WO2011093358A1

WO2011093358A1 - Information retrieval device, information retrieval method, information retrieval program, and recording medium

Info

Publication number: WO2011093358A1
Application number: PCT/JP2011/051553
Authority: WO
Inventors: 佳代子髙橋
Original assignee: 楽天株式会社
Priority date: 2010-01-27
Filing date: 2011-01-27
Publication date: 2011-08-04
Also published as: JP2013077041A

Abstract

Information relating to a feeling of a user is retrieved. An information retrieval device for retrieving information in response to a request from a terminal device connected via a network is characterized by comprising an image display information transmission means for transmitting, to the terminal device, image display information for causing the terminal device to display a plurality of images which are stored in a keyword storage means for storing an image and a keyword relating to the image with the image and the keyword associated with each other, a selection image information reception means for receiving, from the terminal device which has received the image display information, selection image information which indicates an image selected by the user from the images displayed by the terminal device, and a retrieval means for retrieving information by using, as a retrieval keyword, the keyword associated with the image indicated by the selection image information received.

Description

Information search apparatus, information search method, information search program, and recording medium

The present invention relates to a technical field of an information search device, an information search method, and an information search program for searching for information in response to a request from a terminal device.

Conventionally, a technique for performing a fuzzy search based on a keyword input by a user is known. For example, in the technique described in Patent Document 1, synonym expansion of an input keyword is performed, a text search of an electronic document is executed, and then a nearby figure is searched with reference to the allocation information of the electronic document. Further, in this technique, the narrower term of the keyword is expanded according to the meaning of the keyword, and the lower term is searched.

Japanese Patent Laid-Open No. 7-85092

By the way, when a user wants something such as a product or a service, something that the user wants is not concrete in the mind, and the content of the desire may be vague and sensory (for example, “sweet” I want something ").

However, even if the technique described in Patent Document 1 can be applied to the search for information on things such as commercial transactions, it is not possible to perform a search corresponding to the vague psychological feeling of the user. This is because, for example, the technique described in Patent Document 1 merely performs a search by expanding the keyword “mouse” to a synonym or expanding it to a lower term.

Accordingly, the present invention has been made in view of the above points, and an object thereof is to provide an information search device, an information search method, an information search program, and a recording medium that can search for information related to a user's sense. And

In order to solve the above-described problem, the invention according to claim 1 is an information search device for searching for information in response to a request from a terminal device connected via a network, and relates to an image and the image. Image display information transmitting means for transmitting to the terminal device image display information for displaying a plurality of images stored in the keyword storage means for storing the keyword in association with each other, and the image display information received Corresponding to the selected image information receiving means for receiving selected image information indicating an image selected by the user from among the images displayed by the terminal device from the terminal device, and an image indicated by the received selected image information. And a search means for searching for information using the keyword as a search keyword.

According to the present invention, information is searched using a keyword related to an image selected by the user among a plurality of images displayed by the terminal device. Therefore, by selecting an image with the user's sense, it is possible to search for information related to the user's sense.

According to a second aspect of the present invention, in the information search device according to the first aspect, the keyword storage unit stores an image associated with the information to be searched, and the information to be searched or the information is Text information receiving means for receiving text information input by the user from the terminal device with respect to the thing to be shown; a keyword is extracted from the received text information, and the extracted keyword is input with the received text information Or a keyword extracting means for storing in the keyword storage means in association with an image associated with information indicating things.

According to this invention, a keyword extracted from text information input by a certain user for information to be searched is associated with an image corresponding to the information, and a plurality of images displayed by the terminal device, Information is searched using a keyword related to the image selected by the user using the terminal device. Therefore, by selecting an image with the user's sense, it is possible to search for information related to the user's sense.

According to a third aspect of the present invention, in the information search device according to the second aspect, for each keyword stored in the keyword storage means, a counting means for counting the number of times of extraction by the keyword extracting means, and the number of times of extraction. Determining means for determining the level of the relationship between the image stored in the keyword storage unit and the keyword, and the keyword associated with the image indicated by the received selected image information is related to the image. First selection means for selecting a keyword determined to be high, and the search means uses the keyword selected by the first selection means as a search keyword.

According to the present invention, based on how many times each keyword is extracted, the level of association between each keyword and an image associated with each keyword is determined. Then, keywords that are highly relevant to the image selected by the user are used for the search. Therefore, it is possible to search for information highly relevant to the user's sense.

According to a fourth aspect of the present invention, in the information search device according to the third aspect, the information processing device further comprises selection information receiving means for receiving selection information indicating selection by a related high / low user from the terminal device, and the first selection. The means is characterized in that, when the selection information indicates that a lower association is selected, a keyword determined to have a low association with the image is selected.

According to the present invention, information having a low relationship with the user's sense is searched by the user's selection. Therefore, it is possible to search for an object of an unexpected commercial transaction.

According to a fifth aspect of the present invention, in the information search apparatus according to any one of the first to fourth aspects, the search means is selected from keywords associated with an image indicated by the received selected image information. And a second selection unit that selects a keyword used by the search unit, wherein the search unit uses the keyword selected by the second selection unit as a search keyword, and the second selection unit selects at least one of the keywords to be selected. Is changed for each search by the search means.

