JP4892971B2 - Information retrieval apparatus and program - Google Patents

Description

  The present invention relates to an information search apparatus and a program.

  Information retrieval in the Internet and various databases (hereinafter abbreviated as “DB” as appropriate) is mainly performed by the user by inputting a keyword (character string). The information including the input keyword and the keyword are supported. The attached information is searched and displayed.

  As an approach for obtaining a desired search result, the user first considers a keyword suitable for the search from his / her knowledge, and executes the search by inputting the keyword. If there are a plurality of search results, the search is further performed by additionally inputting keywords. The keywords input by the user in the past are stored as a history, and the user can select a keyword from the history and perform a search without bothering to input a character string.

In addition, as a technique for facilitating information retrieval from the Internet and databases that have a huge amount of information, information should be hierarchized within a pre-restricted category (topic) such as weather, traffic, and meals. Thus, a technique for assisting the user so that a desired search result can be easily obtained by performing a search within the category has been put into practical use (see Patent Document 1).
JP 2002-169818 A

  However, there are individual differences for each keyword entered by the user, so that the entered keyword is not appropriate, or additional input of the keyword at the time of the refinement search is omitted, and the desired search result May not be obtained. For example, even if the user has input a keyword that has not been input at the time of a past search, the user cannot confirm the keyword unless referring to the input history one by one.

  Further, as in the technique of Patent Document 1, the search range can be narrowed down to the category by hierarchizing information for each category, but this category is statistically obtained as a general classification. However, it is not in line with the interests of each user and preferences such as customs at the time of search. For this reason, for example, when going out, a user who examines traffic information to the destination and the weather searches the traffic information to the destination in the category of traffic, and further determines the weather of the destination in the category of weather. Must search.

  Thus, even if there is a difference in preference for each user in the combination of categories used in the search, the user must select a plurality of categories one by one, enter the same keyword, and repeat the search. The operation was complicated.

  The present invention has been made in view of the above-described problems, and its purpose is to support user keyword input at the time of search and to search for a category according to individual user preferences. It is to realize an information retrieval apparatus that can be used.

In order to solve the above problems, an information search device according to claim 1 is provided:
Category storage means for storing a plurality of preset categories;
Category selection means for selecting a category stored in the category storage means;
Keyword storage means for storing a plurality of keywords associated with the category selected by the category selection means ;
Keyword selection means for selecting a predetermined keyword from a plurality of keywords stored by the keyword storage means;
Information search means for searching for information in the first category based on a predetermined keyword selected by the keyword selection means for a first category that is one category selected by the category selection means;
After performing a search for information in the first category by the information search means, when the second category different from the first category is continuously selected by the category selection means, the first category The third category is merged with the second category to create a third category for performing an information search based on the keywords of both the first category and the second category, Category creating means for adding the category to the category storage means;
It is characterized by having.

The invention according to claim 2 is the invention according to claim 1,
The category creating means creates the third category based on a combination of a plurality of categories selected by the category selecting means.

The invention according to claim 3 is the invention according to claim 1 or 2,
The category creating means creates the third category when a continuous information search of the first category and the second category is performed a predetermined number of times or more .

The invention according to claim 4 is the invention according to any one of claims 1 to 3,
Before SL category creating means, a pre-Symbol plurality of keywords associated with the each of the third category was created from the first category and the second category read from the keyword storage means, the plurality of the same keyword of the keyword in association with the third category are characterized and this with a keyword preparing means you added to the keyword storage means.

The invention according to claim 5 is the invention according to any one of claims 1 to 4,
Before Symbol keyword storing means, a history storage means for storing Kiki keyword before used in information retrieval is our row by prior Symbol information searching means in association with the category,
History notification means for notifying the keyword stored in the history storage means when the keyword selection means selects the keyword;
The a is characterized in Rukoto.

The invention according to claim 6 is the invention according to any one of claims 1 to 5,
The category associated with the selected category by the selecting means keyword read from the keyword storage means, and wherein, further comprising a notification unit configured to notify the user when selecting the pre-crisis keyword.

The invention according to claim 7 is the invention according to claim 6,
The notification unit, when the third category is selected by the pre-Symbol selection means, the keyword associated with the said third category the first category and the second category of creation source The information is read from the keyword storage means and notified.

The invention according to claim 8 is the invention according to any one of claims 1 to 7,
It is characterized by obtaining Bei Table Shimesuru display means in a size of the display in accordance with the selection frequency of the category by the category selection unit.

The information search device according to claim 9 is the invention according to any one of claims 1 to 8,
It is characterized by comprising a keyword acquisition means for acquiring the keyword used when searching for information within the category from another device via a communication line network .

The invention according to claim 10 is the invention according to any one of claims 1 to 9 ,
A category acquisition unit is provided for acquiring a category different from the plurality of categories stored in the category storage unit from a plurality of categories in which the other unit searches for information from the other unit via a communication line network. Yes.

The program according to claim 14 is a computer,
Category storage means for storing a plurality of preset categories;
Category selection means for selecting a category stored in the category storage means;
Keyword storage means for storing a plurality of keywords associated with the category selected by the category selection means;
Keyword selection means for selecting a predetermined keyword from a plurality of keywords stored by the keyword storage means;
Information search means for searching for information in the first category based on a predetermined keyword selected by the keyword selection means for a first category that is one category selected by the category selection means;
After performing a search for information in the first category by the information search means, when the second category different from the first category is continuously selected by the category selection means, the first category The third category is merged with the second category to create a third category for performing an information search based on the keywords of both the first category and the second category, Category creation means for adding the category of the category to the category storage means ,
It is characterized by making it function as.

