JP2013250608A - Store information search system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a store information search system for searching information which is concretely unspecifiable with any word or phrase or information corresponding to vague needs.SOLUTION: A store information search system includes: a store database 20 for storing store information; a feature database 30 for storing the degree of association and appearance frequency of each feature data calculated from the store information in association for each feature data extracted from the store information; and an information provision part 11 for providing a Web page for accepting store search to a user terminal 3, and for accepting a search request from the Web page. When a search word is input on the Web page, the information provision part 11 selects the feature data associated with the input search word by referring to the feature database 30, and displays the feature data as a keyword for searching the store so as to be selectable on the Web page of the user terminal 3.

Description

  The present invention relates to a store information search system that provides store information to a user terminal via a network.

  2. Description of the Related Art Conventionally, a search system that provides an information search service such as a restaurant to a user terminal such as a personal computer or a mobile phone via a communication network is known. This search system generally provides a Web site having a search window for inputting a word or phrase (or a logical combination thereof) to a user terminal, and the word input from the user terminal to the search window And phrases are accepted as search criteria. The search system searches for information corresponding to the received search condition and provides the searched information to the user terminal.

  Patent Document 1 discloses a store search technique using the above-described search system. The search system (information search display server) described in Patent Document 1 is connected to a user terminal via a communication network. In addition, the search system accepts “keyword” and “region designation” input from the user terminal as search conditions, searches for stores corresponding to the search conditions, and on the map of the area specified by the user terminal, Display data in which the location of the store is arranged is transmitted. Further, in the above-described search system, when “keyword (search word)” is input from the user terminal, it is similar to the search word and has a high input frequency from the history information of the search word input in the past. Those that provide a search suggestion function for presenting keywords have also been put into practical use.

JP 2002-82958 A

  By the way, in information search, if information that cannot be specifically specified by words or phrases or information corresponding to vague needs can be searched, it is very useful in various fields. For example, when searching for restaurants for eating out, there are vague and potential needs such as "I want to eat something (unknown) and delicious", "I want to go to something (unknown) and interesting" However, there are many cases where a specific keyword or phrase for identifying a restaurant cannot be conceived. In such a case, it would be convenient if a restaurant satisfying vague needs and potential needs could be searched.

  However, information retrieval by the above-described conventional retrieval system has a technical problem that information that cannot be specifically specified by words or phrases or information corresponding to vague needs cannot be retrieved. Specifically, in order to search for and find desired information using a conventional search system, it is necessary to know words and phrases representing the desired information when searching. That is, the search system of the prior art can search for information that can be specified by words or phrases, but cannot search for things that are unknown, unconscious, forgotten, and the like. Note that the above-described search suggestion function of the prior art extracts and presents keywords from history information of search words. However, the keywords for extracting “information corresponding to user needs” do not necessarily match the history information, and the keywords from the history information alone are information corresponding to the vague needs and potential needs of the users. I can't search.

  The present invention has been made in view of the above technical problem, and provides a store information search system capable of searching for information that cannot be specifically specified by words or phrases or information corresponding to vague needs. With the goal.

  The present invention for solving the above technical problem is a store information search system connected to a user terminal via a network and providing store information to the user terminal, the store database storing store information, For each feature data indicating the feature of the store extracted from the store information, a feature database in which the relevance and appearance frequency of each feature data calculated from the store information are stored in association with each other, and a Web page that accepts store searches in the user terminal And providing a search request from the Web page, search for store information satisfying the search request with reference to the store database, and providing the user terminal with the searched store information. And when the search word is input on the Web page, the information providing unit is input with reference to the feature database. Feature data related to a search word is selected, the selected feature data is transmitted to the user terminal, and the feature data is displayed as a keyword for searching for a store on the Web page so as to be selectable. And

  As described above, in the present invention, when a search word is input on the Web page of the user terminal, feature data related to the input search word is selected with reference to the feature database, and the search result is displayed on the Web page of the user terminal. The feature data is displayed so as to be selectable as a keyword for searching for a store. In other words, in the present invention, “search feature data extracted from store information” that is not presented in the prior art is presented as a search suggestion function. Therefore, when searching, the user sees the feature data of the store related to the search word that he / she entered, and as a result, his / her potential needs and direction can be recognized. In addition, since feature data is presented as a keyword for searching, the store information that the user did not imagine when searching may be reached. Moreover, since the feature data of the store is presented by the search suggestion function, the feature of the store is promoted to the user.

