JPH1166078A - Retrieval request embodiment method/device and storage medium storing retrieval, request embodiment program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To realize efficient information retrieval by automatically embodying the vague retrieval request of a user to a retrieval request word group fitted to the user through the use of individual information of the user and information on an environment where the user exists. SOLUTION: A word for the retrieval request is inputted to a request embodiment part 3 from the user. The thesaurus knowledge retrieval part of the request embodiment part 3 detects a partial tree where the word of the retrieval request inputted from object thesaurus knowledge 5 as a root and generates the set of words adjusted to parts equivalent to leaves. The restriction knowledge retrieval part of the request embodiment part 3 takes out one word from the set, retrieves an individual restriction knowledge base 4 by using it and obtains knowledge on the object which the word shows. The judgment part of the request embodiment part 3 compares the condition of the respective item descriptions of obtained individual restriction knowledge with individual information which is obtained by an individual/environment information acquirement part and is described in a user information part 1 and environment information described in an environment information part 2. When the condition is satisfied and the word which is not processed does not exist, the word group for the retrieval request is outputted.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a method and an apparatus for embodying a search request and a storage medium storing a search request embodying program, and more particularly to an information search system for embodying a user's vague search request. And a storage medium storing a search request materializing program.

[0002]

2. Description of the Related Art The recent development of the Internet has made it possible to easily access various types of information.
There is a great deal of difficulty for a user to obtain necessary information accurately. For this reason, information is classified into a hierarchical structure, such as “yahoo”, so that the user can easily find necessary information, or a search is performed using a keyword specified by the user.
A system for presenting a WWW page including the keyword is used.

[0003]

However, such a system works well if the user's requirements are sufficiently embodied, but has a problem that it is difficult to use when the requirements are vague. . For example, suppose a person who wants to go somewhere every weekend wants information. In this case, even if there is a hierarchical classification, there are so many things under the play that it is very difficult to look around. Also, even if a page containing "play" is presented in a keyword search,
Then, information or the like of “ski” which is a specific candidate of “play” cannot be obtained.

Here, in order to present information on specific candidates such as "ski", there is a method of searching for all the words under "play" as keywords using a thesaurus or the like. That's still a huge number of pages, and searching for the right one requires a lot of work. As described above, in the conventional information system, there is a problem that when a user has a vague request, it is not possible to efficiently obtain appropriate information.

[0005] The present invention has been made in view of the above points,
When the user's request is vague, the information presented to the user can be limited by using the user's personal information and information on the environment where the user is located and not presenting anything that seems unrelated to the user at that time. It is an object of the present invention to provide a search request materializing method and apparatus capable of presenting such a keyword group and a storage medium storing a search request materializing program.

[0006]

SUMMARY OF THE INVENTION The present invention provides a method for embodying a user's vague search request and extracting a keyword that allows easy access to information in an information search system. When a word is input, a lower-level word group of the input word is obtained, words that are incompatible with the user information or the user's environment information are deleted from the word group based on a predetermined condition, and a matching word group is deleted. Output as a keyword for searching the information search system.

FIG. 1 is a diagram for explaining the principle of the present invention. According to the search request embodying method of the present invention, when a word of a search request is input (step 1), a search is made for an object thesaurus using the input word, and a lower word group of the word is obtained (step 2). Acquire constraint knowledge on the thing represented by each word by the word group (step 3), acquire information on the user, and acquire environmental information of the user (step 4).
It is determined whether the information satisfies the constraint knowledge (step 5). If the information is not satisfied, the word expressing the unsatisfied thing is deleted from the word group (step 6). , A word group is output as a keyword for a search request (step 7).

FIG. 2 is a diagram showing the principle of the present invention. The present invention relates to a search request embodying device that embodies a user's vague search request and extracts a keyword that allows easy access to information in an information search system, wherein a search request word input that inputs a search request from a user is provided. Means 10;
A request materializing unit that deletes a word that is incompatible with the user information or the user's environment information from the word group based on a predetermined condition, and outputs the matching word group as a keyword for searching the information search system. .