According to this invention, since at least one of the keywords used for the search changes each time the search is performed, even if the user selects the same image as the previous time, the search result may change from the previous time. Therefore, it is possible to search for unexpected information.

According to a sixth aspect of the present invention, in the information search device according to any one of the first to fifth aspects, user information acquisition means for acquiring information on a user who has selected an image indicated by the received selected image information. Text information acquisition means for acquiring text information associated with the content of the acquired user information, and the acquired keyword from the keywords associated with the image indicated by the received selected image information. And a third selection means for selecting a keyword included in the text information, wherein the search means uses the keyword selected by the third selection means as a search keyword.

According to the present invention, search results personalized for the user can be provided to the user.

According to a seventh aspect of the present invention, in the information search device according to the sixth aspect, for each image indicated by the received selected image information, the third selection unit selects a keyword associated with the image. First generation means for generating first condition information indicating that all of the generated keywords are included in the information to be searched, and the received selection of the generated first condition information Second generation means for generating second condition information indicating a search condition used by the search means in combination between images indicated by image information, wherein the search means includes the generated second condition information It is characterized by searching information based on the above.

When there are a plurality of keywords related to an image selected by the user, there may be a plurality of keywords included in the text information among the plurality of keywords related to the image. According to the present invention, in this case, the first condition information indicating that all of the plurality of keywords included in the text information are included in the information to be searched is generated. When a plurality of images are selected by the user, the second condition information is generated by combining the first condition information generated for each image. Information is searched based on the generated second condition information. Therefore, a personalized search condition can be generated as a search condition in which a plurality of keywords are combined only by selecting an image by the user.

The invention according to claim 8 is characterized in that in the information search device according to any one of claims 1 to 7, the information to be searched is information for a commercial transaction.

According to the present invention, it is possible to search for information on a commercial transaction related to the user's feeling.

The invention according to claim 9 is an information search method for searching for information in response to a request from a terminal device connected via a network, and stores an image and a keyword related to the image in association with each other. An image display information transmitting step for transmitting to the terminal device image display information for displaying the plurality of images stored in the keyword storage means on the terminal device, and the terminal device from the terminal device receiving the image display information. The selected image information receiving step for receiving the selected image information indicating the image selected by the user among the images displayed by the above, and the keyword associated with the image indicated by the received selected image information as a search keyword And a search step for searching for information.

According to a tenth aspect of the present invention, a computer included in an information search device that searches for information in response to a request from a terminal device connected via a network is associated with an image and a keyword related to the image. Image display information transmitting means for transmitting to the terminal device image display information for displaying a plurality of images stored in the keyword storage means for storing the terminal device, and the terminal device receiving the image display information from the terminal device A selection image information receiving unit that receives selection image information indicating an image selected by a user from among images displayed by the apparatus, and a search for the related keyword associated with the image indicated by the received selection image information As a keyword, it is made to function as a search means for searching for information.

According to an eleventh aspect of the present invention, a computer included in an information retrieval device that retrieves information in response to a request from a terminal device connected via a network is associated with an image and a keyword related to the image. Image display information transmitting means for transmitting to the terminal device image display information for displaying a plurality of images stored in the keyword storage means for storing the terminal device, and the terminal device receiving the image display information from the terminal device A selection image information receiving unit that receives selection image information indicating an image selected by a user from among images displayed by the apparatus, and a search for the related keyword associated with the image indicated by the received selection image information As an information keyword, an information retrieval program that functions as a retrieval means for retrieving information is recorded in a computer-readable manner. Characterized in that it is.

It is a figure showing an example of outline composition of shopping system S concerning one embodiment. It is a figure which shows an example of the function outline | summary of the shopping system S which concerns on one Embodiment. It is a block diagram which shows an example of schematic structure of the shopping server 1 which concerns on one Embodiment. It is a flowchart which shows the process example of the goods review registration and attribute information extraction process in the system control part 20 of the shopping server 1 which concerns on one Embodiment. It is a flowchart which shows the process example of the search process in the system control part 20 of the shopping server 1 which concerns on one Embodiment. It is an example of a screen display of an image selection page.

Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. The embodiment described below is an embodiment when the present invention is applied to a shopping system in which goods are bought and sold by exchanging electronic information on a network.

[1. Overview of shopping system configuration and functions]
First, the structure and outline | summary function of the shopping system S which concern on this embodiment are demonstrated using FIG.1 and FIG.2.

FIG. 1 is a diagram illustrating an example of a schematic configuration of a shopping system S according to the present embodiment.

As illustrated in FIG. 1, the shopping system S includes a shopping server 1 as an example of an information search device and a plurality of user terminals 2 as an example of a terminal device. And the shopping server 1 and each user terminal 2 can mutually transmit / receive data via the network NW, for example using TCP / IP etc. as a communication protocol. The network NW is constructed by, for example, the Internet, a dedicated communication line (for example, a CATV (CommunityCommunAntenna Television) line), a mobile communication network (including a base station, etc.), a gateway, and the like.

In the shopping system S having such a configuration, the shopping server 1 is a Web server that transmits a Web page constituting the shopping site to the user terminal 2 in response to a request from the user terminal 2. Then, the shopping server 1 performs processing such as product search and purchase based on a request from the user terminal 2 of the user who browsed the Web page.