  The store information includes store PR information, event information performed by the store, and feature information that the store participates in. The information providing unit includes the search word, the selected feature data, and the selected feature data. Using the store database, the feature information related to the search word is extracted from the store database, the extracted feature information is transmitted to the user terminal, and the keywords for searching the store on the Web page are as follows: It is desirable to display the extracted feature information together with the feature data in a selectable manner.

  As described above, according to the present invention, not only a keyword for a word but also feature information related to a search word input by a user is presented as a keyword for searching for a store. Therefore, in the present invention, information that was not imagined when the user searches can be presented, and the field of view of the user when searching for a store can be expanded.

  Further, for each predetermined time, the store information stored in the store database is analyzed, feature data is extracted from the store information, and the relevance and appearance frequency of each feature data are calculated. It is desirable to provide an information update unit that stores the association degree and the appearance frequency in association with each other.

  The reason for this configuration is that store information that introduces stores has a characteristic that it is frequently changed according to the season and time. For example, if the store information is restaurant information, menus and services provided by the season are often changed (even in the same restaurant, if the store is winter, the feature data of the store may be “nabe” etc. In summer, the characteristic data may be “cold noodles”). Therefore, by comprising in this way, the characteristic data of a shop can be provided as what respond | corresponds to the characteristic of the present shop. As a result, a store search adapted to the current store situation can be performed.

  Further, the web page is configured so that the store area selection and the dish genre selection can be accepted together with the input of the search word, and the feature data includes all store information registered in the store database. The entire feature database generated from the feature data obtained by analysis and the store information registered in the store database are classified for each area, and generated from the feature data obtained by analyzing the store information for each area. Area-specific feature data, store information registered in the store database for each genre, and feature-specific data generated from feature data obtained by analyzing store information for each genre, and the store Store information registered in the database is classified for each area and each cooking genre, and for each area and each cooking genre If the information providing unit receives a search word in which the area selection and the dish genre selection are not performed, the feature data is generated from the feature data obtained by analyzing the store information. When the feature data is selected from the overall feature database and the area-selected search word is received, the feature data is selected from the corresponding area-specific feature data, and the dish genre-selected search word is received In this case, feature data is selected from the corresponding feature data by genre, and when the search word selected by the area and dish genre is received, feature data is selected from the corresponding feature data by area / genre. desirable.

  As described above, in the present invention, a feature database reflecting areas and cooking genres is constructed. Therefore, when the user selects an area or a cooking genre and inputs a search word, feature data reflecting the area characteristics and the cooking genre is presented as a search keyword. As a result, according to the present invention, it is possible to cause the user to perform store information search in which the characteristics of the area and the cooking genre are recognized.

  ADVANTAGE OF THE INVENTION According to this invention, the shop information search system which can search the information corresponding to the information which cannot be specified concretely with words, phrases, etc., or vague needs can be provided.

It is the block diagram which showed the structure of the shop information search system of embodiment of this invention. It is the schematic diagram which showed an example of the data structure of the whole characteristic database which comprises the characteristic database of embodiment of this invention. It is the schematic diagram which showed an example of the data structure of the feature database classified by area which comprises the feature database of embodiment of this invention. It is the schematic diagram which showed an example of the data structure of the database classified by genre which comprises the characteristic database of embodiment of this invention. It is the schematic diagram which showed an example of the data structure of the feature database classified by area and genre which comprises the feature database of embodiment of this invention. It is a flowchart for demonstrating the search suggestion process which the shop information search system of embodiment of this invention performs. It is a schematic diagram for demonstrating the process which extracts the information relevant to the search word which the shop information search system of this embodiment performs, (a) is a model for demonstrating the extraction process of the characteristic data relevant to a search word. It is a figure, (b) is a schematic diagram for demonstrating the extraction process of the special feature relevant to a search word. It is the schematic diagram which showed an example of the web page for shop information search which the shop information search system of this embodiment provides. It is the schematic diagram which showed an example of the web page shown by the search suggestion process by the shop information search system of this embodiment.

