Cn is calculated. Further, the content set Cn + 1 searched by the search query Qn + 1 is calculated. Using the calculated content sets, the content sets C1∪C2∪. . . A content set (C1∪C2∪ ... ∪Cn) ∩ (¬Cn + 1) that is included in ∪Cn but not in the content set Cn + 1 is calculated. Then, it progresses to step S23-2.

(Step S23-2)
In step S23-2, content to be presented to the user is determined from the contents included in the content set (C1∪C2∪ ... ∪Cn) ∩ (¬Cn + 1) calculated in step S23-1. Various determination methods can be considered. For example, a method of determining popular content, a content registered recently, a content not yet viewed by the user, and the like can be cited as an example. Then, it progresses to step S23-3.

(Step S23-3)
In step S23-3, the content determined in step S23-2 is presented to the user.

  As described above, in the information presentation device according to this application example, the content list searched by the search query input by the user is presented to the user, and the content determined according to the search session determined by the search session calculation unit 11 is displayed to the user. Can be recommended. Since the content recommended to the user is related to the search query input by the user, it is possible to recommend the content that matches the search purpose of the user.

[Application Example 5-2]
A second application example in the present embodiment will be described. In this application example, another variation of the method of determining the content recommended to the user in the recommended content generation unit 1212 of the presentation information generation unit 12 will be described.

  FIG. 21 is a flowchart showing an exemplary operation of a method for determining content recommended to a user in accordance with a search session in the recommended content generation unit 1212 in this application example. The operation of the recommended content determination method will be described with reference to the flowchart of FIG.

  In the description of the operation, a search query included in the search session is assumed to be a search query Qi (1 ≦ i ≦ n). The current search query input by the user is set as a search query Qn + 1. A content set searched by the search query Qi is set as a content set Ci (1 ≦ i ≦ n + 1).

  FIG. 22 shows an advertisement registration table 121 that defines the relationship between advertisement content recommended to the user and a search query (target search query) for which each advertisement content is desired to be presented.

(Step S23-4)
In step S23-4, the search queries Q1,. . . , Qn, Qn + 1, the content set C1∪C2∪. . . ∪Cn∪Cn + 1 is calculated. Then, it progresses to step S23-5.

(Step S23-5)
In step S23-5, for each advertisement content Aj (1 ≦ j ≦ m) registered in the advertisement registration table 121, a content set TCj searched by the target search query Tj defined in each advertisement content Aj is calculated. To do. Then, it progresses to step S23-6.

(Step S23-6)
In step S23-6, for the content set TCj calculated in step S23-5, the content set C1∪C2∪. . . The product set TCj∩ (C1∪C2∪ ... ∪Cn∪Cn + 1) with ∪Cn∪Cn + 1 is calculated, and the number of contents belonging to the calculated product set Vj = | TCj∩ (C1∪C2∪ ... ∪Cn ∪Cn + 1) | is calculated. Then, it progresses to step S23-7.

(Step S23-7)
In step S23-7, the number of contents Vj (1 ≦ j ≦ m) is compared, and the advertisement content with the maximum Vj is determined as the advertisement content to be presented to the user. Then, it progresses to step S23-8.

(Step S23-8)
In step S23-8, the advertising content determined in step S23-7 is presented to the user.

  As described above, in the information presentation device according to the application example, the content list searched by the search query input by the user is presented to the user, and the advertisement content determined according to the search session determined by the search session calculation unit 11 is displayed. Can recommend to users. Since the advertising content to be presented to the user is related to the search query input by the user, it is possible to present the advertising content that matches the user's search purpose.

  FIG. 23 is a block diagram showing the configuration of the sixth embodiment of the present invention.

  The information presentation apparatus according to this embodiment includes a content storage means 1201, a condition instruction means 1202, and a content list generation means 1203 in the presentation information generation means 12 in the configuration of the information presentation apparatus according to the fifth embodiment shown in FIG. , Content list presentation means 1204, presentation history management means 1205, and correction sorting control means 1206.

  The content storage unit 1201 stores content information to be presented. The condition instruction unit 1202 determines a search query, that is, a condition for selecting content. The content list generation unit 1203 generates a content list from the contents stored in the content storage unit 1201 in accordance with the search query determined by the condition instruction unit 1202. The content list presentation unit 1204 presents the content list generated by the content list generation unit 1203 to the user, and the user browses the content. The presentation history management unit 1205 manages the content list generated by the content list generation unit 1203 and the information presentation history of the content list presentation unit 1204. The correction sorting control unit 1206 controls whether the correction sorting unit 34 can operate.

  The content list generation unit 1203 includes a condition matching unit 12031, a content selection unit 12032, a base sorting unit 12033, and a correction sorting unit 12034.

  The condition collation unit 12031 collates whether the search query determined by the condition instruction unit 1202 is a search query presented in the past. The content selection unit 12032 selects content that matches the search query from the content stored in the content storage unit 1201 when the corresponding query is not found by the condition matching unit 12031. The base sorting unit 12033 sorts the content selected by the content selection unit 12032 in order of popularity, price, dictionary, or the like. The correction sorting unit 12034 corrects the sort order determined by the base sorting unit 12033 with reference to the presentation status of each content stored in the presentation history management unit 1205, and preferentially presents the content not presented to the user. Correct sorting.

  The presentation history management unit 1205 includes a presentation history storage unit 12051, a condition designation presentation history extraction unit 12052, and a content designation presentation history extraction unit 12053.

  The presentation history storage unit 12051 stores the content list generated by the content list generation unit 1203 together with the search query, and further stores the content list presentation state in the content list presentation unit 1204 and the user browsing history as presentation history information. . The condition designation presentation history extraction unit 12052 extracts a content set with a presentation order from the presentation history information stored in the presentation history storage unit 12051 using a search query as a key. The content designation presentation history extraction means 12053 includes the condition designation presentation history extraction means 12052 and the information stored in the presentation history storage means 12051, using the content as a key, what search query the content is sorted in, what content Information such as whether the list has been presented to the user by the list presenting means 1204 and whether the user has selected is extracted.

  FIG. 24 is a flowchart showing the operation of the information presentation apparatus according to this embodiment. The operation of this embodiment will be described with reference to FIG.