  According to the present invention, when a new category is created based on a plurality of selected categories, and the new category is selected at the time of information search, the category of the creation source associated with the new category In order to perform information search, a category corresponding to the category selection pattern, that is, a category according to the user's preference is created. Further, when the user inputs a keyword, the keyword used when the other device searches for information in the category is acquired and notified, so that a keyword input hint at the time of search can be proposed to the user. Therefore, it is possible to realize an information search device that supports the user's keyword input and can perform a search according to the user's preference during search.

Embodiment
Hereinafter, embodiments of the information search apparatus of the present invention will be described in detail with reference to FIGS. The information retrieval device 1 (see FIG. 2 ) is a personal computer, car navigation device, mobile phone, PDA (Personal Digital Assistance) that can be connected to a communication network N such as the Internet, a public telephone line, or a LAN (Local Area Network). It is comprised by electronic devices, such as.

[Outline of information retrieval device]
First, an outline of the information search apparatus 1 will be described with reference to FIG. The information search device 1 performs various information searches based on the keyword input by the user, and acquires information including the keyword and information associated with the keyword from various servers and databases on the communication network N. The search result is provided to the user by display output or voice output. In order to narrow down the search range of the information search in advance, a plurality of categories for each search range are set.

  When the user selects a desired category and inputs a keyword to instruct execution of the search, an information search based on the keyword is performed within the selected category (hereinafter referred to as “selected category”). In the information search apparatus 1, the agent character proposes a keyword to the user so that the search result desired by the user can be reached, and performs information search according to the user's instruction.

  An agent character (hereinafter abbreviated as “character” as appropriate) is a general term for a function (software) that supports user information retrieval in an interactive format, and an anthropomorphic character is displayed on the display unit 34 (see FIG. 2). Display and interact with the user.

  In the present embodiment, the information search apparatus 1 includes a weather character CR1 that searches for information related to “weather”, a traffic character CR2 that searches for information related to “traffic”, and a meal character CR3 that searches for information related to “meal”. The character is preset in the initial state. This preset character is hereinafter referred to as “basic character”.

  FIG. 1 is an example of a character selection screen 340 for selecting a character. For example, as shown in FIG. 1A, the character selection screen 340 includes a weather character CR1, a traffic character CR2, and a meal character CR3, a main character display area AR1, a shopping character CR10, a leisure (amusement park) character CR12, and the like. Are displayed in the character list display area AR3 so that they can be selected.

  The main character display area AR1 is an area for displaying characters frequently used by the user for searching. Characters displayed in the main character display area AR1 are displayed in a size corresponding to the usage frequency. That is, in FIG. 1A, the weather character CR1 is most frequently used by the user, and is then used in the order of the traffic character CR2 and the meal character CR3. Characters that cannot be displayed in the main character display area AR1 are displayed in the character list display area AR3.

  When the user selects a desired character from a plurality of characters displayed on the character selection screen 340 and further inputs a keyword, information search within a category corresponding to the character is performed based on the keyword. . As a result, the information search device 1 performs a process in which the character performs an information search in the category that the character is good at, acquires information from the Internet or a database, and provides the information to the user. An anthropomorphic function is realized.

  When the user selects a character in a certain combination and the number of times the search is executed exceeds a predetermined number, a new character is created according to the combination. For example, when the search is performed five times or more using the weather character CR1 and the traffic character CR2, the outing character CR4 is newly created as shown in FIG.

  This outing character CR4 is a so-called fused character that inherits the category and the like that the weather character CR1 and the traffic character CR2 are good at. The outing character CR4 performs an information search based on the input keyword in two categories of weather and traffic, and provides a search result in the two categories to the user.

  In this way, a new character is created according to the user's preference such as the combination at the time of character selection and the usage frequency, and the user frequently inherits the characteristics of the character used for the search, so that the user can Search can be performed without selecting characters one by one.

[Configuration of information retrieval device]
Next, the functional configuration of the information search apparatus 1 will be described with reference to FIG. As illustrated in FIG. 2, the information search device 1 includes a control unit 10, an input unit 20 having a voice input unit 22 and an operation unit 24, an output unit 30 having a voice output unit 32 and a display unit 34, and communication. The unit 40 and the storage unit 50 are provided.

  The audio input unit 22 includes, for example, a microphone, an AF (Audio Frequency) amplifier, a sampler, an A / D (Analog-to-Digital) converter, and the like. The voice input unit 22 amplifies a voice signal representing the voice collected by the microphone, samples and performs A / D conversion. Then, the digital audio data representing the audio is generated and output to the control unit 10.

  The operation unit 24 includes a keyboard having cursor keys, numeric keys, various function keys, and the like, and outputs an operation signal of a key pressed by the user to the control unit 10. The operation unit 24 includes a pointing device such as a mouse or a touch panel, and controls a position signal of a position clicked on the screen of the display unit 34 by the mouse or a position touched by a user's fingertip or the like. To the unit 10.