  Hereinafter, embodiments of the present invention will be described with reference to the drawings. First, the structure of the store information search system of this embodiment is demonstrated based on FIG. FIG. 1 is a block diagram showing the configuration of the store information search system of this embodiment.

  As shown in the figure, the store information retrieval system of this embodiment includes a server group 1 connected to a communication network NW such as the Internet, and a user terminal 3 connected to the communication network NW. The server group 1 can perform data communication with the user terminal 3 via the communication network NW, and searches for restaurant information in response to a request from the user terminal 3. The server group 1 includes one or a plurality of server devices.

  In addition, the user terminal 3 includes an information processing apparatus (a personal computer, a PDA, Information processing devices such as smartphones and mobile phones).

  Then, the user terminal 3 can access the server group 1 via the communication network NW, receive the web page constituting the website transmitted from the server group 1, and display the web page on the display unit. ing. Further, the user terminal 3 accepts an operation from the user via the input unit, selects various data displayed on the Web page, transmits the selected data to the server group 1, A search word (free word) can be input to the search window formed in (1) and transmitted to the server group 1.

  Moreover, the server group 1 provides a web site (screen information (see FIGS. 8 and 9)) for restaurant search to the user terminal 3 in response to a request from the user terminal 3, and the user terminal 3 Data input on information is accepted as a search condition. The server group 1 uses the store database 20 to search for restaurant information according to the received search conditions, and transmits the searched restaurant information to the user terminal 3. In addition, the store information search system of this embodiment has the characteristics in the server group 1, and the user terminal 3 is the same as that of a well-known technique. Therefore, in the following, the configuration of the server group 1 will be described in detail, and the description of other configurations will be simplified.

  The server group 1 includes a control unit 10, an information providing unit 11, an information updating unit 12, and a storage unit 13. The storage unit 13 stores a store database 20, a feature database 30, Web page data 40, and search word history data 50.

  The store database 20 includes, for each restaurant, contents introducing the restaurant (information indicating the store's characteristics such as a PR statement of the store, event information performed by the restaurant, special features of the restaurant (feature information), This is a database in which store information that associates a restaurant business type, a restaurant area), a restaurant address and telephone number, a menu provided by the restaurant, a homepage address, and the like is registered. Store information registered in the store database 20 includes store detail information (store word-of-mouth information, etc.) where the details of the store are shown in detail, and store guide information created by extracting certain items from the store information. include. In addition, this store information is comprised so that it can change suitably according to the request | requirement from a restaurant or the like.

  In addition, the feature database 30 analyzes the store information (for example, automatic analysis by a computer such as text mining), and the feature data indicating the feature of the store based on the store information (words to be a store uri (Uri word)) Are extracted, and each feature data is registered in association with the degree of association and the appearance frequency (frequency) of each feature data calculated by the analysis. The feature data extracted by the analysis includes a keyword indicating the contents that can be experienced at the restaurant, a keyword indicating the merit when using the restaurant, a specific feature of the dish, and the like. For example, if the PR statement for Store A was “Italian famous Ginza store and wine is delicious hot stone oven pizza”, and the PR statement for Store B was “hot silver fluffy oyakodon”, “stone kiln pizza” “Atsatsu”, “Stone kiln pizza”, “Famous store”, etc. are extracted as feature data. In this embodiment, for example, when the difference between the degree of association of “stone kiln pizza” and the degree of association of “hot” is small, it is recognized as feature data in which both are associated.

  The feature data 30 of the present embodiment includes an overall feature database 30A, an area-specific feature database 30B, a genre-specific feature database 30C, and an area-genre-specific feature database 30D. Here, the overall feature database 30A is a database generated from feature data obtained by analyzing all store information registered in the store database 20. The area-specific feature database 30B is a database generated from feature data obtained by classifying store information registered in the store database 20 for each area and analyzing the store information for each area. The genre-specific feature database 30 </ b> C is a database generated from feature data obtained by classifying store information registered in the store database 20 for each cooking genre and analyzing store information for each cooking genre. The feature database 30D classified by area / genre is a feature obtained by classifying store information registered in the store database 20 for each area and each cooking genre, and analyzing the store information for each area and each cooking genre. A database generated from data.