(Step S101)
In step S101, a search query is set in the condition instruction unit 1202. The search query set in the condition instruction unit 1202 has the same content as the search query stored in the search query storage unit 1 of the search session calculation unit 11. The search query can be set by a method in which a user sets a search keyword or a method in which the system automatically sets using various recommendation engines such as collaborative filtering. However, any setting method may be used. After setting, the process proceeds to step S102.

(Step S102)
In step S102, the condition collating unit 12031 collates whether or not the search query set by the condition instructing unit 1202 has been used in the past. The collation is performed via the condition designation presentation history extraction unit 12052. The condition designation presentation history extraction unit 12052 extracts a history related to the search query from the presentation history stored in the presentation history storage unit 12051. The range of the history to be verified is assumed to be within the same session, within a fixed time such as within 24 hours, unlimited, etc., and the user is basically assumed to target the history of the same user. However, the range of the history to be verified is set by the service.

  When the history is found by the condition designation presentation history extraction means 12052, the search query and the content with the presentation order are sent from the condition designation presentation history extraction means 12052 to the condition matching means 12031. Then, the process proceeds to step S108.

  If no history is found by the condition designation presentation history extraction means 12052, for example, information indicating that no empty set is found is sent from the condition designation presentation history extraction means 12052 to the condition matching means 12031. Then, the process proceeds to step S103.

(Step S103)
In step S103, when the condition matching unit 12031 determines that the search query set by the condition instruction unit 1202 has not been used in the past, the content selection unit 12032 selects the condition from the contents stored in the content storage unit 1201. The content that matches the search query set by the instruction unit 1202 is selected. The selected content is sent to the base sorting means 12033. Then, it progresses to step S104.

(Step S104)
In step S104, the base sorting unit 12033 sorts the content selected by the content selection unit 12032 using a sorting method (called a base sorting method) such as a predetermined popularity order, price order, dictionary order, or the like. In this embodiment, the sorting method used in the base sorting means 12033 does not affect the effect of the present invention, and any of them may be used. Then, it progresses to step S112.

(Step S112)
In step S112, it is determined whether or not correction sorting is performed according to the correction sorting control unit 1206.

  When it is determined that the search session determined by the search session calculation unit 11 is the same as the search session based on the immediately preceding search query, the correction sorting control unit 1206 determines to perform correction sorting. On the other hand, when it is determined that the search session determined by the search session calculation unit 11 is different from the search session based on the previous search query, the correction sorting control unit 1206 determines not to perform correction sorting.

  As a result, if it is determined by the correction sorting control means 1206 that correction sorting is to be performed, the process proceeds to step S105. On the other hand, if the correction sorting control unit 1206 determines not to perform correction sorting, the process proceeds to step S108.

(Step S105)
In step S <b> 105, the presentation history is acquired for each content sent from the base sorting unit 12033 by the correction sorting unit 12034. Content presentation history acquisition is performed via the content designation presentation history extraction means 12053. The content designation presentation history extraction unit 12053 extracts the presentation history of each content from the presentation history stored in the presentation history storage unit 12051. The range of the target history is assumed to be unlimited within a certain time, such as within the same session, within 24 hours, etc., and the user is basically assumed to target the history of the same user. However, the target history range is set by the service. Then, it progresses to step S106.

(Step S106)
In step S106, the correction sorting unit 12034 refers to the presentation history of each content acquired via the content designation presentation history extraction unit 12053, and corrects the sorting order using the presentation history for the contents sorted by the base sorting unit 12033. Do. In the correction, priority is given to content that has not been presented to the user, and correction is made so that the content is placed at the top of the content list.

  Here, an example of a correction method in the correction sorting means 12034 is shown. The correction sorting unit 12034 acquires the presentation history for each content to be presented via the content designation presentation history extraction unit 12053 of the presentation history management unit 1205. The obtained presentation history is sorted in ascending order (in ascending order). The correction sorting unit 12033 sorts the content with the same number of presentation histories, for example, by sorting with a stable algorithm such as distribution counting sort, so that the order sorted by the base sorting unit 12033 is maintained.

  Note that this correction sorting method is an example, and another method may be used to prioritize content that has not been presented to the user and place it at the top of the content list. Then, it progresses to step S107.

(Step S107)
In step S107, the search query and the content set with a presentation order are stored in the presentation history storage unit 12051. Then, it progresses to step S108.

(Step S108)
In step S108, the content list presentation unit 1204 presents a content list. The list is generated according to the presentation order sorted by the correction sorting means 12034 or the presentation order sent from the condition collating means 12031 and presented to the user. The content presented to the user on the content list page is recorded in the presentation history of the presentation history storage unit 12051. If the content does not fit on one page due to the content presentation area, the content list page is divided into a plurality of pages. When the content list page is divided into a plurality of pages, it is desirable to record that only the content included in the actually presented content list page has been presented. Then, it progresses to step S109.

(Step S109)
In step S109, when a page turning operation such as next page presentation / all page presentation occurs when content selection or content list pages are divided into a plurality of pages, the content list presentation unit 1204 performs processing according to the user's interaction. I do. Then, it progresses to step S110.

(Step S110)
In step S110, the content of the interaction performed in step S109 is recorded in the presentation history storage unit 12051. For example, when a page turning operation occurs as an interaction from the user, the content list presentation unit 1204 updates the content list page and records information indicating that the content is presented in the presentation history storage unit 12051. As another example, when the user selects the detailed information presentation of content from the content included in the content list page, information indicating that the content has been presented in detail is recorded in the presentation history storage unit 12051. To do. Then, it progresses to step S111.

(Step S111)
If it is determined in step S111 that the user has changed the search query in order to change the content search method, the process proceeds to step S101. Otherwise, the process proceeds to S108.

  In the information presentation device according to the present embodiment, when it is determined that the search query input by the user is the same search session as the search query input by the user immediately before, the content not yet presented to the user is displayed with priority. In other words, when it is determined that the search session is the same, the content is sorted based on the content presentation history, and the content presented in another search query is lowered in priority at the time of sorting. Priority can be raised.

  By adopting this mechanism, the user has the advantage of being able to know different types of content if the selection method is changed. Furthermore, from the perspective of the service provider, it is possible to present more content to the user, so it is possible to introduce not only popular content but also so-called long tail products to the user. Has the effect of increasing.

[Application Example 6-1]
A first application example in this embodiment will be described.