  The audio output unit 32 includes, for example, a D / A (Digital-to-Analog) converter, an AF amplifier, a speaker, and the like, and D / A converts and amplifies the audio data supplied from the control unit 10, Sound is reproduced from the speaker based on the obtained analog signal.

  The display unit 34 is configured by a color LCD (Liquid Crystal Display) or an ELD (Electronic Luminescent Display), and displays various information related to the information search apparatus 1 and character images based on display data output from the control unit 10. indicate. The display unit 34 is integrally formed with a touch panel that is one of the operation units 24.

  The communication unit 40 includes a modem, a LAN interface, a USB, and the like, and performs data communication with various servers and other devices on the Internet via the communication network N according to instructions from the control unit 10.

  The control unit 10 includes, for example, a processor including a CPU (Central Processing Unit), a non-volatile memory such as a ROM (Read Only Memory) that stores a program executed by the processor, and a storage area serving as a work area of the processor And a volatile memory such as a RAM (Random Access Memory) having a memory, and generally manages instruction input to each functional unit and data communication between the functional units.

  More specifically, a program stored in the ROM is read based on an operation signal and a position signal output from the operation unit 24, and processing is performed based on the program. In addition, when voice data is output from the voice input unit 22, by performing a process for performing known voice recognition on the voice data, word candidates that may be represented by the voice data, The candidate score (degree of approximation) is specified.

  Then, reading a word whose score is equal to or greater than a predetermined value is read from a word database (not shown), a series of text data is created, the text data is analyzed, and an instruction content by a user's voice input The program is processed according to the contents of the instruction. The control unit 10 outputs, to the display unit 34, image data for causing the display unit 34 to display the processing result performed based on the user's instruction, and also outputs audio data for outputting the processing result as audio. 32.

The storage unit 50 is configured by a non-volatile memory such as a flash memory or an HD (Hard Disk), and stores data necessary for realizing various functions of the information search apparatus 1 in a readable / writable manner. As shown in FIG. 2, the storage unit 50 stores a character database (hereinafter abbreviated as “DB”) 60, a character derivation pattern 62, and an auxiliary keyword list 64.

  FIG. 3 shows an example of the data configuration of the character DB 60. The character DB 60 is a data table that stores basic character data 70 and derived character data 80 for each of a plurality of characters in a searchable manner.

  Basic character data 70 includes weather character data 70a, traffic character data 70b, meal character data 70c,... Corresponding to basic characters such as weather character CR1, traffic character CR2, meal character CR3,. It is set in advance at the initial stage.

In addition, as the derivative character data 80, go out character CR4, leisure (amusement park) character data, leisure (movie) character data, character newly created from the basic character, such as ... (hereinafter referred to as "derived character".) out character data 80a corresponding to leisure amusement park character data 80b, leisure (movie) character data 80c, there is ... etc., the basic character data 70 and the derived character data 80 in response to the user's search operation These two character data are synthesized and generated.

  FIG. 4A shows a data configuration example of the basic character data 70, and FIG. 4B shows a data configuration example of the derived character data 80. As shown in FIG. 4, each of the basic character data 70 and the derived character data 80 includes category information 71, a fixed keyword 73, a user keyword 75, and a character coefficient 77. , And a sub-keyword 79.

  Although not shown, the basic character data 70 and the derived character data 80 are image data for displaying a still image or a moving image of the character as shown in FIG. 1 and text data generated by the control unit 10. Is stored for each character as voice data obtained by digitally converting the voice to be read out by a recording editing method or a rule synthesis method.

  The category information 71 is data indicating a field, classification, and topic category that is a search range for the character to search for information. The fixed keyword 73 is a keyword related to the search in the category indicated by the category information 71. . The category information 71 and the fixed keyword 73 of the basic character data 70 are preset in the initial state.

  For example, “weather” is set as the category information 71 of the weather character data 70a, and keywords related to the weather such as “typhoon”, “precipitation probability”, “temperature”, and “alarm” are set in advance as the fixed keyword 73. The When the user selects a character by operating the operation unit 24 and performs a search, the user selects and searches for a desired keyword from among the fixed keywords 73 of the selected character (hereinafter referred to as “selected character”). Can be executed.

  In addition to this fixed keyword 73, keywords that the user inputs independently are stored as user keywords 75 in an accumulative manner. The user keyword 75 of each of the basic character data 70 and the derived character data 80 is an input history of keywords input by the user for each character. The user refers to and selects the user keyword 75 as the history.

  Note that selection of characters and input and selection of the user keyword 75 may be performed by voice input from the voice input unit 22. Specifically, the control unit 10 can perform voice recognition on the voice data output from the voice input unit 22 and determine which character has been designated based on the recognition result. Sometimes the recognition result is set as a keyword.

  The character coefficient 77 is a parameter indicating the frequency of search execution using the character, and is added every time the character is selected and search is executed. Based on the character coefficient 77 and the number of keywords stored as the user keyword 75, the control unit 10 adds a predetermined number (for example, three) of characters frequently used for searching by the user in the main character display area AR1. To display.

  Derived character data 80 creates a new character in accordance with the character derivation pattern 62 shown in FIG. 5 when the number of times that information retrieval using two characters has been performed simultaneously exceeds a predetermined number of times. Derived character data 80 for each character is generated. The character from which the derived character data 80 is created is referred to as “parent character”, and the basic character data 70 and the derived character data 80 of the character are referred to as “parent character data”.