  The web page data 40 is a database in which various data such as images and icons necessary for generating a web page to be transmitted to the user terminal 3 are registered. The search word history data 50 is a database in which search words input (transmitted) from the user terminal 3 are registered as history information.

  In addition, the hardware configuration of the server device configuring the server group 1 is not particularly limited. For example, the server group 1 includes a CPU, an auxiliary storage device, a main storage device, a network interface, and an input / output interface. (One or a plurality of computers). In this case, the auxiliary storage device stores a program for realizing the functions of the control unit 10, the information providing unit 11, and the information updating unit 12. In addition, the storage unit 13 described above is formed in a predetermined area of the auxiliary storage device. The functions of the control unit 10, the information providing unit 11, and the information updating unit 12 are realized when the CPU loads the program stored in the auxiliary storage device to the main storage device and executes it. Hereinafter, functions of the server group 1 will be described in detail.

  Specifically, the control unit 10 controls the operation of the entire server group 1 and accepts various settings of the server group 1.

  The information providing unit 11 configures a restaurant search Web site (restaurant search site) provided to the user terminal 3 using the store database 20, the feature database 30, and the Web page data 40 stored in the storage unit 13. Web page (screen information) to be generated is generated. Further, the information providing unit 11 transmits the generated web page to the user terminal 3 and causes the display unit of the user terminal 3 to display the web page 100 (see FIG. 8). The Web page 100 is provided with a search window (input field) 101 that accepts input of a search word (free word). When the information providing unit 11 receives the search word input in the search window 101, the information providing unit 11 refers to the store data 20 and the feature database 30, and “feature data” and “special feature of store” related to the input search word. Information "is extracted. In addition, the information providing unit 11 uses the extracted “feature data (see reference numeral 105b in FIG. 9)” and “special information on the store (reference numeral in FIG. 9) as keywords for searching the store on the display unit of the user terminal 3. 105c) ”is displayed in a selectable manner.

  In addition, the information updating unit 12 extracts a plurality of feature data indicating the features of the store from the store information registered in the store database 20. As an extraction method, automatic analysis, particularly text mining is preferable. When extracting the feature data, the relevance level and appearance frequency (frequency) of the feature data are calculated. Then, the information update unit 12 associates the calculated frequency and the degree of association with each feature data and stores them in the feature database 30.

  Specifically, the information update unit 12 generates an overall feature database 30A (see FIG. 2) for all store information through automatic analysis for all store information registered in the store database 20. . Further, the information updating unit 12 classifies the store information registered in the store database 20 for each predetermined area (area) (the areas include other nations such as Kanto, Tokyo, Ginza, Kansai, Osaka, Kyoto, etc.). After that, automatic analysis is performed for each classified store information, and an area-specific feature database 30B (see FIG. 3) for each region is generated. Further, the information updating unit 12 classifies the store information registered in the store database 20 for each predetermined cooking genre (the cooking genre includes non-genre in addition to Japanese food, Italian, etc.) and then stores the classified store. Automatic analysis is performed for each piece of information to generate a genre-specific feature database 30C for each cooking genre. The information updating unit 12 classifies the store information registered in the store database 20 for each predetermined region and each predetermined dish genre, performs automatic analysis for each classified store information, and performs each region and each dish genre. The feature database 30D for each area and genre is generated. In addition, the information updating unit 12 automatically analyzes the store information and generates the feature database 30 every predetermined time (for example, every 24 hours), and updates the registered content of the feature database 30. Yes.

  The automatic analysis such as text mining performed by the information updating unit 12 is realized by a well-known technique, and thus detailed description thereof is omitted. In the present embodiment, feature data indicating the features of the store is extracted from the store information registered in the store database 20 by automatic analysis. However, the present invention is not limited to this. For example, in addition to the store information, the data to be subject to automatic analysis may include information on advertisements of the restaurant and information posted on a homepage (Web site).

  Next, the data structure of the feature database 30 will be described with reference to FIGS. FIG. 2 is a schematic diagram showing an example of the data structure of the entire feature database constituting the feature database of the present embodiment. FIG. 3 is a schematic diagram showing an example of the data structure of the area-specific feature database constituting the feature database of the present embodiment. FIG. 4 is a schematic diagram showing an example of the data structure of a genre-specific database constituting the feature database of the present embodiment. FIG. 5 is a schematic diagram showing an example of the data structure of the feature database classified by area / genre constituting the feature database of the present embodiment.