  FIG. 25 is a diagram showing an example of content information stored in the content storage unit 1201. FIG. 25 shows an example in which content data 120100 with video content as the content to be presented is stored in the content storage unit 1201. The content data 120100 stores attribute values such as content ID, title, genre, recommendation level, and popularity level for each content. The content data 120100 is an example of content information stored in the content storage unit 1201, and the storage form may be any form as long as the content search is possible.

  FIG. 26 is a diagram showing an example of the presentation history information stored in the presentation history storage unit 12051. FIG. 26 shows an example in which the presentation history information 120101 when the user browses the content data 120100 shown in FIG. 25 is stored.

  In the example of FIG. 26, the presentation history information 120101 includes a search query-content list correspondence table 1201011 and a content-history correspondence table 1201012. The search query-content list correspondence table 1201011 stores a search query for each user and a content list in the search query. The content-history correspondence table 1201012 stores the number of times of presentation of each content and the number of times of browsing detailed information for each user. The browsing history information 120101 shown in FIG. 26 is an example in this application example, the content list generated by the content list generating unit 1203 is stored together with the search query, and the content list presenting state in the content list presenting unit 1204 is stored. Any form can be used as long as it can store the user's browsing history.

  Next, the operation of this application example will be described with reference to the flowchart of FIG.

  Here, content data 120100 is stored in the content storage unit 1201, presentation history information 120101 is stored in the presentation history storage unit 12051, and the user USER-01 uses the information presentation device. Will be described.

(Step S101)
In step S101, it is assumed that the user sets “genre = animation” as a search query in the condition instruction unit 1202. Then, it progresses to step S102.

(Step S102)
In step S102, the condition collating unit 12031 collates whether the search query “genre = animation” set by the condition instructing unit 1202 has been used in the past. The collation is performed via the condition designation presentation history extraction means 12052. The condition designation presentation history extraction means 12052 refers to the search query-content list correspondence table 1201011 in the presentation history information 120101 stored in the presentation history storage means 12051, and uses the search query “genre = The contents list (in order of presentation) “CONT-01, CONT-02, CONT-03” is sent to the condition matching means 12031. Then, it progresses to step S108.

(Step S108)
In step S108, the content list presentation unit 1204 presents a content list. The list is generated according to the order sent from the condition matching unit 12031.

  FIG. 27 shows a screen 120110 of a content list presentation example. In addition, the fact that the contents CONT-01, CONT-02, and CONT-03 are presented is recorded in the presentation history of the presentation history storage means 12051.

  FIG. 28 shows the presentation history information 120102 stored in the presentation history storage means 12051 after recording.

(Step S101) Second Time In step S101, it is assumed that the user changes the search query to “popularity = A” in the condition instruction unit 1202. Then, it progresses to step S102 (2nd time).

(Step S102) Second Time In step S102, the condition collating unit 12031 collates whether the search query “popularity = A” set by the condition indicating unit 1202 has been used in the past. The collation is performed via the condition designation presentation history extraction means 12052. The condition-designated presentation history extraction unit 12052 refers to the search query-content list correspondence table 1201011 in the presentation history information 120101 stored in the presentation history storage unit 12051, and the search query “popularity = Confirm that there is no “A”. Then, the condition designation presentation history extraction means 12052 sends an empty set indicating that no search query was found to the condition matching means 12031. Then, it progresses to step S103.

(Step S103)
In step S103, the content selection unit 12032 selects the content of “popularity = A” from the content information 120100 stored in the content storage unit 1201. In this case, contents CONT-01, CONT-04, CONT-07, CONT-10, and CONT-11 are selected, and the selected contents are sent to the base sorting means 12033. Then, it progresses to step S104.

(Step S104)
In step S104, it is assumed that the base sorting unit 12033 defines that the sorting is performed in the dictionary order of titles as the base sorting method. In this case, the contents are CONT-01 (title = Alps girl), CONT-07 (title = Gots), CONT-10 (title = football legs vs Sakura), CONT-11 (professional baseball digest), CONT-04 ( Sorted in the order of crossing demons. Information on the sorted contents is sent to the correction sorting means 12034. Then, it progresses to step S112.

(Step S112)
In step S112, it is determined whether or not correction sorting is performed according to the correction sorting control unit 1206.

  In this embodiment, it is assumed that the search session calculation unit 11 determines that the immediately preceding search query “genre = animation” and the current search query “popularity = A” are the same search session. In this case, the correction sorting control unit 1206 determines to perform correction sorting. If the correction sorting control unit 1206 determines to perform correction sorting, the process proceeds to step S105.

(Step S105)
In step S105, the content designation presentation history extraction means for each content (CONT-01, CONT-07, CONT-10, CONT-11, CONT-04) sent from the base sorting means 12033 by the correction sorting means 12034. A presentation history is acquired via 12053. In this case, the content designation presentation history extraction means 12053 refers to the content-history correspondence table 1201022 of the presentation history information 120102, CONT-01 is presented twice, browsing history is once, and CONT-04 is presented zero times. Presentation history such as browsing history 0 times is acquired. Then, it progresses to step S106.

(Step S106)
In step S106, the correction sorting means 12034 corrects the sorting order using the presentation history for the contents sorted by the base sorting means 12033 with reference to the presentation history of each content acquired via the content designation presentation history extraction means 12053. Do. In the correction, priority is given to content that has not been presented to the user, and correction is made so that the content is placed at the top of the content list.

  Here, an example of a correction method in the correction sorting means 12034 is shown. The correction sorting unit 12034 acquires the presentation history of USER-01 for each content to be presented via the content designation presentation history extraction unit 12053 of the presentation history management unit 1205. Here, CONT-01 = 2 times of presentation, CONT-07 = 0 times of presentation, CONT-10 = 0 times of presentation, CONT-11 = 0 times of presentation, and CONT-04 = 0 times of presentation are acquired. . Then, the acquired presentation history is sorted in ascending order (in ascending order), the presentation order of CONT-01 with two presentations is lowered, and CONT-07, CONT-10, CONT-11, CONT-04, CONT- Rearrange 01.

  Note that this correction sorting method is an example, and another method may be used to prioritize content that has not been presented to the user and place it at the top of the content list. Then, it progresses to step S107.