  The character derivation pattern 62 is a data table that stores a combination pattern of a parent character that is a creation source for creating a new character, a derived character, and the number of searches. The number of searches is the number of times that the search is performed simultaneously with the combination pattern of the parent characters, and is set / updated by the control unit 10.

  The “simultaneous period” here means the same search timing. Specifically, after selecting one character and searching, the second character is continuously selected and searched. The timing when a plurality of characters are continuously selected and searched.

  For example, when the information search by the weather character CR1 and the information search by the traffic character CR2 are performed at the same time, the control unit 10 performs the search count corresponding to the combination pattern of the parent characters of “weather” and “transport”. Is added. Then, when the number of searches is 5 or more, for example, the derived character corresponding to the pattern, that is, the outing character CR4 is created and the outing character data 91 is generated.

  Here, a method of generating the derived character data 80 will be described with reference to FIG. It should be noted that since the character from which the derived character is created, that is, the parent character, is assumed to be both a basic character and a derived character, the constituent elements of the parent character data are the category information 71, the fixed keyword 73, the user keyword 75, and the character. This will be described as a coefficient 77.

  First, the category information 71 of the derived character data 80 is set by combining the respective category information of the first parent character data 91 and the second parent character data 92 of the two parent characters. In addition, an overlapping keyword is extracted from the user keyword 75 of the first parent character data 91 and the user keyword 75 of the second parent character data 92, and the keyword is set as the fixed keyword 73 of the derived character data 80.

Further, a user keyword 75 of the first parent character data 91, of the user keyword 75 of the second parent character data 92, extracts a different keyword without duplication, it sets the user keyword 75 derived character data 80 . Further, the larger one of the character coefficient 77 of the first parent character data 91 and the character coefficient 77 of the second parent character data 92 is set as the character coefficient 77 of the derived character data 80. In addition, as the character coefficient 77 of the derived character, an average value of the character coefficient 77 of the parent character may be set, or a total value may be set.

  For example, when the search of the weather character CR1 and the traffic character CR2 at the same time exceeds a predetermined number, first, the outing character data 91 of the outing character CR4 is generated as shown in FIG. Then, the category information 71 “weather” and “traffic” of the weather character data 70 a and the traffic character data 70 b are combined and set in the category information 71 of the outing character data 91.

  Next, overlapping keywords “Maihama” and “Chigasaki” are extracted from the user keywords 75 of the weather character data 70 a and the traffic character data 70 b, and set as the fixed keywords 73 of the outing character data 91. Further, “Hachioji” and “Laundry” of the user keyword 75 of the weather character data 70a and “Toilet” and “Fee” of the user keyword 75 of the traffic character data 70b are different keywords that do not overlap. Extracted and set as the user keyword 75 of the outing character data 91.

  As a result, the outing character CR4 is a character that takes over the categories of “weather” and “traffic” from the parent character and searches for information in this category. In addition, since the keywords “Maihama” and “Chigasaki” that the user has searched for at the same time with the weather character CR1 and the traffic character CR2 are set as the fixed keyword 73, the search is performed with the outing character CR4. The search can be executed by an easy operation such as selecting from preset fixed keywords 73 without inputting the keywords to be entered one by one.

  Further, since the user keyword 75 input by the weather character CR1 and the traffic character CR2 is also inherited and set as the user keyword 75 of the outing character CR4, it is possible to save time and effort for inputting the keyword input when searching for the parent character. The name of the derived character may be changeable by the user, so that the user can individually set an easy-to-understand name according to his / her preference.

  On the other hand, in FIG. 4B, the sub-keyword 79 included in the derived character data 80 is a keyword that is not owned by the own device acquired from another device. The information retrieval apparatus 1 acquires a different keyword and acquires a sub-keyword if a character held by another device connected to the communication network N has a fixed keyword and a user keyword different from the character of the own device. Store as 79.

  For example, as shown in FIG. 8, the fixed keyword 73 and the user keyword 75 of the outing character CR4 of the information search device 1 are compared with the fixed keyword and the user keyword of the other device 1a, and are keywords that the own device does not have. “Hotel”, “Amusement park” and “Train” are acquired and set as the sub-keyword 79.

  The auxiliary keyword list 64 is a keyword list that the character (information retrieval apparatus 1) proposes to the user in order to assist the user in inputting the keyword. The keyword (user keyword 75) that the user has input in the past or the same character is used. A keyword (sub-keyword 79) input by the user of the other machine in the information search is set.

  For example, when the selected character is a basic character, the user keyword 75 of the basic character data 70 is read and stored as the auxiliary keyword list 64 so that the keyword input by the user in the past need not be referred to from the history. Presented to the user by voice output and display output.

  Further, when the selected character is a derived character, a sub-keyword 79 is further added as an auxiliary keyword list 64 in order to propose a keyword that the user has not assumed in the keyword input using the derived character, that is, a newly created character. It is memorized and presented as audio output and display output.

  For example, when the outing character CR4 is selected as shown in FIG. 8, the keywords different from those of the other machine 1a are “hotel”, “amusement park”, and “train”, and these are acquired as the sub-keyword 79. Then, the user of the other device 1a is presented to the user of the own device that the user is searching for the keywords “hotel”, “amusement park”, and “train”, and the user's keyword input hint is given. Therefore, the user can know a keyword that he could not have imagined from the sub-keyword 79.