  The overall feature database 30A shown in FIG. 2 is a database generated from feature data extracted by automatic analysis for all store information registered in the store database 20, and for each of the extracted feature data, the feature data Are stored in association with the frequency and degree of association. Specifically, the entire feature database 30A registers a field 31 for registering feature data extracted by automatic analysis, a field 32 for registering frequency calculated by automatic analysis, and a degree of relevance calculated by automatic analysis. And a field 33 to form one record, and a plurality of records are registered. The entire feature database 30A is used for search suggestion processing when a “search word (free word)” that does not specify an area and a cooking genre is input.

  Further, the area-specific feature database 30B shown in FIG. 3 extracts the feature data by automatically analyzing the store information registered in the store database 20 for each area, and adds the feature data to each feature data extracted for each area. Are stored in association with the frequency and degree of association. Specifically, the area-specific feature database 30 </ b> B includes a plurality of databases 30 </ b> B <b> 1 to Bn (n is a natural number equal to or greater than “2”) generated for each area. Each database 30B1 to Bn is provided with area information for identifying an area. The records of the databases 30B1 to Bn each include a field 31 for registering feature data, a field 32 for registering frequency, and a field 33 for registering the degree of association, as in the case of the entire feature database 30A. Thus, by analyzing store information for each area, a database reflecting the characteristics of the area (region) can be constructed. This area-specific feature database 30B is used for search suggestion processing when a “search word (free word)” designating an area is input.

  The feature database 30C by genre shown in FIG. 4 automatically analyzes store information registered in the store database 20 for each dish genre, and extracts the frequency and relevance of the feature data for each feature data extracted for each dish genre. It is stored in association with each other. Specifically, the genre-specific feature database 30 </ b> C includes a plurality of databases 30 </ b> C <b> 1 to Cn (n is a natural number equal to or greater than “2”) generated for each cooking genre. Each database 30C1 to Cn is provided with genre information for identifying a cooking genre. The records of the databases 30C1 to Cn each include a field 31 for registering feature data, a field 32 for registering a frequency, and a field 33 for registering a degree of association, like the entire feature database 30A. Thus, by analyzing store information for each cooking genre, a database reflecting the characteristics of the cooking genre can be constructed. This genre-specific feature database 30C is used for search suggestion processing when a “search word (free word)” specifying a cooking genre is input.

  The feature database 30D by area / genre shown in FIG. 5 automatically analyzes store information registered in the store database 20 for each area and each dish genre, and extracts it from the store information for each area and each dish genre. Each feature data is stored in association with the frequency and relevance of the feature data. Specifically, the area-genre-specific feature data 30D includes a plurality of databases 30D1 to Dn (n is a natural number equal to or greater than “2”) generated for each area and for each cooking genre. Each database 30D1 to Dn is provided with area / genre information for identifying the area and the cooking genre. The records of the databases 30D1 to Dn each include a field 31 for registering feature data, a field 32 for registering frequency, and a field 33 for registering the degree of association, as in the case of the entire feature database 30A. As described above, by analyzing the store information for each area and for each cooking genre, a database reflecting the characteristics of the cooking genre in each area can be constructed. This area / genre-specific feature database 30D is used for search suggestion processing when a “search word (free word)” specifying “area and dish genre” is input.

  Next, search suggestion processing performed by the store information search system of this embodiment will be described with reference to FIGS. Here, FIG. 6 is a flowchart for explaining the search suggestion process performed by the store information search system of the present embodiment. FIG. 7 is a schematic diagram for explaining processing for extracting data related to the input search word, and FIG. 7A is a schematic diagram for explaining selection processing of feature data related to the search word. And (b) is a schematic diagram for explaining a selection process of a special feature related to a search word. FIG. 8 is a schematic diagram showing an example of a store information search Web page provided by the store information search system of the present embodiment. FIG. 9 is a schematic diagram showing an example of a Web page presented by the search suggestion process by the store information search system of the present embodiment.