(Step S107)
In step S107, the result of sorting by the correction sorting unit 12034 is stored in the presentation history storage unit 12051.

  FIG. 29 shows browsing history information 120103 stored in the presentation history storage means 12051. Then, it progresses to step S108 (2nd time).

(Step S108) Second Time In step S108, the content list presenting means 1204 presents a content list. The list is generated according to the order sent from the correction sorting unit 12034. In addition, the fact that the contents CONT-07, CONT-10, CONT-11, CONT-04, and CONT-01 are presented is recorded in the presentation history of the presentation history storage unit 12051.

  FIG. 30 shows a screen 120111 of a content list presentation example. On screen 120111, a content list is generated so that CONT-01 that has been presented to user USER-01 in the past is presented at the bottom of the list.

[Application Example 6-2]
A second application example in the present embodiment will be described. In this application example, a variation of the information presentation apparatus according to this embodiment will be described.

  In the present embodiment, there are several possible variations of the content sorting method in the content list generation unit 1203. For example, as a variation of one configuration, a configuration in which the content list generation unit 1203 does not include the base sorting unit 12033 is also conceivable.

  FIG. 31 is a block diagram that does not include the base sorting means 12033. In the case of the configuration of FIG. 31, the content selected by the content selection unit 12032 is directly sent to the correction sorting unit 12034, and the content is sorted by the correction sorting unit 12034. Since the function for sorting contents based on the presentation history of contents, which is a feature of the present invention, is provided in the correction sorting means 12034, the same function can be provided in the configuration of FIG.

  In another variation, in this embodiment, as the presentation history managed by the presentation history management unit 1205, the last presentation date and time of each content is recorded together, and when the content is sorted by the correction sorting unit 12034, It is also conceivable that a content with a new final presentation date / time is more difficult to be presented, and a content with an old final presentation date / time is easier to be presented than a content with a new final presentation date / time. By taking such a form, it becomes possible to present in order from the content with the lightest memory of the user in the next priority of the content that the user does not know at all.

  In another variation, in this embodiment, as the presentation history managed by the presentation history management unit 1205, the importance level of each content for each user is recorded together, and the content is sorted by the correction sorting unit 12034. At this time, a form in which content with high importance of content is more easily presented than content with low importance of content may be considered. For example, when corrective sorting is performed, each content is scored in proportion to the presentation history, and sorting is performed in the ascending order of the content score. This variation is realized by providing a mechanism that reduces the content according to the importance of the content. However, the method of reflecting the importance of content described here is an example, and other reflection methods may be used. The content importance input methods include a method in which the user explicitly inputs, a method in which the user's access history and preference information are automatically generated, and a method in which the importance of purchased content is extremely reduced. However, it may be any including other methods.

  In another variation, in the present embodiment, the correction sorting unit 12034 prioritizes the content that has not been presented to the user and presents it to the user when correcting the content so that it is placed at the top of the content list. A configuration is possible in which a certain content is deleted from the content list. In general, since the amount of calculation is smaller when the content is deleted than when the content is sorted, this variation is particularly effective when the information presentation apparatus of the present invention is applied to a large-scale service.

[Application Example 6-3]
A third application example in the present embodiment will be described. In this application example, a variation of the information presentation apparatus according to this embodiment will be described.

  In this application example, in a portal site on the Internet, the content provider that provides the content while retaining the content right and the content provided by the plurality of content providers are integrated to unify the attribute information of each content Explains the case where the service is provided by a primary service provider that enables content search, etc., and a secondary service provider that provides value-added services to the content service provided by the primary service provider. To do.

  The primary service provider is assumed to be a service provider that performs a content providing service using the conventional technology, and manages the content storage unit 1201, the condition instruction unit 1202, the content selection unit 12032, and the base sorting unit 12033, and is arbitrarily selected. The search keyword is selected and contents are sorted in order of popularity, price, dictionary, etc., and a list of contents is provided to the user.

  The secondary service provider provides an additional function to the content providing service provided by the primary service provider, and manages the condition matching unit 12031, the correction sorting unit 12034, the content list presentation unit 1204, and the presentation history management unit 1205. In consideration of situations such as content presentation history and user content selection history, for example, content that is not presented to the user is preferentially presented and content is presented in the same order in the same search query. As such, it provides the added value of properly presenting content.

  From the viewpoint of the user, it is possible to obtain higher convenience than the service provided by the primary service provider by using the content providing service provided by the secondary service provider. For this reason, it is considered that the secondary service provider can obtain compensation from the user. From the perspective of the primary service provider, the content service channel will be expanded by allowing secondary service providers for users who want to use highly convenient services even if they pay for the service to use the content. There is an advantage that you can.

  In addition, a form in which the primary service provider does not provide the content providing service directly to the user but only provides the content to the secondary service provider is also conceivable. From the perspective of secondary service providers, the primary service provider can take care of collecting the content from multiple content providers and unifying the attribute information of various types of content. Specialized function maintenance can be performed.

[Application Example 6-4]
A fourth application example in this embodiment will be described. In this application example, a variation of the information presentation apparatus according to this embodiment will be described.

  FIG. 32 is a block diagram illustrating a configuration of this application example. The information presentation apparatus according to this application example is characterized in that the presentation history management unit 1205 includes a presentation history correction unit 12054 that corrects the presentation history information stored in the presentation history storage unit 12051.

  Next, features of the information presentation apparatus according to the configuration of this application example including the presentation history correction unit 12054 will be described.

[Application Example 6-4-1]
In this application example, the presentation history correction unit 12054 deletes the presentation history stored in the presentation history storage unit 12051 after a certain period of time since the last use. By providing the presentation history correction unit 12054, even once presented content can be presented to the user again after a certain period (history deletion period). That is, even if the content has been presented once but not selected by the user, the opportunity to be presented again to the user is given after the history deletion period.

  Although various patterns can be considered for the history deletion period depending on the application service, the length of the period is not particularly limited in the implementation of the present invention. For example, it is usually possible to set a day (24 hours) to about one week, but if the service provider wants to introduce as much content as possible to the user, the history deletion period is set to one week or one. It is also possible to set a long period such as a month.

  Note that this history deletion period can be set according to the method of setting all the history deletion periods, the method of setting the history deletion period for each content, the method of changing the history deletion period according to the season, or the number of days from the content registration date. Various methods such as a method of changing the deletion period are conceivable.