  In this way, a character having a function of assisting the user in inputting the keyword is realized by presenting / proposing the keyword input history or the keyword input by the user of the other device to the user as the auxiliary keyword list 64.

  In addition, when the user registers a category of interest in the information search apparatus 1 in advance, the control unit 10 may have a character that performs a division search using the registered category, The character data of the character is acquired and stored in the character DB 60 of the own device.

  For example, assume that the other device 1a has a health character CR5a as shown in FIG. When “health” is registered as a category of interest to the user, health character data 95a including “health” in the category information 71 is acquired from the other device 1a, and the basic character data 70 of the own device is acquired. Remember as. If the acquired character data category information 71 is composed of a plurality of categories, it is stored as derived character data 80.

  When the user selects the acquired health character CR5a and executes a search, an information search regarding “health” is performed by the health character CR5a. In this way, by registering a category in which the user is interested, a character who is good at information search in that category is brought from another machine, and a search for a category that the character is good at is performed on the own machine side. Such a function can be provided.

[Specific operation of information retrieval device]
Next, specific operations of the information search apparatus 1 will be described with reference to the flowcharts of FIGS. 9 to 11 and display screen examples of FIGS.

  First, the control unit 10 displays an initial screen 340 (see FIG. 12A) on the display unit 34 when the information retrieval apparatus 1 is turned on, and starts a retrieval process according to a program stored in the ROM. Wait until a search instruction for information search is given (step A1). For example, the control unit 10 determines that the touch panel on the information search button BT2 on the initial screen 340 is in contact with the position signal from the operation unit 24, or based on the audio data from the audio input unit 22. If it is determined that the voice input “JOHOKEN Saku” is input, it is determined that a search instruction for information search has been issued.

  Then, a character selection screen 342 shown in FIG. 12B is displayed and, for example, a voice output “Please select a dedicated agent.” Is performed to prompt the user to select a character. When one character is selected from a plurality of characters in the character selection screen 342 (step A3), a keyword input screen 344 as shown in FIG. 12C is displayed on the display unit 34, and fixed keyword selection and , Prompting for the input of user keywords. At this time, the fixed keyword 73 of the character data of the selected character is read out and displayed so as to be selectable by the check box 346.

  When the user uses a fixed keyword (step A5; Yes), the check keyword 346 is turned ON to acquire the selected fixed keyword (step A11), and then the user keyword input in the input area 348 Is acquired (step A13).

  In FIG. 12C, when the keyword list display button 350 is designated by touching the touch panel or the like, the keyword stored as the user keyword 75 is displayed by the list list 352 as shown in FIG. . The user can select a desired keyword from the list 352.

  Then, the control unit 10 performs an information search within the category of the category information 71 of the selected character based on the fixed keyword selected and input by the user and the user keyword (step A15), and the search result is displayed on the display unit 34. Display and output from the audio output unit 32 (step A17).

  When the user does not use the fixed keyword 73 (step A5; No), the input user keyword is acquired, and information search is performed within the category of the selected character based on the user keyword (step A9). The search result is notified from the display unit 34 and the audio output unit 32 (step A17). The control unit 10 stores the keyword input in steps A7 and A13 as the user keyword 75 of the selected character (step A19), and adds 1 to the character coefficient 77 of the selected character (step A21).

FIG. 13B shows an example of the search result display screen 354 when the traffic character CR2 is selected, the keyword “Chigasaki” is input, and the search is executed. On this search result display screen 354, a search result 356 of information related to “Chigasaki” within the category “transportation” is displayed. Further, along with this display, voice data for outputting the search result as voice is read from the character DB 60 and output to the voice output unit 32, so that the traffic character CR2 is output as if the search result 356 is read out.

  The control part 10 starts the auxiliary keyword proposal process shown in FIG. 9 after the process of step A21, creates the auxiliary keyword list 64, and presents the list to the user. Here, the auxiliary keyword proposal process will be described as follows.

  First, the control unit 10 accesses the character of the other device that is communicatively connected to the communication network N via the communication unit 40 (step B1), and sets the character data character stored in the other device. Then, it is determined whether or not there is the same character as the character held by the own device (step B3). If the same character is not held in another machine (step B3; No), it is determined whether or not a character that matches the user's interest is held in another machine (step B9).

  When it is determined that the character that matches the user's interest is held (step B9; Yes), the control unit 10 acquires the character data of the character from the other device via the communication network N, and Is stored in the character DB 60 (step B11). Thus, by acquiring the character data held by another device, a new character that performs information search on behalf of the user on a new category that the information search device 1 has not been set from the initial state becomes an information search device. 1 is provided.

  When the control unit 10 determines in step B3 that the derived character possessed by the other device and the derived character possessed by the own device are the same (step B3; Yes), the derived character data 80 of each other. The fixed keyword 73 and the user keyword 75 are compared to determine whether they are the same (step B5). If it is determined that they are not the same (step B5; No), the sub-keyword 79 is acquired from the fixed keyword 73 and the user keyword 75 of the own device and stored for each character data (step B7).

  When the control unit 10 determines that the keywords owned by the own device and the other device are the same after the processing of Steps B7 and B11 or at Step B5 (Step B5; Yes), the current information search is performed with the basic character. If it is a basic character (step B13; Yes), the user keyword 75 is stored in the auxiliary keyword list 64 (step B15).