  First, the user terminal 3 accesses the server group 1 via the communication network NW, and requests the server group 1 to present a restaurant search site. When receiving the presentation request, the information providing unit 11 of the server group 1 generates a Web page that configures the restaurant search site, and transmits the generated Web page to the user terminal 3 (S1).

  Specifically, when receiving the presentation request, the information providing unit 11 uses the store database 20, the feature database 30, and the Web page data 40 to provide a restaurant search Web site (restaurant search) to be provided to the user terminal 3. A web page 100 (see FIG. 8) constituting the site is generated and transmitted to the user terminal 3. As a result, the web page 100 is displayed on the display unit of the user terminal 3.

  The Web page 100 includes a search window (input field) 101 that accepts an input of a search free word (search word), an area designation field 102 that accepts an area designation of a restaurant, and a designation of a restaurant genre. A genre designation field 103 for accepting and a search button 104 for accepting search processing are provided. In response to a request from the user terminal 3, the information providing unit 11 searches for restaurant information by “free word”, searches for restaurant information by “area designation”, and stores restaurant information by “designation genre designation”. The restaurant information is searched by searching and “combination of free word, area, and cooking genre”, and the searched restaurant information can be provided to the user terminal 3.

  Next, the information provision part 11 receives the input of the search word from the user terminal 3 (S2), and progresses to the process after S3. Specifically, when the user operates the user terminal 3 and inputs a search word into the search window 101 of the Web page 100 displayed on the display unit, the search word is transmitted to the information providing unit 11.

  In this step, when a search word is input to the search window 101, if an area is specified in the area specification field 102, the specified area and the search word are transmitted to the information providing unit 11. (In this case, the information providing unit 11 receives an input of a search word designating an area). Further, when the search word is input, if a food genre is specified in the genre specification field 102, the specified food genre and the search word are transmitted to the information providing unit 11 (in this case). The information providing unit 11 receives an input of a search word specifying a cooking genre). In addition, when the search word is input, if an area is specified in the area specification field 101 and a food genre is specified in the genre specification field 102, the specified area and food are displayed in the information providing unit 11. The genre and the search word are transmitted (in this case, the information providing unit 11 accepts the input of the search word specifying the area and the cooking genre). In the following description, a case where “Shinjuku” is input in the area designation field 102 and “Yakiniku” is input as a search word in the search window 101 will be described as an example.

  Next, the process of S3, S4, S5 performed after S2 is demonstrated. The order of processing of “S3” and “S4 and S5” is not particularly limited. For example, the processing of “S3” may be performed after “S4, S5”, or the processing of “S3” and “S4 and S5” may be performed in parallel.

  Specifically, in S3, the information providing unit 11 searches the search word history data 50 stored in the storage unit 13 for a word similar to the search word accepted in S2 (hereinafter referred to as “similar word”). To extract. Note that the information providing unit 11 extracts words having a large number of inputs when extracting similar words. In this case, from the search word history data 50, among similar words similar to “Yakiniku”, words (such as a yakiniku private room, all-you-can-eat all-you-can-eat) are extracted. A known technique (for example, a similarity determination technique using a vector space model) is used to extract a word similar to the search word, and thus detailed description thereof is omitted. In this step, the information providing unit 11 registers the search word received in S2 in the search word history data 50.

  In S4, the information providing unit 11 uses the feature database 30 stored in the storage unit 13 to extract feature data (uriword) related to the search word received in S2.

  Specifically, in S4, the information providing unit 11 selects a corresponding database from the databases 30A, 30B, 30C, and 30D constituting the feature database 30 according to the data received in S2. Here, an example is shown in which “Shinjuku as an area” is designated and “Yakiniku as a search word” is designated, so the area of the feature database 30B by area selects the database 30B1 in Shinjuku (see FIG. 3). To do.

  Note that the information providing unit 11 selects the entire feature database 30A when receiving a search word in which no area or genre is specified in S2. In S2, when the information providing unit 11 receives a search word in which a genre is specified, the information providing unit 11 selects a corresponding genre database (any of databases 30C1 to Cn) in the genre-specific feature database 30C. Further, for example, when the information providing unit 11 receives a search word in which an area and a genre are specified in S2, the information providing unit 11 corresponds to the area / genre characteristic database 30D (any one of the databases 30D1 to Dn). ) Is selected.