The user terminal 2 is a terminal device used by a user who purchases a product at a shopping site. As the user terminal 2, for example, a personal computer, a PDA (Personal Digital Assistant), a mobile phone or the like is used.

FIG. 2 is a diagram illustrating an example of a functional outline of the shopping system S according to the present embodiment.

As a search for merchandise in the shopping system S, it is possible to search based on an image selected from a plurality of images with a psychological feeling of the user. The outline of this search will be described below.

The user can input a product review text (an example of text information in the present invention) for a product that can be purchased using the shopping system S. The content indicated by the product review text is, for example, an opinion such as an impression of using the product, a critique of the product, or a comment. The product review text input by the user is transmitted from the user terminal 2 to the shopping server 1 and registered by the shopping server 1. The user can browse the registered product review text by accessing the shopping server 1 using the user terminal 2.

When the product review text is input, keywords are extracted from the product review text in the shopping server 1 (FIG. 2 (1)). The extracted keyword is registered as attribute information (an example of a keyword in the present invention) of an image of a product for which a product review text is input. The product images are stored in the shopping server 1 in association with the products. The product image may be, for example, a photographic image obtained by photographing the product itself or a photographic image obtained by photographing an item related to the product. The product image is, for example, an image displayed on a product detail page, which is a Web page that displays product description information and the like. The product description information is text information that describes the product specifications, contents, and other information related to the product. Since a plurality of images may be associated with one product, the keyword extracted as attribute information from the product review text corresponds to one representative image (an example of an image in the present invention) that exists for each product. Attached and registered. The product representative image is an image that is displayed on the search result page, which is a Web page that displays the search result of the product, together with the product name of the searched product.

Shopping server 1 transmits an image selection page, which is a Web page for selecting an image, to user terminal 2 in response to a request from user terminal 2. A plurality of representative images of products are displayed on the image selection page. A user arbitrarily selects an image from the displayed images on the image selection page (FIG. 2 (2)). The reference for selecting an image at this time is left to the user. For example, the user may select an image that suits his / her mood, or may select an image that recalls a matter close to his / her desire.

When an image is selected by the user, the shopping server 1 searches for a product in which the same word as the attribute information of the selected image is included in the description information ((3) in FIG. 2). That is, the shopping server 1 performs a keyword search for products using the attribute information as a keyword. And the shopping server 1 transmits the search result page which displays the list of searched goods as a search result to the user terminal 2, and the user terminal 2 displays a search result page (FIG. 2 (4)).

When the user selects any product from the search result page, the product detail page of the selected product is displayed on the user terminal 2. On the product detail page, the product name, price, product description information, product image, etc. of the selected product are displayed. Information on these products displayed on the product detail page is an example of information to be searched in the present invention. Moreover, some information is displayed on a search result page among the information of the product displayed on the product detail page of each searched product. That is, the shopping server 1 searches for product information and transmits a search result page displaying the searched information to the user terminal 2. In the present embodiment, the search for a product is a search for product information.

In the attribute information, which is a keyword extracted from the product review text that expresses what the user felt or thought about the product, or thought about the product, the user expressed it sensuously to the product It is thought that the word is included. Therefore, by searching for the product with the attribute information of the image selected by the user for his / her sense, it is possible to search for the product related to the user's sense. There are no specific conditions for inputting the product review text. That is, regardless of whether or not the user has purchased the target product using the shopping system S, and whether or not the user has used the target product, the user can read the product review text. Can be entered. Of course, a condition for inputting the product review text may be imposed. For example, only the person who purchased the target product using the shopping system S may be able to input the product review text.

[2. Shopping server configuration]
Next, the structure of the shopping server 1 is demonstrated using FIG.

FIG. 3 is a block diagram illustrating an example of a schematic configuration of the shopping server 1 according to the present embodiment.

As shown in FIG. 3, the shopping server 1 includes an operation unit 11, a display unit 12, a communication unit 13, a drive unit 14, a storage unit 15, an input / output interface unit 16, a system control unit 20, It has. The system control unit 20 and the input / output interface unit 16 are connected via a system bus 21.

The operation unit 11 includes, for example, a keyboard, a mouse, and the like, and receives an operation instruction from an operator or the like, and outputs the instruction content to the system control unit 20 as an instruction signal. The display unit 12 includes, for example, a CRT (Cathode Ray Tube) display, a liquid crystal display, and the like, and displays information such as characters and images. The communication unit 13 is connected to a network NW or the like and controls the communication state with the user terminal 2 or the like. For example, the drive unit 14 reads data from a disk DK such as a flexible disk, a CD (Compact Disc), a DVD (Digital Versatile Disc), and the like, and records data on the disc DK. The storage unit 15 is configured by, for example, a hard disk drive or the like, and stores various programs, data, and the like. The input / output interface unit 16 performs interface processing between the operation units 11 to 15 and the system control unit 20. The system control unit 20 includes a CPU (Central Processing Unit) 17, a ROM (Read Only Memory) 18, a RAM 19, and the like.