[Application Example 6-4-2]
In this application example, the presentation history correction unit 12054 deletes the presentation history related to specific content from the presentation history stored in the presentation history storage unit 12051 with a special intention of the service provider.

  By providing the presentation history correction means 12054, it is possible to make it easier for the service provider to be presented to the user again even after the content that the service provider has strongly recommended to the user is presented once.

  Since the presentation history related to the specific content can be deleted by the presentation history correction means 12054, the following service operation can be performed.

  For example, in the portal site on the Internet, the service provider that operates the portal is different from the content provider that manages the content provided on the portal, and the service provider takes the content from multiple content providers to the portal. There is a case of building a site. From the viewpoint of content providers, in order to improve the sales of the content to be provided, it would be a good idea to present it to the user as much as possible. Based on this, by receiving additional costs from the content provider who wants to expose as much content as possible, the service provider explicitly deletes the presentation history for the content owned by the content provider, and the content provider Make it easier for the provider's content to be presented. Various methods such as a method of deleting all of the presentation histories at once and a method of deleting the presentation history in a period shorter than a normal deletion period can be considered.

  The service operation described above is one application example of this embodiment, and this embodiment is not limited to this application example.

[Application Example 6-4-3]
In this application example, the presentation history correcting unit 12054 deletes the presentation history related to a specific search query from the presentation histories stored in the presentation history storage unit 12051 with a special intention of the service provider. Alternatively, the presentation history correcting unit 12054 explicitly describes the presentation history intended by the service provider for a specific search query. For example, in a portal site on the Internet, there are cases where a service provider operating a portal arbitrarily controls the order of content presentation when searching for content using a certain search keyword. In such a case, by providing the presentation history correcting means 12054, the search history of the search query is unconditionally deleted so as to be in the content presentation order intended by the service provider. It becomes possible to present a content list.

[Application Example 6-4-4]
In this application example, the presentation history correction unit 12054 deletes the presentation history stored in the presentation history storage unit 12051 according to the size of the presentation history stored in the presentation history storage unit 12051.

  For example, when the size of the presentation history stored in the presentation history storage unit 12051 exceeds a set threshold, the information presentation device makes a determination and deletes the presentation history. As for the threshold value, a case where the threshold value is set for the entire information presentation apparatus, a case where the threshold value is set for each user, and the like can be considered, but any other method may be used. In addition, as a method for determining the presentation history to be deleted, a method of deleting from an old presentation history, a method of deleting from a presentation history with low importance, and the like can be considered, but any method including other methods may be used.

[Application Example 6-4-5]
In this application example, in a portal site on the Internet, the service provider that manages the portal is different from the content provider that manages the content provided on the portal. A case where a portal site is constructed by depositing content will be described.

  The service provider usually collects the value from the content provider by presenting the content owned by the content provider to the user, or collects the value from the content provider or the user when the user uses the content. And gain profits.

  In the information presentation apparatus according to this application example, when the service provider collects additional consideration from the content provider who desires to present more content to the user, the service provider provides the content provider via the presentation history correction unit 12054. The presentation history stored in the presentation history storage means 12051 related to the content owned by is deleted. Thereby, it becomes possible to make it easier for the user to present the content of the content provider who has collected additional consideration.

[Application Example 6-4-6]
In this application example, in a portal site on the Internet, the service provider that manages the portal is different from the content provider that manages the content provided on the portal. A case where a portal site is constructed by depositing content will be described. The content provider pays the service provider when the content entrusted to the service provider is presented to the user, or pays the service provider when the user uses the content. As a result, the owned content can be provided to the user.

  In the information presentation apparatus according to this application example, when the content provider wants to present more content to the user, the content provider pays an additional fee to the service provider, and the content is displayed via the presentation history correction unit 12054. The presentation history stored in the presentation history storage unit 12051 related to the content owned by the provider is deleted. Thereby, it becomes possible to make it easy to present the owned content to the user by paying an additional consideration. Note that the entity that instructs the presentation history correction unit 12054 to delete the history may be either a service provider or a content provider.

  FIG. 33 is a block diagram showing the configuration of the seventh embodiment of the present invention.

  In the information presentation apparatus according to the present embodiment, each user uses the search session calculation means 11 by the information presentation apparatus according to the first embodiment and the determination results by the search session determination means 7 of the search session calculation means 11. A search session counting means 13 for counting the number of search sessions (number of times of use) is further provided.

  FIGS. 34A and 34B show search session count tables 122 and 123 used for search session counting as an example of search session counting in the search session counting means 13.

  The search session count unit 7 sequentially updates the search session count tables 122 and 123 in response to the search session determination result in the search session calculation unit 11.

  For example, in the state of the search session count table 122 shown in FIG. 34A, the user USER1 newly performs a search with a search query corresponding to the search session SESSION1, and the search session calculation means 11 is the search session SESSION1. Assume that you have determined. At this time, the search session count means 13 adds the number of times of use corresponding to SESSION1 in the search session corresponding to the user USER1 to the search session count table 122, and the result is the search session count shown in FIG. Store as table 123.

  Note that the search session counting method based on the search session count tables 122 and 123 is an example of an implementation method, and the counting may be performed by another method.

[Application Example 7-1]
A first application example in this embodiment will be described. In this application example, a method of using a search session counted by the search session counting unit 13 will be described.

  As a usage method of the search session counted by the search session counting means 13, a usage method in which the user is charged for the usage fee of the information providing apparatus according to the number of search sessions used by the user can be considered. Until now, since it was not possible to accurately count search sessions, when trying to charge the usage fee of the information providing device to the user, a method of charging based on the usage time of the information providing device or use of a search query A method of charging based on the number was adopted. On the other hand, in the information presentation apparatus according to this application example, it is possible to realize a charging process for each search session.

  Further, as a method of using the search session counted by another search session counting means 13, a method of using the user to give points for using the information providing apparatus to the user according to the number of search sessions used by the user is considered. It is done. Until now, since it was not possible to accurately count the search sessions, when trying to give a point as a use benefit of the information providing device to the user, a method of granting based on the use time of the information providing device, a search query A method of granting based on the number of uses has been adopted. On the other hand, in the information presentation device according to this application example, it is possible to realize point grant processing for the user in units of search sessions.