Further, the request embodying device of the present invention comprises a user information storage means 1 which describes personal information about a user;
Environment information storage means 2 which describes environment information relating to the situation where the user is placed, and object thesaurus knowledge 5 representing the classification of the object
And an individual constraint knowledge base 4 describing individual constraint knowledge about an object, and a general constraint knowledge base 6 describing a relationship between personal information and environmental information and individual constraint knowledge. Reference numeral 20 includes a request creating unit that creates a more specific search request from a user's vague search request based on personal information, environmental information, knowledge of a physical thesaurus, and knowledge of individual constraints.

[0010] The above-mentioned request creating means searches the thesaurus knowledge searching means 21 using the word of the input search request, and searches the individual constraint knowledge base by the word of the search request and is expressed by each word. Constraint knowledge search means 22 for obtaining constraint knowledge about an object, personal / environment information search means 23 for searching environment information storage means and user information storage means to obtain user's environment information and personal information, and thesaurus knowledge search means 21 and personal / environmental information search means 23
Determining means 24 for determining whether or not the information obtained from the information matches the constraint condition obtained by the constraint knowledge searching means 22
If the words are not matched by the determination means 24, the words that express an object that does not satisfy the constraint conditions from the word group are deleted from the word group, and the words are deleted by the word deletion means 25. And a keyword output unit 26 that outputs the remaining word group as a keyword.

[0011] The present invention is a storage medium storing a search request materializing program for embodying a user's vague search request and extracting a keyword capable of easily accessing information in an information search system. A search request input process for inputting a request, and a word that is incompatible with information about the user and the user's environment information is deleted from the word group based on a predetermined condition, and the matching word group is used as a keyword for searching the information search system. And a request materialization process for outputting.

[0012] The above-described request embodying process includes personal information describing personal information about the user, environmental information describing environmental information regarding the situation where the user is placed, a thesaurus knowledge representing the classification of a product, and a product related to the product. There is a request creation process for creating a more specific search request from a user's vague search request based on individual constraint knowledge.

The request creation process includes a thesaurus knowledge search process for searching for the object thesaurus knowledge using the input search request word, and a search for individual constraint knowledge using the search request word, and searching for the object represented by each word. Constraint knowledge retrieval process for acquiring constraint knowledge, environmental information retrieval process for acquiring environmental information, and whether the information acquired from the thesaurus knowledge retrieval process and the environment information retrieval process matches the constraints acquired by the constraint knowledge retrieval process And a word deletion process of deleting a word expressing an object that does not satisfy the constraint condition from the word group, and deleting the word by the word deletion process if the word does not satisfy the constraint condition by the judgment process. And a keyword output process of outputting the remaining word group as a keyword as a result.

As described above, according to the present invention, when a word of a user's search request is input, the request embodying means checks the object thesaurus knowledge, finds a match with the input word, and Get a group of words located below the word. With this, the request embodying means searches the individual constraint knowledge base using the word group, acquires knowledge about an object represented by each word in the word group, acquires user information and user environment information. By using the knowledge in the general constraint knowledge base, it is determined whether or not the condition of each item of the constraint knowledge on the object is satisfied based on the above information. For those that do not satisfy the constraints, the words that express them are removed from the above word group, and the finally left word group is output as a keyword for a specific search request. This makes it possible to automatically embody the search request so as not to present unnecessary information to the user by removing information that is incompatible with the user's information and the environment at that time.

[0015]

FIG. 3 shows a system configuration to which the present invention is applied. As shown in the figure, when a search request is input from the user to the search request input device 10, the search request embodying device 20 uses a word group constituting the search request to perform an appropriate search according to the state and environment of the user. Generate keywords,
The keyword is sent to the information search system 30.
Thereby, the information search system 30 can directly or directly connect the information database 40,
50 or another information search system 60 connected to the information search system 30 via a network or the like.

FIG. 4 shows the configuration of a search request embodying apparatus of the present invention. The search request materializing device 20 shown in FIG. 1 includes a user information unit 1, an environment information unit 2, a request materializing unit 3, an individual constraint knowledge base 4, an object thesaurus knowledge 5, and a general constraint knowledge base 6. The user information unit 1 is configured by a database or a file, and stores personal information (age, gender, etc.) of a user who uses the present apparatus 20.