The storage unit 15 constitutes an example of a keyword storage unit in the present invention. Various databases such as a member DB (database) 101, a store DB 102, a product DB 103, and a member use DB 104 are constructed in the storage unit 15. In the member DB 101, information related to users registered as members in the shopping site (for example, member ID, password, name, address, telephone number, mail address, etc. as member identification information) is registered in association with each user. Has been. Further, in the store DB 102, information about stores that sell products on the shopping site (for example, store ID, password, store name, address, telephone number, e-mail address, store guidance information, etc. as store identification information) is stored. It is registered in association with each store. Further, in the product DB 103, information on products sold on the shopping site (for example, store ID of a store that sells products, product ID as product identification information, and genre ID as identification information indicating a genre to which the product belongs) , Product name, price, product description information, image ID as product image identification information, representative image image ID, product review text, etc.) are registered in association with each store and each product. In the product DB 103, image data of each product is registered in association with the image ID of the image. Further, in the product DB 103, attribute information of the representative image of each product is registered in association with the image ID of the representative image. In addition, in the member use DB 104, user information related to the use of the shopping site (for example, a member ID, purchase history, browsing history, etc.) is registered for each user.

Also, the storage unit 15 stores various HTML (Hyper Text Markup Language) documents, image data, audio data, text data, and the like that constitute the Web page of the shopping site.

Further, the storage unit 15 stores a predetermined operating system. In addition, the storage unit 15 uses the HTTP (Hyper Text Transfer Protocol) protocol to store Web pages based on various data stored in the storage unit 15 in response to a request transmitted from the user terminal 2 or the like. A WWW (World Wide Web) server program for transmission to the user terminal 2 or the like is stored. Furthermore, the shopping server 1 stores a shopping processing program that performs processing related to product search, purchase, and the like on the shopping site. The various programs may be acquired from other server devices or the like via the network NW, or may be recorded on a disk DK such as a CD-ROM and read via the drive unit 14. May be.

The system control unit 20 controls each unit of the shopping server 1 by the CPU 17 reading and executing various programs stored in the ROM 18 and the storage unit 15. The system control unit 20 also includes text information receiving means, keyword extracting means, image display information transmitting means, selected image information receiving means, search means, counting means, determination means, first selection means, and second selection means in the present invention. , Functioning as user information acquisition means, text information acquisition means, and third selection means.

In addition, you may comprise the shopping server 1 by several server apparatuses, such as a server which manages various databases, a search processing server which performs the search process of goods, and a WWW server which provides various information, for example.

[3. Operation of shopping system]
Next, operation | movement of the shopping system S is demonstrated using FIG. 4 thru | or FIG.

FIG. 4 is a flowchart showing a processing example of product review registration / attribute information extraction processing in the system control unit 20 of the shopping server 1 according to the present embodiment.

For example, the user designates a desired product and performs an operation for instructing input of a product review. Then, the user terminal 2 transmits a request for a product review input page, which is a Web page for inputting a product review text. The product review registration / attribute information extraction process is started when a request for a product review input page is received.

First, the system control unit 20 transmits a product review input page to the user terminal 2 (step S1). The user terminal 2 receives the product review input page and displays it on the screen. When the user inputs the product review text, the user terminal 2 transmits the product review text data and the product ID of the product specified by the user to the shopping server 1. Then, the system control unit 20 receives the product review text data and the product ID, and registers the product review text data in the product DB 103 in association with the product ID (step S2).

Next, the system control unit 20 performs morphological analysis on the received product review text data, and extracts keywords from the product review text (step S3). For example, the product review text for the product “XX Zenzai” says, “The taste is moderately sweet. For example, “Taste”, “Sweet”, “Texture”, “Shiratama”, “Smooth”, “Delicious”, “Everyone”, “Once”, “Eat”, “O Keywords such as “recommended” are extracted. In addition, since the technique itself which extracts a keyword based on a morphological analysis etc. is well-known, detailed description of the process which extracts a keyword is abbreviate | omitted.

Next, the system control unit 20 registers each extracted keyword as attribute information of the representative image of the product (step S4). Specifically, the system control unit 20 acquires the image ID of the representative image corresponding to the received product ID from the product DB 103. Then, the system control unit 20 registers each extracted keyword in the product DB 103 in association with the acquired image ID.

The system control unit 20 ends the product review registration / attribute information extraction process when the process of step S4 is completed.

FIG. 4 is a flowchart showing a processing example of search processing in the system control unit 20 of the shopping server 1 according to the present embodiment. FIG. 5 is a screen display example of an image selection page.

For example, the user performs an operation of selecting a search. Then, the user terminal 2 transmits to the shopping server 1 a request for a genre selection page that is a Web page for selecting a genre to which a product for which an image is displayed belongs. The search process is started when the shopping server 1 receives a request for a genre selection page.

First, the system control unit 20 transmits a genre selection page to the user terminal 2 (step S11). The user terminal 2 receives the genre selection page and displays it on the screen. A list of genres is displayed on the genre selection page. The user can select any genre on the genre selection page. The user can also select all genres. When the user selects any genre, the user terminal 2 transmits a genre ID indicating the selected genre to the shopping server 1, and the system control unit 20 receives the genre ID (step S12).