  Further, by collecting and analyzing the search sessions counted by the search session counting means 13, the user tendency in the search behavior in the information providing device can be analyzed accurately. Until now, since the search session was not determined, the usage situation was analyzed for each search query. However, in the analysis of only the search query, the analysis was limited to a discrete state. On the other hand, in the information presentation device according to this application example, it is possible to integrate a series of input search queries and collect them as a search session, and analyze a usage situation for each search session. Can be continuously analyzed.

[Other Examples]
The first to seventh embodiments described above are flowcharts shown in FIG. 2, FIG. 10, FIG. 12, FIG. 14, FIG. 15, FIG. It can also be realized by executing a program for implementing the above-described operation. In this case, the search session calculation unit 11, the presentation information generation unit 124, the search session count unit 13, and the like are mounted on a computer.

  In addition, the information presentation apparatus according to each of the embodiments described above has a physical configuration, a hardware, and the like as long as the processing (function) of each unit (which constitutes each unit, each unit, and the like) that is a component can be realized. The hardware (circuit) and software (program) configurations are not particularly limited. For example, it can be provided in any form such as individually configuring individual circuits, units, or program parts (program modules, etc.), or integrally configured in one circuit or unit. These forms can be appropriately selected according to circumstances such as the function and application of the apparatus actually used.

  In addition, an information presentation method including processing steps for performing the same processing as these corresponding to the function of each unit included in the information presentation device described above is also included in the scope of the present invention.

  Furthermore, at least a part of the processing included in each of the units included in the information presentation device described above may be realized by software processing by a computer such as a processor having a CPU (Central Processing Unit). In this case, a program (information presentation program) for causing the computer to function is included in the category of the present invention. This program is not limited to a program in a format that can be directly executed by the CPU, but includes a program in various forms such as a source format program, a compressed program, and an encrypted program. In addition, this program operates in cooperation with a control program such as an OS (Operating System) and firmware that controls the entire apparatus, or is incorporated into a part thereof and constitutes an application program that operates integrally. Any form such as a software component (software module) can be provided. Furthermore, when this program is used by being mounted on a computer such as a terminal or a server having a communication function for communicating with an external device via a wireless or wired line, for example, from an external node such as a server connected on the line. It can also be downloaded and installed on a recording medium in its own device. These forms can be appropriately selected according to circumstances such as the function and application of the apparatus actually used.

  A computer-readable recording medium that records the above computer program is also included in the scope of the present invention. In this case, the recording medium can be provided in any form such as a memory such as a ROM (Read Only Memory) that is used in a fixed manner in the apparatus or a portable type that can be carried by the user. is there.

  The present invention has been described with reference to the embodiments. However, the present invention is not limited to the above embodiments, and the configuration and details of the present invention can be understood by those skilled in the art within the scope of the present invention. Various changes can be made.

  The present invention can be applied to applications such as an information search device that searches data stored in a database and an information recommendation device that recommends suitable information from data stored in a database. The target data is not limited to the restaurant information used in the description in the above embodiment, but is video content such as TV programs, commercials, and movies, text information content such as patent documents and papers, and office business applications. Any application can be applied as long as the application information is stored in a database and a content list is presented to the user.

It is a block diagram which shows the whole structure of the information presentation apparatus which concerns on the 1st Example of this invention. It is a flowchart which shows operation | movement of the information presentation apparatus which concerns on 1st Example of this invention. It is a figure which shows the example of the search query stored in the search query history storage means in the information presentation apparatus which concerns on 1st Example of this invention. (A) And (b) is a Venn diagram describing the relationship between a content set and a synthesized content set in the information presentation apparatus according to the first embodiment of the present invention. (A) And (b) is a figure which shows the relationship between a content set and the number of contents contained in the information presentation apparatus which concerns on 1st Example of this invention. In the information presentation apparatus concerning the 1st example of the present invention, it is a figure showing the example of the totalization table which totaled the number of the contents set which belongs about each contents. It is a figure explaining the table | surface which shows the score of each content in each content set searched with the search query in the information presentation apparatus which concerns on 1st Example of this invention. (A) And (b) is the Venn diagram expressing the content set shown in the table of Drawing 7 in the information presentation device concerning the 1st example of the present invention. It is a block diagram which shows the whole structure of the information presentation apparatus which concerns on the 2nd Example of this invention. It is a flowchart which shows operation | movement of the information presentation apparatus which concerns on the 2nd Example of this invention. It is a block diagram which shows the whole structure of the information presentation apparatus which concerns on the 3rd Example of this invention. It is a flowchart which shows operation | movement of the information presentation apparatus which concerns on the 3rd Example of this invention. It is a block diagram which shows the whole structure of the information presentation apparatus which concerns on the 4th Example of this invention. It is a flowchart which shows operation | movement of the information presentation apparatus which concerns on the 4th Example of this invention. It is a flowchart explaining the operation | movement of the variation of the search session determination in a comprehensive determination means in the information presentation apparatus which concerns on the 4th Example of this invention. It is a flowchart explaining the operation | movement of the variation of the search session determination in a comprehensive determination means in the information presentation apparatus which concerns on the 4th Example of this invention. It is a block diagram which shows the whole structure of the information presentation apparatus which concerns on the 5th Example of this invention. It is a flowchart which shows operation | movement of the information presentation apparatus which concerns on the 5th Example of this invention. In the information presentation apparatus concerning the 5th example of the present invention, it is a block diagram showing the internal configuration of the presentation information generation means. It is a flowchart which shows operation | movement of an example of the method of determining the content recommended to a user according to a search session by the recommended content production | generation means in the information presentation apparatus which concerns on 5th Example of this invention. It is a flowchart which shows operation | movement of an example of the method of determining the content recommended to a user according to a search session by the recommended content production | generation means in the information presentation apparatus which concerns on 5th Example of this invention. In the information presentation apparatus concerning the 5th example of the present invention, it is a figure explaining the advertisement registration table which defines the relation between the advertising content recommended to a user, and the target search query which wants to show each advertising content. It is a block diagram which shows the whole structure of the information presentation apparatus which concerns on the 6th Example of this invention. It is a flowchart which shows operation | movement of the information presentation apparatus which concerns on the 6th Example of this invention. It is a figure which shows an example of the content information stored in a content storage means in the information presentation apparatus which concerns on the 6th Example of this invention. In the information presentation apparatus concerning the 6th example of the present invention, it is a figure showing an example of presentation history information stored in a presentation history storage means. In the information presentation apparatus concerning the 6th example of the present invention, it is a figure showing the example of a screen of the contents list presented by contents list presentation means. In the information presentation apparatus concerning the 6th example of the present invention, it is a figure showing presentation history information stored in presentation history storage means. In the information presentation device concerning the 6th example of the present invention, it is a figure showing browsing history information stored in presentation history storage means. In the information presentation apparatus concerning the 6th example of the present invention, it is a figure showing the example of a screen of the contents list presented by contents list presentation means. In the information presentation apparatus concerning the 6th example of the present invention, it is a block diagram showing composition at the time of not providing a base sorting means. In the information presentation apparatus concerning the 6th example of the present invention, it is a block diagram showing composition when a presentation history correction means is provided in a presentation history management means. It is a block diagram which shows the whole structure of the information presentation apparatus which concerns on the 7th Example of this invention. (A) And (b) is a figure explaining an example of the search session count table used for the search session count counting by the search session count means in the information presentation apparatus according to the seventh embodiment of the present invention. It is a figure for demonstrating the problem of related technology.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Search query storage means 2 Content search means 3 First content set storage means 4 Search query history storage means 5 Content set composition means 6 Second content set storage means 7 Search session determination means 8 Composition history storage means 9 Search result history storage means 10 Comprehensive determination means 11 Search session calculation means (processing means)
12 presentation information generation means 1201 content storage means 1202 condition instruction means 1203 content list generation means 12031 condition collation means 12032 content selection means 12033 base sorting means 12034 correction sorting means 1204 content list presentation means 1205 presentation information management means 12051 presentation history storage means 12052 Condition designation presentation history extraction means 12053 Content designation presentation history extraction means 12054 Presentation history correction means 1206 Correction sorting control means 1211 Content list generation means 1212 Recommended content generation means