  Then, the auxiliary keyword list 64 is displayed on the display unit 34 as, for example, the keyword candidates 362 in FIG. 13B, and the keywords input by the user in the past by the message 358 are presented (step B17). When a desired keyword is found in the keyword candidates 362, the user turns on the check box 364 and instructs the re-search button 360 to further narrow down the information search as shown in FIG. The search result 356 can be obtained. In this way, the user can easily search for information by selecting a suggestion from a character as a hint without referring to the history of keywords input in the past.

  If it is determined in step B13 that the current information search is performed on the derived character (step B13; No), the user keyword 75 and the sub-keyword 79 of the selected character are stored in the auxiliary keyword list 64. Further, the fixed keyword 73 of the parent character of the selected character is additionally stored in the auxiliary keyword list 64 (step B21).

  If the user keyword 75 of the parent character has been updated (step B23; Yes), the updated portion of the user keyword 75 is additionally stored in the auxiliary keyword list 64 (step B25), and the list is stored in step B17. Present to the user.

  The reason why the process is performed according to whether or not the user keyword 75 of the parent character is updated is as follows. That is, since the fixed keyword 73 of the derived character is set based on the user keyword 75 of the parent character as described above, the user keyword at the time of creation of the derived character is reflected in the fixed keyword 73. Since the updated contents of the parent character are not reflected after the creation of the derived character, the updated part of the user keyword of the parent character is acquired in step B25 and presented as the user auxiliary keyword list 64.

  For example, when the outing character CR4 is selected on the character selection screen 342 displayed as shown in FIG. 14A, the keyword input screen 344 shown in FIG. When the user inputs “amusement park” in the input area 348 of the screen and turns on the check box 346 of “Maihama” from the fixed keywords to instruct execution of the search, the category information 71 of the outing character CR4 Within the categories of “weather” and “traffic”, an information search using “amusement park” and “Maihama” as keywords is performed, and a search result 356 as shown in FIG. 14C is obtained.

  Thus, when going out, a user who frequently searches for information on the weather and traffic of the destination does not need to repeatedly select and select the weather character CR1 and the traffic character CR2 one by one. The search operation is facilitated by the character according to the preference.

  Further, when the outing character data 91 of the outing character CR4 has the data structure as shown in FIG. 8, the user keyword and sub-keyword of the outing character data 91 in FIG. 8, the weather character data 70a and the traffic character data 70b shown in FIG. Each fixed keyword is stored as an auxiliary keyword list 64. Then, among the keywords in the auxiliary keyword list 64, for example, a keyword frequently used by the user is selected and displayed as the keyword candidate 362 on the display unit 34, and the message 364 is displayed and output.

  For this reason, for example, the user can input a keyword using the keyword candidate 362 proposed by the outing character CR4 as a hint. The auxiliary keyword list 64 is displayed in the order of frequency of use for each keyword, thereby making it easier to select frequently used keywords by the user. The display order may be in the order of the Japanese syllabary.

When the auxiliary keyword suggestion process is completed, the control unit 10 shifts the process to step A25 of the search process of FIG. 9, and based on the operation signal from the operation unit 24 and the voice data from the voice input unit 22, It is determined whether or not a character change instruction has been made (step A25). Then, for example, is input voice data of "Character strange Kou", the character changes when the instruction is determined to have been performed (step A25; Yes), the display unit a character selection screen 342 as shown in FIG. 12 (b) 34, and the user's character selection is accepted (step A27).

  If the selected character and the character selected at the previous search are used as parent characters and the combination of the parent characters is in the character derivation pattern 62, the search count corresponding to the combination is incremented by one ( Step A29). Then, the derived character creation process shown in FIG. 11 is performed (step A31). After the derived character is created, the process proceeds to step A5, and information retrieval is performed again using the fixed keyword 73 and the user keyword 75.

  Here, the derived character creation process will be described as follows. First, it is determined whether or not the number of searches added in step A29 in FIG. 9 is a predetermined number (for example, 5 times) or more (step C1). If the number of searches is less than the predetermined number (step C1; No), the derived character creation process is terminated as it is.

 On the other hand, for example, when the parent character of the character derivation pattern 62 shown in FIG. 5 is 5 times or more like the search number of the combination of “weather” and “traffic” (step C1; Yes), it corresponds to the combination. And a memory area for storing the derived character data 80 is secured in the storage unit 50 (step C5).

  The control unit 10 sets the category information 71 and the character coefficient 77 of the derived character data 80 as described above based on the category information 71 and the character coefficient 77 of the basic character data 70 or the derived character data 80 serving as the parent character. (Step C5). Then, the user keywords 75 of the parent characters are compared to determine the presence / absence of overlapping keywords (step C7).

  When it is determined that there are no overlapping keywords (step C7; No), the control unit 10 sets the user keyword 75 of the parent character as the user keyword 75 of the derived character (step C9). If it is determined that there is a duplicate keyword (step C7; Yes), the keyword is extracted and set as a derived character fixed keyword 73 (step C11). Set (step C13).