Next, the system control unit 20 generates an HTML document of the image selection page (an example of image display information in the present invention) (step S13). Specifically, the system control unit 20 selects a predetermined number of representative images of products belonging to the selected genre indicated by the received genre ID among the images of products registered in the product DB 103. At this time, the system control unit 20 selects representative images, for example, in ascending order of product price based on the preset sort condition. Then, the system control unit 20 generates an HTML document including an <img> tag for displaying each selected representative image.

Next, the system control unit 20 transmits an image selection page to the user terminal 2 (step S14). Specifically, the system control unit 20 transmits the generated HTML document to the user terminal 2. Then, the system control unit 20 receives a request for each representative image transmitted from the user terminal 2 that has received the transmitted HTML document, and the representative image corresponding to the URL (Uniform Resource Locator) included in the received request. Is acquired from the product DB 103 and transmitted to the user terminal 2.

The user terminal 2 displays an image selection page based on the received HTML document and image data. As shown in FIG. 6, the image selection page includes an image group display field 201, a selected image display field 202, a sort link group 203, a next image group display link 204, a search button 205, and the like. In the image group display column 201, a plurality of representative images of products belonging to the genre selected by the user are displayed. When the user selects one of the images displayed in the image group display column 201 by operating the user terminal 2, the selected image is displayed in the selected image display column 202. Note that the user can select one or a plurality of images. The sort link group 203 is a plurality of links for selecting sort conditions that determine the priority order and arrangement order of images displayed in the image group display field 201. For example, the sort conditions that can be selected include, for example, the order in which the price of the product is low, the order in which the price of the product is high, the new arrival order (the order in which the product information is registered in the product DB 103 is the newest), There are the order with the largest number of cases and the order with the smallest number of registered product review sentences. When the user selects any link in the sort link group 203, the shopping server 1 displays an image selection page in which the representative image of the product is displayed in the image group display column 201 under the sort condition corresponding to the selected link. 2 to send. The next image group display link 204 is a link for causing the current image group display field 201 to display an image different from the image displayed in the current image group display field 201.

Here, when the user selects an image and selects the search button 205, the user terminal 2 shopping for the image ID of the selected image displayed in the selected image display field 202 (an example of selected image information in the present invention). The system control unit 20 receives the image ID and transmits it to the server 1 (step S15).

Next, the system control unit 20 acquires attribute information corresponding to the received image ID from the product DB 103 (step S16).

Next, the system control unit 20 refers to the product DB 103 and selects one product belonging to the selected genre indicated by the genre ID received in step S12 (step S17).

Next, the system control unit 20 acquires the description information of the selected product from the product DB 103, and searches the keyword as the attribute information of the selected image acquired from the product DB 103 from the acquired description information (step S18).

Next, the system control unit 20 determines whether or not the description information of the selected product includes the same word as the attribute information of the selected image based on the search result of Step S18 (Step S19). At this time, when the description information of the selected product includes the same word as the attribute information of the selected image (step S19: YES), the system control unit 20 obtains the product ID of the selected product as a search result. It adds to the list (step S20). The search result list is information indicating a list of product IDs of the searched products. If there are a plurality of pieces of attribute information of the selected image, the fact that all the attribute information is included in the description information may be a condition for adding to the search result list (that is, performing an AND search) The condition that at least one attribute information is included in the description information may be a condition for adding to the search result list (that is, OR search is performed). Moreover, you may enable it for a user to select which conditions are used.

The system control unit 20 receives the information when the description information of the selected product does not include the same word as the attribute information of the selected image (step S19: NO) or when the process of step S20 is completed. It is determined whether or not there is a product that has not yet been selected among the products that belong to the selected genre indicated by the genre ID (step S21). At this time, if there is a product that has not yet been selected among the products that belong to the selected genre (step S21: YES), the system control unit 20 has not yet been selected from the products that belong to the selected genre. One non-commodity product is selected (step 17), and the processing of steps S18 to S21 is executed.

Then, the system control unit 20 generates a search result page based on the search result list when there is no product that has not yet been selected among the products belonging to the selected genre (step S21: NO). Step S22). Specifically, the system control unit 20 generates an HTML document for displaying a list of products indicated by the product ID set in the search result list based on information registered in the product DB 103 or the like.

Next, the system control unit 20 transmits the generated search result page to the user terminal 2 (step S23). On the search result page displayed by the user terminal 2, the representative image of the searched product, the product name, the price, the name of the store to be sold, and the like are displayed.

The system control unit 20 ends the search process when the process of step S23 is completed.

[4. Modified example]
[4.1 Modification 1]
In the above embodiment, the system control unit 20 searches for products by determining whether or not all the attribute information of the selected image selected on the image selection page is included in the description information. . However, the attribute information used for the search may be selected based on the degree of association between the image and the attribute information.

Specifically, the system control unit 20 registers the number of extractions of each attribute information in the product DB 103 in association with each attribute information of the representative image. When the system control unit 20 extracts a keyword from a product review text and registers it as attribute information, the same attribute information as the attribute information to be registered is the representative image of the product targeted by the input product review text 1 is added to the number of extractions associated with the attribute information.

The system control unit 20 determines whether the relationship with the selected image with which the attribute information is associated is high or low based on the number of extractions of the attribute information. For example, the system control unit 20 determines that attribute information whose extraction count value is larger than a preset threshold value is highly related to the selected image, and attribute information whose extraction count value is less than or equal to the threshold value. Determines that the relationship with the selected image is low.