Claims (33)

A content set to be searched in a search query that is specified by a user of the search operation, and one content set synthesized using multiple contents set to be searched in the specified search query by past search operation of the user In comparison, the information presentation apparatus includes means for determining whether or not the specified search query is specified by a search operation for the same search purpose as the user's past search operation. First content set storage means for holding a content set searched by a search query designated by a user search operation;
Content set synthesizing means for synthesizing into one content set using a plurality of content sets searched by the search query specified by the user's past search operation;
Second content set storage means for holding the content set generated by the content set combining means;
The content set stored in the first content set storage unit and the content set stored in the second content set storage unit are compared, and the specified search query belongs to a search session for the same search purpose as the user's past search operation. Search session determination means for determining whether or not,
An information presentation device comprising: Search query storage means for storing a search query input by a user search operation;
Content search means for searching for content according to the search query;
First content set storage means for holding a content set searched by the content search means using a search query stored in the search query storage means;
Search query history storage means for storing a search query input by a past search operation of the user;
Content set combining means for combining a plurality of content sets searched by the content search means using a past search query stored in the search query history storage means into one content set;
Second content set storage means for holding the content set generated by the content set combining means;
The search query stored in the search query storage unit by comparing the content set of the first content set storage unit and the content set of the second content set storage unit is the same search purpose as the past search operation of the user. Search session determination means for determining whether or not the search session belongs;
An information presentation device comprising:   The information presentation apparatus according to claim 3, wherein an upper limit of the number of search queries stored in the search query history storage unit is set.   5. The information according to claim 3, further comprising synthesis history storage means for storing a history of the content set synthesized by the content set synthesis means in association with a history of a search query. Presentation device.   The search result history storage means for storing the content set searched by the content search means in association with the search query is further provided. Information presentation device.   The search session determination unit further includes a comprehensive determination unit that determines a plurality of search sessions, comprehensively determines a plurality of determination results, and outputs a final search session determination result. The information presentation device according to any one of claims 3 to 6.   The presentation information generation unit that generates information to be presented to the user based on the result of the search session determination unit, further comprising a presentation information generation unit according to any one of claims 3 to 7. Information presentation device.   The information presentation apparatus according to claim 8, wherein the presentation information generation unit performs information recommendation based on a result of the search session determination unit. A content set searched by a search query specified by a user search operation is compared with one content set synthesized using a content set searched by a search query specified by a past search operation of the user. Determining whether the specified search query is specified by a search operation for the same search purpose as the user's past search operation, and the specified search query is the user's past search operation. 2. A content list in which content is sorted with priority given to content that has not yet been presented is determined when it is determined that the search is designated by a search operation for the same search purpose. Information presentation device. A presentation history management means for managing a content list and an information presentation history;
Correction sorting means for referring to the presentation status of each content stored in the presentation history management means and sorting the contents so as to preferentially present the contents not presented to the user;
Correction sorting control means for controlling whether or not the correction sorting means is operable,
3. The information presentation according to claim 2, wherein the correction sorting control unit permits the operation of the correction sorting unit when the search session determination unit determines that the search purpose is the same as a past search operation. apparatus. The presentation history management means includes
Presentation history storage means for storing a list of contents generated by the correction sorting means together with the search query;
Condition-designated presentation history extraction means for extracting a content set with a presentation order from the presentation history information stored in the presentation history storage means using a search query as a key;
Using the content as a key from the information stored in the presentation history storage means, what number the search condition was sorted in what search condition, whether the content list presentation means presented to the user, whether the user selected Whether or not the content designation presentation history extraction means for extracting other information,
The information presentation apparatus according to claim 11, wherein A content list presenting means for presenting the content list sorted by the correction sorting means to the user and allowing the user to browse the content;
13. The information including at least one of a content list presentation state and a user browsing history in the content list presentation unit is included in the presentation history information and stored in the presentation information management unit. The information presentation apparatus described in 1. A base sorting means for sorting the contents using the attribute values of the contents in order of popularity, price, dictionary, or other predetermined order;
12. The contents are sorted by combining the sorting method defined by the base sorting means and the sorting method by the correction sorting means when sorting the contents to be presented. 14. The information presentation device according to any one of items 13. Further comprising a condition matching means for checking against the presentation history management means whether the search query is a search query used in the past,
If the search query is a search query stored in the presentation history management means, content is presented using the content list stored in the presentation history management means, and the search query is stored in the presentation history management means 15. The information presentation apparatus according to claim 11, wherein if the query is not a query, the content is sorted by the correction sorting means to sort the content. Content storage means for storing content information to be presented;
Condition indicating means for determining a search query as a condition for selecting content,
Content list generation means for generating a content list from the contents stored in the content storage means in accordance with the search query determined by the condition instruction means;
Content list presenting means for presenting the content list generated by the content list generating means to the user and allowing the user to browse the content;
A display history management means for managing a content list generated by the content list generation means and an information presentation history in the content list presentation means;
The content list generation means includes
Correction sorting means for referring to the presentation status of each content stored in the presentation history management means and sorting the contents so as to preferentially present the contents not presented to the user;
Correction sorting control means for controlling whether or not the correction sorting means can be operated,
4. The information presentation apparatus according to claim 3, wherein when the search session determination unit determines that the search purpose is the same as a past search operation, the correction sorting control unit permits the operation of the correction sorting unit. . The content list generation means includes