The environment information section 2 is composed of a database or a file, and stores environment information (season, weather, etc.) when the apparatus 20 is used. The request materializing unit 3 uses the knowledge of the individual constraint knowledge base 4, the thing thesaurus knowledge 5, and the general constraint knowledge base 6 to input the words for the search request into the user information unit 1, the environment information unit. 2 is embodied into a word group serving as a search request suitable for a situation determined by each information.

The individual constraint knowledge base 4 is composed of a database and files, and describes various necessary conditions for each object. The thing thesaurus knowledge 5 is constituted by a database or a file, and is knowledge that classifies objects hierarchically. The general constraint knowledge base 6 is composed of a database or a file, and describes knowledge about the relationship between each item of personal information and environmental information and each item of individual constraint knowledge.

FIG. 5 shows a detailed configuration of the above-mentioned request embodying unit. The request materializing unit 3 is a thesaurus knowledge searching unit 3
1. Constraint knowledge search unit 32, personal / environment information acquisition unit 33,
It comprises a determination unit 34 and a keyword output unit 35. Next, the operation of the search request embodying apparatus having the above configuration will be described. FIG. 6 is a flowchart showing the operation of the search request embodying apparatus of the present invention.

Step 101) First, the request specifying unit 3
The user inputs a word for a search request. Step 102) Next, the thesaurus knowledge search unit 31 of the request specification unit 3 finds a subtree rooted at the word of the search request input from the physical thesaurus knowledge 5, and finds a part corresponding to the leaf (thesaurus) A word set U corresponding to the lowermost portion is generated, and the output word group set O is made an empty set.

Step 103) The constraint knowledge search unit 32 of the request specification unit 3 selects one word from the set U (referred to as t).
The individual constraint knowledge base 4 is retrieved using the word t and the knowledge about the object represented by t is obtained. Step 104) The determination unit 34 of the request embodying unit 3 compares the condition described in each item of the individual constraint knowledge acquired in step 103 with the user information unit 1 acquired by the individual / environment information acquisition unit 33. The described personal information is compared with the environment information described in the environment information section 2 to determine whether or not the condition is satisfied. In this determination, the knowledge described in the general constraint knowledge base 6 is used as needed.

Step 105) If the condition is satisfied, add the word t to the set O. Step 106) If there is an unprocessed word in the set U, go to step 103; otherwise, go to step 1
Shift to 07. Step 107) The keyword output unit 35 of the request specifying unit 3 outputs the set O as a group of words for the specified search request.

[0023]

An embodiment of the present invention will be described below with reference to the drawings. FIG. 7 shows an example of personal information in the user information section according to one embodiment of the present invention. The user's personal information is represented by a set of item names such as age and gender, and values of the items. The user information may be registered in advance, or may be input for each item determined by the system when the user uses the system.

FIG. 8 shows an example of environment information of the environment information section according to one embodiment of the present invention. The environment information is represented by a set of item names such as season and weather, and a value of the item name. It is assumed that the environment information has been given in advance. FIG. 9 is an example of knowledge of the individual constraint knowledge base according to one embodiment of the present invention. The individual constraint knowledge describes a value to be satisfied or a value range for each item such as the required number of people and cost for what is expressed by each word such as skiing and sea bathing. Individual constraint knowledge is given in advance.

FIG. 10 shows an example of knowledge of an object thesaurus according to one embodiment of the present invention. The thesaurus has a hierarchical structure in which the higher ones represent abstract and vague concepts, and the lower ones represent concrete ones. The thing thesaurus knowledge is given in advance. FIG. 11 is an example of knowledge in a general constraint knowledge base according to one embodiment of the present invention. The general constraint knowledge is knowledge for determining whether the user information and the environment information satisfy the individual constraint knowledge. For example, in the same figure, IF kankyo (season, X) AND kobetsu (season, Y) AND
X {Y AND Y} "Yearly" THEN NOGOOD indicates that the value of the item "season" in the environmental information section 2 is X, and the value of the item "season" in the individual constraint knowledge base 4 is Y when "Yearly" is not used. Means that the condition is not satisfied.