Then, the system control unit 20 determines whether or not only the attribute information determined to be highly related to the selected image among the attribute information associated with the selected image is included in the description information. Search for products.

Note that the method for determining the relationship with the selected image based on the number of extractions of attribute information is not limited to the above. For example, the system control unit 20 calculates the total value of the number of extractions of attribute information for the selected image, and calculates the ratio of the number of extractions of each attribute information to the total value. Then, the system control unit 20 determines that the attribute information in which the calculated ratio is larger than the preset threshold value is highly related to the selected image, and the attribute information in which the calculated ratio is equal to or less than the threshold value is It may be determined that the association with the selected image is low. Further, the system control unit 20 determines that the attribute information from the first place to the preset order in the descending order of the number of extractions for the selected image is highly related to the selected image, and the order is set to the preset order. It may be determined that the attribute information not included in is not related to the selected image.

Further, the user may select whether to use attribute information that is highly related to the selected image or to use attribute information that is not related to the selected image. For example, on the image selection page, a button or the like for selecting a high and low relationship is provided. The shopping server 1 receives, from the user terminal 2, information indicating which one of the higher or lower relationship is selected together with the image ID of the selected image. Then, when the more related one is selected, the system control unit 20 searches for the product using only the attribute information determined to be highly related to the selected image. On the other hand, when the lower association is selected, the system control unit 20 searches for products using only the attribute information that is determined to be less relevant to the selected image.

Attribute information that has a low association with the selected image is a keyword extracted at least once from the product review text, so it is considered that there is at least some relationship with the selected image. However, since the degree of association is low, it is possible to present an unexpected product to the user as a search result.

[4.2 Modification 2]
Even when the same search word is input, a different search result may be presented to the user each time.

Specifically, the system control unit 20 changes the attribute information used for the search from the attribute information of the selected image registered in the product DB 103 and selects it every time the search is performed. For example, at least one of the attribute information used for the search may be changed between the previous search and the current search, or the number of attribute information used for the search may be changed. Moreover, you may determine the attribute information selected as attribute information used for a search at random, for example.

Thus, since the search result may change every time, the unexpectedness of the search result can be emphasized.

[4.3 Modification 3]
When selecting the attribute information used for the search from the attribute information of the selected image selected on the image selection page, the attribute information may be selected based on the user information. This selection aims to personalize the search results. The user information includes, for example, information indicating user attributes included in the user's personal information and profile, usage history and usage status of the shopping system S by the user, and the like. Information indicating user attributes is registered in the member DB 101, for example. The usage history and usage status of the shopping system S are registered in the member usage DB 104, for example.

First, a case where attribute information of a selected image is selected based on information indicating user attributes will be described. In this case, a keyword DB is constructed in the storage unit 15 of the shopping server 1. The keyword DB is a keyword (text information associated with the content of the user information in the present invention) related to the content of the information indicating the user's attribute for each content of information predetermined as a keyword registration target. Is an example of a registered database. Examples of information indicating user attributes include sex, age, and address. The keyword registration is performed by an administrator, for example.

For example, it is assumed that the information to be registered as a keyword is the gender of the user. In this case, in the keyword DB, keywords related to men are registered in association with identification information indicating men. In the keyword DB, keywords related to women are registered in association with identification information indicating women. The keyword related to a certain gender includes, for example, a keyword associated with the gender, a keyword related to things that are easily interested in people of that gender, and the like. A plurality of keywords may be registered for each gender.

Upon receiving the image ID of the selected image from the user terminal 2, the system control unit 20 of the shopping server 1 acquires the member ID of the user who uses the user terminal 2 that has transmitted the image ID of the selected image. Next, the system control unit 20 acquires the gender of the user corresponding to the acquired member ID from the member DB 101. Next, the system control unit 20 acquires a keyword corresponding to the acquired user's gender from the keyword DB. Further, the system control unit 20 acquires attribute information corresponding to the image ID of the received selected image from the product DB 103. Then, the system control unit 20 compares each attribute information acquired from the product DB 103 with each keyword acquired from the keyword DB.

Here, when certain attribute information matches any keyword acquired from the keyword DB, the system control unit 20 selects the attribute information as a keyword used for the search. Further, even when certain attribute information is included in any keyword acquired from the keyword DB, the system control unit 20 selects the attribute information as a keyword used for the search. Then, the system control unit 20 searches for product information using the selected attribute information. Thereby, it is possible to provide the user with a search result including information related to the user, such as information on products that the user is interested in.

Note that there may be a plurality of pieces of information to be registered as keywords. For example, a keyword for each gender, a keyword for each age group, a keyword for each prefecture where the address is located, and the like may be registered in the keyword DB. In this case, the system control unit 20 acquires the gender, age, address, and the like of the user who uses the user terminal 2 that has transmitted the image ID of the selected image from the member IDB 101. Next, the system control unit 20 acquires a keyword corresponding to the user's sex, a keyword corresponding to the age group to which the user's age belongs, and a keyword corresponding to the prefecture where the user's address is located from the keyword DB. Then, when the attribute information corresponding to the image ID of the received selected image matches or is included in any keyword acquired from the keyword DB, the system control unit 20 displays the attribute information. Select as a keyword to be used for search.