  Then, the information providing unit 11 refers to the selected database 30B1 (feature database 30), reads the relevance level of the search word “Yakiniku” accepted in S2, and uses the read relevance level to read “Yakiniku”. Extract feature data related to. For example, the information providing unit 11 extracts feature data having a relevance degree within a predetermined range based on the relevance degree of “Yakiniku” (extracts a plurality of feature data). Moreover, the information provision part 11 calculates a score (score) about each of the extracted some feature data. Although the score calculation method is not particularly limited, for example, the information providing unit 11 calculates a score using the degree of association and the frequency of each feature data (for example, adding after weighting each degree of association and frequency) Etc. to calculate the score). Further, the information providing unit 11 selects a predetermined number of feature data in descending order of the score as the feature data to be presented to the user (for example, selects the top ten feature data of the score), Associate selected feature data. For example, as shown in FIG. 7A, the information providing unit 11 generates suggestion feature data 60a in which selected feature data having a higher score is associated with the received search word.

  In this step (S4), when the search word received in S2 is not found in the selected database 30B1 (feature database 30), for example, the information providing unit 11 searches from the database 30B1 in S2. Feature data similar to the word is selected, and feature data related to the selected feature data is selected. In this case, a known technique is used to extract a feature word similar to the search word.

  Next, the process of S5 will be described. In S5, the information providing unit 11 uses the feature data selected in S4, the feature database 30 and the store database 20 stored in the storage unit 13, and the “special feature of the store” related to the search word received in S2. Information "is extracted.

  For example, the information providing unit 11 calculates a score for each feature information registered in the store database 20 using the feature data extracted in S4 (feature data related to the search word). The score calculation method is not particularly limited. For example, the information providing unit 11 calculates the score of each feature information by using the score for each feature data calculated in S4 and the frequency of the feature data in each feature information. To do. Further, the information providing unit 11 selects a predetermined number of feature information in descending order of score as feature information to be proposed to the user (for example, selects the top ten feature information in the score), Associate selected feature information with higher scores. For example, as illustrated in FIG. 7B, the information providing unit 11 generates the suggestion special feature information 60 b in which the selected special feature information is associated with the received search word.

  Next, the information providing unit 11 sends to the user terminal 3 the similar word obtained by the process of S3, the feature data (uriword) obtained by the process of S4, and the special feature information obtained by the process of S5. Then, similar words, feature data, and special feature information are displayed on the display unit of the user terminal 3 so as to be selectable as keywords for searching for a store (S6). For example, as shown in FIG. 9, the information providing unit 11 displays the similar word, the feature data (uriword), and the feature in the lower area of the search window 101 of the Web page 100 displayed on the user terminal 3. A list box 105 presenting information to be selectable is displayed.

  In the example of FIG. 9, “similar words extracted from user input history (private room, all-you-can-eat, etc.)” are listed in the upper area 105 a of the list box 105. Also, “characteristic data related to the input search word (smokeless roaster, one person, Ishigaki beef, etc.)” is displayed in a list in the middle area 105 b of the list box 105. In the lower area 105c of the list box 105, “special information related to the input search word (“ special stamina! Special yakiniku / hormones ”,“ special hot Korean food ”, etc.) is displayed in a list.

  Then, when the user operates the user terminal 3 to select data displayed in the list box 105 (selects one of similar words, feature data, and feature information) and clicks the search button 104, the information providing unit 11 accepts the selected data as a search condition and executes a search process. Specifically, the information providing unit 11 refers to the store database 20, searches for store information that satisfies the received search condition, and transmits the searched store information to the user terminal 3.

  As described above, in this embodiment, when a user inputs a search word, the search suggestion function can select not only the similar word extracted from the input history held by the server group 1 but also store feature data and store feature information. It is possible to search for restaurants based on the feature data and feature information. In other words, according to this search suggestion function, even when the user is not able to materialize his / her needs or when an appropriate search word cannot be conceived to find a desired store, a simple search word is input. Only feature data and feature information for searching for a store are presented to be selectable. Therefore, according to this embodiment, the user can materialize his vague needs by referring to the feature data (restaurant of restaurant) and feature information displayed by the search suggestion function. .