  After the processes of steps C9 and C13, the control unit 10 compares the character coefficient 77 of the character displayed in the main character display area AR1 with the character coefficient 77 of the created derived character. Specifically, a variable a representing the order of the displayed character size is provided, and “1” is set in the variable a (step C15). Then, it is determined whether or not the character coefficient 77 of the derived character is larger than the character coefficient 77 of the character that is displayed a-th largest in the main character display area AR1 (step C17).

  When it is determined that the character coefficient 77 of the derived character is smaller than the character coefficient 77 of the character that is displayed as the a-th largest (step C17; No), the control unit 10 increments the variable a (step C19). If the variable a is less than or equal to the main displayable number, the process proceeds to step C17 and is compared again.

  If the variable a exceeds the main displayable number (step C21; Yes), that is, if the character coefficient 77 of the derived character is smaller than the character coefficient 77 of the character displayed in the main character display area AR1, it is created. An image of the derived character is added and displayed in the character list display area AR3 (step C23).

  In Step C17, when it is determined that the character coefficient 77 of the derived character is larger than the character coefficient 77 of the character that is displayed as the a-th largest (Step C17; Yes), the control unit 10 displays that character in the a-th. The derived character is displayed in the main character display area AR1 with an image larger than the currently displayed character.

  For example, when the number of consecutive searches by the weather character CR1 and the traffic character CR2 is 5 or more, the control unit 10 creates an outing character CR4 with these characters as the parent character, and goes out as described above. Character data 91 is generated. As shown in FIG. 7, when the character coefficient 77 of the outing character data 91 is “10”, the traffic character among the weather character CR1, the traffic character CR2, and the meal character CR3 displayed in the main character display area AR1. Since it is larger than the character coefficient 77 of CR2 (a = 2), the outing character CR4 is displayed larger than the traffic character CR2.

  When the main character display area AR1 has three main displayable numbers, the meal character CR3 having the smallest character coefficient 77 is moved to the character list display area AR3 and displayed, and the character list display area is displayed. When the number of characters that can be displayed in AR3 is exceeded, the character having the smallest character coefficient 77 is deleted from the area.

  In this way, by changing the character size and display area according to the character coefficient 77, it is possible to distinguish between characters frequently used by the user at the time of search and characters that are not frequently used. In addition, a character frequently used by the user can be known at a glance, and for example, when a character is selected with a touch panel or the like, it becomes easy to select the character.

  As described above, according to the present embodiment, when a new character is created according to the character selected by the user and the character is selected during information search, information search within a plurality of categories inherited from the parent character is performed. Therefore, the user does not have to bother to select a plurality of characters and perform a search, so that the operability during the search can be improved. Therefore, it is possible to perform information retrieval according to the user's preference by multiplying a plurality of characters according to the user's preference at the time of search and using a new character inheriting the characteristics of the original character.

  In addition, by acquiring a character that is not possessed from the character possessed by the other device, or by acquiring a keyword that the character of the own device does not have from the same character of the other device and presenting it to the user, It is possible to perform a search with a character that is not possessed by the own device or to input a keyword using the presented keyword as a hint. Accordingly, it is possible to support keyword input by proposing a keyword that is not expected by the user, and it is possible to provide a new approach for obtaining a search result desired by the user.

  In addition, since the character is displayed in a size corresponding to the number of the character coefficient 77 and the user keyword 75, a character that is frequently used by the user can be displayed in a large size, and a user interface that matches the user's preference can be realized. The operability can be improved.

[Modification]
Note that this embodiment is an example to which the present invention is applied, and the scope of the application is not limited to the above-described content. For example, in the embodiment described above, a derived character is created from two basic characters. However, a new derived character may be created from a combination of a basic character and a derived character, or a new derived character may be created from two derived characters. It may be possible to create a derived character. It should be noted that the creation of the derived character in this case can be realized by the derived character creating process shown in FIG. 11, and it goes without saying that the same effects as those of the above-described embodiment can be obtained.

  Further, the memory area secured as an area for storing the basic character data 70 and the derived character data 80 may be changed according to the character coefficient 77. Specifically, character data with a large character coefficient 77 sets the maximum number of keywords that can be stored as the user keyword 75 (for example, 100), and character data with a small character coefficient 77 decreases the number of keywords. Set. As described above, a large number of keywords are accumulated as a history for characters that are frequently used by the user, and the number of keywords is limited for characters that are not frequently used, so that the memory area of the storage unit 50 can be used efficiently. it can.

  Further, when different characters are selected continuously and the number of times of information retrieval by the combination exceeds a predetermined number of times, the derived character is created. For example, the number of times of information retrieval by the same keyword is the predetermined number of times. It is good also as producing a derived character according to the combination of the character beyond. Specifically, a combination of characters having a keyword associated with the number of times of retrieval exceeding a predetermined number is stored in association with the number of times of retrieval for the keyword input when performing information retrieval with the selected character. Create a derived character based on.

  For example, it is assumed that the user performs information retrieval of “Maihama” five times with the traffic character CR2 and performs information retrieval of “Maihama” five times with the weather character CR1 on another occasion. At this time, since “5” is associated as the number of searches with “Maihama”, which is the user keyword of each character, the outing character CR4 is created based on this combination. For this reason, a new character can be created based on a keyword often used by a user for a search.