Next, the case where the attribute information of the selected image is selected based on the use history of the shopping system S by the user will be described. For example, assume that a purchase history of a product is used as the usage history. The product purchase history includes, for example, the product ID of the product purchased by the user.

Upon receiving the image ID of the selected image from the user terminal 2, the system control unit 20 of the shopping server 1 acquires the member ID of the user who uses the user terminal 2 that has transmitted the image ID of the selected image. Next, the system control unit 20 acquires a user purchase history corresponding to the acquired member ID from the member use DB 104. Next, the system control unit 20 acquires the product ID of the purchased product from the acquired purchase history. Next, the system control unit 20 acquires product description information (an example of text information associated with the content of user information in the present invention) corresponding to the acquired product ID from the product DB 103. That is, the system control unit 20 acquires the description information of the product that the user has purchased in the past. Then, the system control unit 20 searches the attribute information of the selected image using the attribute information as a keyword from the acquired description information.

Here, based on the search result, the system control unit 20 selects the attribute information as a keyword used for the search when the description information of the product includes the same word as the certain attribute information. Then, the system control unit 20 searches for product information using the selected attribute information. The product that the user has purchased in the past is, for example, a product that the user has demanded or a product that the user has been interested in. Therefore, it is possible to provide the user with a search result including information related to the user, such as information on a product that the user potentially desires or is interested in.

Further, the system control unit 20 may generate a search expression for each image selected by the user. Then, the system control unit 20 may generate a final search formula by combining the search formulas generated for each image. Specifically, the system control unit 20 performs a search using the attribute information of the selected image as a keyword from the description information of the product that the user has purchased in the past. Next, among the attribute information of the selected image, the attribute information retrieved from the product description information is grouped for each selected image. Next, the system control unit 20 generates a search expression indicating AND conditions for all keywords belonging to the group. That is, the system control unit 20 generates a search expression on the condition that all the attribute information searched from the product description information is included. Then, the system control unit 20 connects the search expressions generated for each selected image with the AND condition or the OR condition, and generates a final search expression.

Specific examples are shown below. For example, assume that the user selects image A and image B on the image selection page. As the attribute information of the image A, “sweet”, “banana”, “ripe”, and the like are associated. Further, as the attribute information of the image B, “Karai”, “Chilli”, “High Quality” and the like are associated. In addition, the description information of the product X that the user has purchased in the past includes words such as “mandarin orange”, “banana”, “sweet”, “karakashi”, “chili”, “karai”, etc. Yes.

The system control unit 20 retrieves each attribute information of the image A as the selected image from the description information of the product X. Then, the system control unit 20 determines that “sweet” and “banana” are included in the product information. Then, the system control unit 20 generates “sweet * banana” as a retrieval formula. Here, “*” indicates a logical product, that is, an AND condition. Further, the system control unit 20 retrieves each attribute information of the image B as the selected image from the description information of the product X. Then, the system control unit 20 determines that “product” and “red pepper” are included in the product information. Then, the system control unit 20 generates “Karai * Chilli” as a search expression. Note that the search formulas for each selected image, that is, “sweet * banana” and “garlic * chili” are examples of the first condition information in the present invention.

Next, the system control unit 20 connects the search expressions generated for each selected image, and selects “(sweet * banana) * (garlic * chili)” or “(sweet * banana) + (garlic * A search expression “Chilli”) is generated. The search formula generated at this time is an example of the second condition information in the present invention. Then, the system control unit 20 searches for a product based on the generated search expression. In other words, the system control unit 20 searches for a product in which all of “sweet”, “banana”, “garai”, and “chili” are included in the product description information. Alternatively, the system control unit 20 searches for a product in which both “sweet” and “banana” are included in the product description information, or both “Karai” and “Chilli” are included. . Note that when connecting a search expression for each selected image, whether to connect using an AND condition or to connect using an OR condition may be set in advance. Moreover, you may enable it for a user to select which conditions are used.

If the products purchased in the past are different for each user, even if the same image is selected on the image selection page, a different search expression may be generated. Therefore, a complicated search expression can be generated only by the user selecting an image. In addition, a search expression personalized for the user can be generated. Further, as the number of items purchased by the user increases, the search formula changes to a search formula suitable for the user. Thereby, search accuracy can be improved.

Further, the system control unit 20 may generate a keyword by connecting a plurality of attribute information for each selected image based on the grammatical rules. Specifically, connection rule information indicating how to connect words and generate a keyword used for search is stored in the storage unit 15. The connection rule information is, for example, information in which information indicating the part of speech of the connected word is listed in the word connection order. This connection rule information conforms to the grammar rules. Note that the connection rule information is set in advance by an administrator, for example. A plurality of connection rule information may be stored.

The system control unit 20 performs morphological analysis and the like when extracting a keyword serving as attribute information of the representative image from the product review text. Therefore, the system control unit 20 can specify the part of speech of the extracted keyword. Then, the system control unit 20 registers the attribute information extracted as a keyword in the product DB 103 in association with information indicating the part of speech.