Next, the operation of the search request embodying apparatus of the present invention will be described using a specific example. Here, a case where the word "play" is input as a word for a search request will be described as an example. First, the word “play” is input to the request embodying unit 3. The request embodying unit 3 refers to the thesaurus in the thing thesaurus knowledge 5 and uses “play” as a route in FIG.
Find a subtree as shown in. Then, a set U of the words (sea bathing, skiing, amusement park, museum, mahjong, etc.) of the leaves of the partial tree is created. The output word set O is an empty set.

Next, the request embodying unit 3 extracts one word from the set U. Here, "sea bathing" is used. The request materializing unit 3 refers to the column of “sea bathing” of the knowledge as shown in FIG.
Get the conditions for seasons, inside and outside. Next, the request specifying unit 3 refers to the personal information shown in FIG. 7 in the user information unit 1 and the environment information shown in FIG. By using the knowledge in the general constraint knowledge base 6, it is determined whether the conditions regarding the required number of people, cost, season, inside and outside, and the like are satisfied. Here, since the condition relating to the season is satisfied, it is determined that “sea bathing” is an unnecessary word in the current situation.

The request materializing section 3 performs the same processing for "skiing". Here, "skiing" does not satisfy the cost condition. Subsequently, when the same processing is performed for the “amusement park”, it is added to the output word set O because all the conditions are satisfied. Similarly, "Museum"
The process proceeds with “Mahjong”, and when there are no more unprocessed words in the set U, an output word set O is output. In this example, "amusement park" and "art museum" are output.

According to the above procedure, a more specific search request word group (keyword for accessing the information search system) “amusement park” and “museum” are automatically obtained from the vague search request word “play”. In the above embodiment, the method of adding words satisfying all the conditions to the output word set O has been described. However, the present invention is not limited to this example. It is also possible to delete unmatched words and output the remaining words as keywords.

In the above embodiment, the search request embodying unit 3 has been described. However, it is assumed that each of the functions is provided as shown in FIG. As described above, according to the present invention, by using thesaurus knowledge, even if an input word for a search request is abstract and vague, it is replaced with a specific word group, and further, individual restrictions By using knowledge and general constraint knowledge, it is possible to automatically materialize a user's search request so as to remove unnecessary search request words in the search situation expressed by user information and environmental information. It becomes possible.

Further, in the present invention, the request embodying unit 3 in the above-described embodiment uses the process of searching for information and the word expressing the thing that does not match the constraint condition from the word group obtained by the search process. A disk drive, floppy disk, CD-ROM, or the like connected to a computer used in the information search system is constructed as a program including a keyword generation process of generating and outputting the remaining words as keywords by deleting the words. Can be stored in a portable storage medium.

The present invention is not limited to the above embodiment, but can be variously modified and applied within the scope of the claims.

[0033]

As described above, according to the present invention, a group of search request words suitable for a user can be obtained by using a user's personal information and information on the environment in which the user is placed. By automatically realizing the information to, even if the user has only a vague search request, there is an effect of enabling efficient information search.

[Brief description of the drawings]

FIG. 1 is a diagram for explaining the principle of the present invention.

FIG. 2 is a principle configuration diagram of the present invention.

FIG. 3 is a system configuration diagram to which the present invention is applied.

FIG. 4 is a configuration diagram of a search request embodying apparatus of the present invention.

FIG. 5 is a detailed configuration diagram of a request embodying unit of the present invention.

FIG. 6 is a flowchart of the operation of the search request embodying device of the present invention.

FIG. 7 is an example of personal information in a user information section according to an embodiment of the present invention.

FIG. 8 is an example of environment information of an environment information unit according to an embodiment of the present invention.

FIG. 9 is an example of knowledge of a personal constraint knowledge base according to an embodiment of the present invention.

FIG. 10 is an example of knowledge of an object thesaurus in one embodiment of the present invention.

FIG. 11 is an example of knowledge in a general constraint knowledge base according to one embodiment of the present invention.