In addition to the correction sorting means, a condition matching means for obtaining a list of contents generated in the past by checking whether the search query determined by the condition input means is a search query used in the past;
A content selection unit that selects content that matches the search query as content for presentation from the content stored in the content storage unit when the search query is not found by the condition matching unit;
The information presentation apparatus according to claim 16, further comprising: In addition to the correction sorting means, the condition matching means, and the content selection means, the content list generation means sorts the contents using the attribute values of the contents in popularity order, price order, dictionary order, or other predetermined order. Further comprising a base sorting means,
17. The content sorting is performed by combining the sorting method determined by the base sorting unit and the sorting method by the correction sorting unit when sorting the contents to be presented. The information presentation device according to 17. The presentation history management means includes
A content list generated by the content list generating means, together with the search query, and further, a presentation history storage means for storing a content list presentation state and a user browsing history in the content list presentation means as presentation history information;
Condition-designated presentation history extraction means for extracting a content set with a presentation order from the presentation history information stored in the presentation history storage means using a search query as a key;
Using the content as a key from the information stored in the presentation history storage means, what number the search condition was sorted in what search condition, whether the content list presentation means presented to the user, whether the user selected Whether or not the content designation presentation history extraction means for extracting other information,
The information presentation apparatus according to claim 16, wherein the information presentation apparatus includes:   The presentation history management means further comprises a presentation history correction means for correcting the presentation history information stored in the presentation history storage means, in addition to the correction sorting means, the condition matching means, and the content selection means. The information presentation device according to any one of claims 16 to 19.   Content that has been recently presented with reference to the final presentation time of the content when the final presentation time of each content is recorded together as the presentation history stored in the presentation history management unit and sorted by the correction sorting unit 21. The information presentation apparatus according to claim 16, wherein the information presentation apparatus is more difficult to be presented. The content set searched by the search query specified by the user search operation is stored in the first content set storage means,
Synthesized one content set with a plurality of content sets to be searched in the past search queries of the user,
The synthesized content set is stored in the second content set storage means,
The content set stored in the first content set storage unit and the content set stored in the second content set storage unit are compared, and the specified search query belongs to a search session for the same search purpose as the user's past search operation. An information presentation method characterized by determining whether or not. The search query input by the user's search operation is stored in the search query storage means,
Search for content according to the search query stored in the search query storage means,
The retrieved content set is stored in the first content set storage means,
Storing the user's past search query in a search query history storage means;
Using a plurality of the search query history storage means is searched using past search queries stored in the content set and combined into a single content set,
The synthesized content set is stored in the second content set storage means,
The content set in the first content set storage unit, the by comparing the contents set in the second content set storage means, past search operations and the same search object of the search query stored in the search query storage means the user A method for presenting information, characterized in that it is determined whether or not it belongs to a search session.   24. The information according to claim 22, further comprising: determining a plurality of search sessions, comprehensively determining the plurality of determination results, and outputting a final search session determination result. Presentation method.   The information presentation method according to any one of claims 22 to 24, further comprising: generating information to be presented to the user based on a determination result of the search session.   Furthermore, when the specified search query is determined as a search operation for the same search purpose as the past search query as the determination result of the search session, the content is sorted with priority on the content not yet presented The information presentation method according to any one of claims 22 to 25, wherein a list is presented.   27. The information presentation according to claim 26, wherein when the content is sorted with priority on the content that has not yet been presented, the content that is recently presented with reference to the last presentation time of the content is less likely to be presented. Method. A process of storing a content set searched by a search query specified by a user search operation in a first content set storage unit;
A process of combining a plurality of content sets searched by the user's past search query into one content set;
A process of storing the combined content set in the second content set storage means;
The content set stored in the first content set storage unit and the content set stored in the second content set storage unit are compared, and the specified search query belongs to a search session for the same search purpose as the user's past search operation. Processing to determine whether or not
A program for presenting information, characterized by causing a computer to execute. A process for storing a search query input by a user search operation in a search query storage means;
Processing for searching for content according to a search query stored in the search query storage means;
Storing the retrieved content set in the first content set storage means;
Storing the user's past search query in a search query history storage means;
A process of combining a plurality of content sets searched using the user's past search queries stored in the search query history storage means into one content set;
A process of storing the combined content set in the second content set storage means;
The content set in the first content set storage unit, the by comparing the contents set in the second content set storage means, past search operations and the same search object of the search query stored in the search query storage means the user The process of determining if it belongs to the search session,
A program for presenting information, characterized by causing a computer to execute. In addition, a process for determining a plurality of search sessions;
A process of comprehensively judging a plurality of judgment results and outputting a judgment result of a final search session;
30. The information presentation program according to claim 28 or 29, wherein the program is executed by a computer.   31. The information presentation program according to claim 28, further causing a computer to execute processing for generating information to be presented to a user based on a determination result of a search session. .   Furthermore, when the specified search query is determined as a search operation for the same search purpose as the past search query as the determination result of the search session, the content is sorted with priority on the content not yet presented 32. The information presentation program according to claim 28, further causing a computer to execute a process of presenting a list.   Furthermore, when sorting content with priority on content that has not yet been presented, the computer is caused to execute a process that makes it more difficult to present a content that has been recently presented with reference to the last presentation time of the content. 32. The information presentation program according to 32.

Images