  Moreover, although it demonstrated as acquiring the user keyword 75 and the fixed keyword 73 from the other machine which has the same character, it is good also as acquiring the category information 71. FIG. For example, in FIG. 8, “route” different from the category information 71 of the own device is set in the category information 71 of the outing character data 91 a of the other device. The control unit 10 acquires the category information 71 different from the character of the own device from the category information 71 of the same character held by the other device, and sets it as subcategory information of the character data.

  Then, the subcategory information is presented as a subcategory when the user searches for information. When the subcategory is designated by the user, information search within the subcategory is performed. As a result, for example, when information retrieval is performed using the outing character CR4, it is possible to perform information retrieval regarding the “route” that the outing character CR4a of the other machine has as a category at the time of retrieval. For this reason, the difference with the character which other machines hold can be compensated, and a new category can be proposed to a user.

The figure which shows the example of a display of the agent character in an information search device. The block diagram which shows an example of a function structure of an information search device. The figure which shows an example of a data structure of a character database. (A) is a figure which shows an example of a data structure of basic character data, (b) is derived character data. The figure which shows an example of the data structure of a character derivation pattern. The figure for demonstrating the production | generation method of derived character data. The 1st figure for demonstrating the production | generation method of the derived character data which used the going-out character data as an example. The 2nd figure for demonstrating the production | generation method of the derived character data which used the going-out character data as an example. The flowchart for demonstrating a search process. The flowchart for demonstrating an auxiliary keyword list creation process. The flowchart for demonstrating a derived character creation process. The 1st figure which shows the example of a transition of the display screen of an information search device. The 2nd figure which shows the example of a transition of the display screen of an information search device. The 3rd figure which shows the example of a transition of the display screen of an information search device.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Information retrieval apparatus 10 Control part 20 Input part 22 Voice input part 24 Operation part 30 Output part 32 Voice output part 34 Display part 40 Communication part 50 Storage part 60 Character database 62 Character derivation pattern 64 Auxiliary keyword list 70 Basic character data 71 Category Information 73 Fixed keyword 75 User keyword 77 Character coefficient 79 Sub keyword 80 Derived character data 91 First parent character data 92 Second parent character data AR1 Main character display area AR3 Character list display area

Claims (11)

Category storage means for storing a plurality of preset categories;
Category selection means for selecting a category stored in the category storage means;
Keyword storage means for storing a plurality of keywords associated with the category selected by the category selection means ;
Keyword selection means for selecting a predetermined keyword from a plurality of keywords stored by the keyword storage means;
Information search means for searching for information in the first category based on a predetermined keyword selected by the keyword selection means for a first category that is one category selected by the category selection means;
After performing a search for information in the first category by the information search means, when the second category different from the first category is continuously selected by the category selection means, the first category The third category is merged with the second category to create a third category for performing an information search based on the keywords of both the first category and the second category, Category creating means for adding the category to the category storage means;
An information retrieval apparatus comprising: The information search apparatus according to claim 1, wherein the category creating unit creates the third category based on a combination of a plurality of categories selected by the category selecting unit. The category creating means creates the third category when a continuous information search between the first category and the second category is performed a predetermined number of times or more. Or the information search device of 2. Before SL category creating means, a pre-Symbol plurality of keywords associated with the each of the third category was created from the first category and the second category read from the keyword storage means, the plurality of the same keyword of the keyword in association with the third category according to any one of claims 1 to 3, characterized that you provided with keyword preparing means you added to the keyword storage means Information retrieval device. Before Symbol keyword storing means, a history storage means for storing Kiki keyword before used in information retrieval is our row by prior Symbol information searching means in association with the category,
History notification means for notifying the keyword stored in the history storage means when the keyword selection means selects the keyword;
Information retrieval apparatus according to any one of claims 1-4, characterized in Rukoto equipped with. Reads the keywords associated with the category selected by the category selection unit from the keyword storage unit, according to claim 1 to 5, characterized in that the previous, further comprising a notification unit configured to notify the user when selecting the crisis keyword The information search device according to any one of the above. The notification unit, when the third category is selected by the pre-Symbol selection means, the keyword associated with the said third category the first category and the second category of creation source The information retrieval apparatus according to claim 6, wherein the information retrieval apparatus is read out from the keyword storage unit and notified. Information retrieval apparatus according to any one of claims 1-7, characterized in that to obtain Bei Table Shimesuru display means in the display of a size corresponding to the selected frequency of the category by the category selection unit. 9. The keyword acquisition unit according to claim 1, further comprising: a keyword acquisition unit configured to acquire the keyword used when performing the information search within the category from another device via a communication line network. Information retrieval device. A category acquisition means is provided for acquiring a category different from the plurality of categories stored in the category storage means from a plurality of categories in which the other machine searches for information from the other machine via a communication network. The information search device according to any one of claims 1 to 9. Computer
Category storage means for storing a plurality of preset categories;
Category selection means for selecting a category stored in the category storage means;
Keyword storage means for storing a plurality of keywords associated with the category selected by the category selection means;
Keyword selection means for selecting a predetermined keyword from a plurality of keywords stored by the keyword storage means;
Information search means for searching for information in the first category based on a predetermined keyword selected by the keyword selection means for a first category that is one category selected by the category selection means;
After performing a search for information in the first category by the information search means, when the second category different from the first category is continuously selected by the category selection means, the first category The third category is merged with the second category to create a third category for performing an information search based on the keywords of both the first category and the second category, Category creation means for adding the category of the category to the category storage means ,
Program to function as.

Images