[Explanation of symbols]

REFERENCE SIGNS LIST 1 user information section, user information storage means 2 environment information section, environment information storage means 3 request materialization section 4 individual constraint knowledge base 5 object thesaurus knowledge 6 general constraint knowledge base 10 search request word input means, search request input device 20 request Materializing means, request materializing section, search request materializing device 21 Thesaurus knowledge searching means 22 Restricted knowledge searching means 23 Personal / environmental information searching means 24 Judging means 25 Word deleting means 26 Keyword output means 30, 60 Information searching system 31 Thesaurus Knowledge search unit 32 Constraint knowledge search unit 33 Personal / environment information acquisition unit 34 Judgment unit 35 Keyword output unit 40, 50, 70 Information database

Claims (8)

[Claims] 1. A search request embodying method for embodying a user's vague search request and extracting a keyword that allows easy access to information in an information search system. A word group lower than the selected word is obtained, a word that is incompatible with user-related information or user environment information is deleted from the word group based on a predetermined condition, and the information search system searches for a word group that matches the word group. A search request embodying method characterized by outputting as a keyword for performing a search. 2. When a word of the search request is input, an object thesaurus knowledge is searched by the input word, a lower word group of the word is acquired, and an object represented by each word by the word group is obtained. Information on the user, it is determined whether or not the environment information of the user satisfies the constraint knowledge. If the information is not satisfied, a word expressing an unsatisfied object is obtained from the word group. 2. The search request materializing method according to claim 1, wherein the word group is output as a keyword for a search request when the word group is deleted and satisfied. 3. A search request embodying device that embodies a user's vague search request and extracts a keyword that can easily access information in an information search system, wherein a search request word for inputting a search request from a user is provided. Request means for deleting words that are incompatible with user-related information or user environment information from the word group based on a predetermined condition, and outputting the matching word group as a keyword for searching the information search system; And a search request embodying device. 4. A user information storage unit that describes personal information about a user, an environment information storage unit that describes environment information about a situation where a user is placed, an object thesaurus knowledge representing an object classification, and an individual constraint on an object. An individual constraint knowledge base describing knowledge, the personal information and the environment information, and a general constraint knowledge base describing a relationship between the individual constraint knowledge, wherein the request embodying unit includes the personal information, Said environmental information,
4. The search request materializing apparatus according to claim 3, further comprising a request creation unit that creates a more specific search request from the user's vague search request based on the thing thesaurus knowledge and the individual constraint knowledge. 5. The request creating means, a thesaurus knowledge searching means for searching the article thesaurus knowledge by the input word of the search request, and searching the individual constraint knowledge base by the word of the search request, Constraint knowledge search means for acquiring constraint knowledge about an object represented by: a personal / environment information search means for searching the user information storage means and the environment information storage means to obtain personal information and environment information of the user; Determining means for determining whether information obtained from the thesaurus knowledge searching means and the individual / environmental information searching means matches the constraint conditions obtained by the constraint knowledge searching means; A word deletion unit for deleting a word expressing an object that does not satisfy the constraint condition from the word group, from the word group; 5. The search request embodying device according to claim 4, further comprising keyword output means for outputting, as a keyword, a group of words remaining as a result of the deletion by the means. 6. A storage medium storing a search request materializing program for embodying a user's vague search request and extracting a keyword capable of easily accessing information in an information search system, wherein the storage medium stores a search request from the user. A search request input process for inputting, a word that is incompatible with information about a user or environment information of the user is deleted from the word group based on a predetermined condition, and a word group that matches is used as a keyword for searching the information search system. A storage medium storing a search request materializing program, comprising: a request materializing process for outputting. 7. The request embodying process includes: personal information describing personal information about a user; environmental information describing environment information about a situation where the user is placed; a product thesaurus knowledge representing a classification of a product; 7. The storage medium according to claim 6, further comprising a request creation process for creating a more specific search request from the user's vague search request based on the constraint knowledge. 8. The request creating process includes: a thesaurus knowledge search process for searching for the object thesaurus knowledge using the input search request word; and searching for the individual constraint knowledge using the search request word. A constraint knowledge search process for acquiring constraint knowledge of an object to be represented; an environment information search process for acquiring the environment information; and information obtained from the thesaurus knowledge search process and the environment information search process. A determination process of determining whether or not the acquired constraint condition is satisfied; and, if the determination process does not match, a word expressing an object that does not satisfy the constraint condition from the word group is extracted from the word group. A word deletion process to be deleted, and a word group remaining as a result of the word deletion process output as a keyword A storage medium storing a search request materializing program according to claim 7, including a keyword output process.