  Further, in the search suggestion function of the present embodiment, it is information that is not presented in the general search suggestion function and is “Uri of the store where the restaurant side is promoting” and is related to the input search word. Information is presented. That is, according to the present embodiment, a word that cannot be thought of even if it is a user's potential needs may be presented as a search keyword, and a store that the user is potentially seeking can be searched. Can increase the sex.

  In addition, the feature database 30 according to the present embodiment includes an area-specific feature database 30B in which feature data obtained by automatically analyzing store information for each area and store information is automatically analyzed for each dish genre. A genre-specific feature database 30C in which the feature data is registered, and an area / dish-specific feature database 30D in which feature data obtained by automatically analyzing store information for each area and genre are included. . By configuring in this way, it is possible to construct a database in which feature data reflecting regional characteristics and cooking genre contents is registered. According to the present embodiment, the feature database 30 is used to extract “feature data” and “special feature information” to be presented to the user. Presentation is possible. In other words, according to the present embodiment, when a user inputs a search word after designating an area (or cooking genre), feature data or feature information indicating features unique to that region (or cooking) is displayed, and the user is prompted. On the other hand, it is possible to propose a store selection that reflects the regional characteristics (or cuisine genre).

  As described above, according to the present embodiment, it is possible to provide a store information search system that can search for information that cannot be specifically specified by words or phrases or information that corresponds to vague needs. In addition, this invention is not limited to embodiment mentioned above, A various deformation | transformation is possible within the range of the summary.

  For example, in the above-described embodiment, the case where the store information is a restaurant is taken as an example, but this is only an example. The present invention is also applied to store information other than restaurants.

DESCRIPTION OF SYMBOLS 1 ... Server group, 3 ... User terminal, 10 ... Control part, 11 ... Information provision part, 12 ... Information update part, 13 ... Memory | storage part, 20 ... Store database, 30 ... Feature database, 40 ... Web page data, 50 ... Search word history data

Claims (4)

A store information search system connected to a user terminal via a network and providing store information to the user terminal,
A store database storing store information;
For each feature data indicating the feature of the store extracted from the store information, a feature database storing the association degree and appearance frequency of each feature data calculated from the store information in association with each other,
When a web page for accepting a store search is provided to the user terminal and a search request is accepted from the web page, store information that satisfies the search request is searched with reference to the store database, and the user terminal searches for the store information. And an information provider that provides store information,
When a search word is input on the Web page, the information providing unit selects feature data related to the input search word with reference to the feature database, and selects the selected feature data in the user terminal. And the feature data is displayed on the Web page as a keyword for searching for the store so as to be selectable. The store information includes PR information of the store, event information that the store performs, and feature information that the store participates in,
The information providing unit extracts feature information related to the search word from the store database using the search word, the selected feature data, and the store database, and transmits the extracted feature information to the user terminal. The store information search system according to claim 1, wherein the extracted feature information is displayed together with the feature data as a keyword for searching for a store on the Web page. Analyzing the store information stored in the store database every predetermined time, extracting feature data from the store information, calculating the relevance and appearance frequency of each feature data, and for each feature data The store search system according to claim 1, further comprising an information update unit that stores the degree and the appearance frequency in association with each other. The web page is configured to be able to accept the store area selection and dish genre selection along with the search word input,
The feature data is
An overall feature database generated from feature data obtained by analyzing all store information registered in the store database;
By classifying the store information registered in the store database for each area, the area-specific feature data generated from the feature data obtained by analyzing the store information for each area,
The store information registered in the store database is classified for each cooking genre, the cooking-specific feature data generated from the feature data obtained by analyzing the store information for each cooking genre,
Area information and genre-specific feature data generated from feature data obtained by classifying store information registered in the store database for each area and for each food genre and analyzing store information for each area and for each food genre Prepared,
The information providing unit selects the feature data from the global feature database when receiving a search word in which the area selection and the cooking genre selection are not performed, and receives the search word selected in the area Selects feature data from the corresponding feature data for each area, and selects the feature data from the corresponding feature data for each genre when the search word selected for the food genre is received, and the area and the food genre are selected. The store information search system according to any one of claims 1 to 3, wherein, when a search word is received, feature data is selected from the corresponding feature data by area and genre.

Images