Specific examples are shown below. For example, connection rule information indicating “adjective + noun” is stored in the storage unit 15. Similarly to the above-described example, it is assumed that the user selects the image A and the image B. Moreover, the user has purchased the product X in the past. The system control unit 20 determines that “sweet” and “banana” are included in the product information of the product X for the image A. Here, the part of speech of “sweet” is an adjective, and the part of speech of “banana” is a noun. Therefore, the system control unit 20 generates a keyword “sweet banana” based on the part of speech of each attribute information and the connection rule information. Similarly, the system control unit 20 generates a keyword “garlic chili” for the image B as well. The keywords generated for each selected image, that is, “sweet banana” and “garlic chili” are examples of the first condition information in the present invention.

Next, the system control unit 20 connects the keywords generated for each selected image, and generates a search expression of “sweet banana * hard pepper” or “sweet banana + hard pepper”. The search formula generated at this time is an example of the second condition information in the present invention. Then, the system control unit 20 searches for a product based on the generated search expression. In this case, the system control unit 20 searches for products for which both “sweet banana” and “garlic chili” are included in the product description information. Alternatively, the system control unit 20 searches for products for which at least one of “sweet banana” and “garlic chili” is included in the product description information. As described above, all the keywords determined to be included in the product information of the product X based on the connection rule information (for example, “sweet” and “banana” in the image A) are connected and the keyword is found. When generated, the generated keyword (for example, “sweet banana”) is provided on condition that all of the keywords included in the product information of the product X are included among the keywords associated with the selected image. The search condition shown (for example, “sweet * banana”) will also be shown.

For example, it is assumed that the description information of the product X does not include “sweet” but includes “banana” and “ripe”. In this case, for the image A, the system control unit 20 determines that “banana” and “completely matured” are included in the product information of the product X. Here, the parts of speech of “banana” and “completely matured” are nouns, and therefore “banana” and “completely matured” cannot be connected based on the connection rule information. There is no adjective word that can be connected to “banana” or “ripe”. In this case, the system control unit 20 does not use “banana” and “completely matured” in the search for merchandise.

If the products purchased in the past are different for each user, different keywords may be generated even if the same image is selected on the image selection page. Therefore, a personalized keyword for the user can be generated only by the user selecting an image. Further, as the user's purchased products increase, the keyword changes to a keyword suitable for the user. Thereby, search accuracy can be improved.

The system control unit 20 may select the attribute information as a keyword used for the search when the same word as the attribute information is included in the product name of the product that the user has purchased in the past. good. Further, the system control unit 20 searches the attribute information when the same word as the attribute information is included in at least one of the product name and the description information of the product that the user has purchased in the past. You may select as a keyword used for.

In addition, the generation of the search expression described above is not limited to the case where the use history of the shopping system S by the user is used. The system control unit 20 may generate the above-described search formula even when using information indicating user attributes.

As described above, according to the present embodiment, the system control unit 20 receives the product review text data input by the user from the user terminal 2, extracts the keyword from the product review text, and extracts the extracted keyword. Is registered in the product DB 103 in association with the product ID of the product for which the product review text is input (corresponding to the representative image associated with the product for which the product review text is input) as a plurality of representatives. An HTML document for displaying an image on the user terminal 2 is transmitted to the user terminal 2, and an image ID indicating an image selected by the user from the representative images displayed on the user terminal 2 is received from the user terminal 2 and received. The product information is searched using the attribute information of the selected image indicated by the image ID as a search keyword.

Therefore, the attribute information, which is a keyword extracted from the text input by the user about the product, is associated with the representative image of the product, and the attribute information of the image selected by the user among the plurality of images displayed by the user terminal 2 is used. The product information is retrieved using the search. The representative image of the product may be a photographic image obtained by photographing the product itself or a photographic image obtained by photographing an item related to the product. Therefore, the keyword extracted from the text input by the user for the product is a keyword related to the representative image. Therefore, by selecting an image with the user's sense, it is possible to search for product information related to the user's sense.

In addition, the product review text is a text that represents at least one of the user's impressions and opinions about the product.

Therefore, the extracted attribute information is likely to include attribute information indicating what the user felt or thought about the product.

Further, the system control unit 20 counts the number of extractions of keywords as attribute information for each attribute information corresponding to the representative image of each product, and based on the number of extractions, the relationship between the selected image and the attribute information is high or low. The attribute information determined to be highly relevant to the selected image among the attribute information registered in the product DB 103 may be selected, and the selected attribute information may be used for the search.

In this case, attribute information determined to be highly relevant to the corresponding selected image based on the number of extractions is used for the search. Therefore, it is possible to search for product information highly related to the user's sense.

In addition, when the system control unit 20 transmits information indicating the selection of the related level from the user terminal 2 and the one with the lower relationship is selected, the attribute information determined to have the lower relationship with the selected image is displayed. The selected attribute information may be used for the search.

In this case, information on a product having a low relevance to the user's sense is searched by the user's selection. Therefore, it is possible to search for information on unexpected products.

Further, the system control unit 20 may select attribute information used for search from the attribute information registered in the product DB 103, and change at least one of the selected attribute information each time the search is performed.

In this case, since at least one of the attribute information used for the search changes each time the search is performed, even if the user selects the same image as the previous time, the search result may change from the previous time. Therefore, it is possible to search for information on unexpected products.

In addition, the system control unit 20 acquires information on the user who has selected the selected image, acquires text information such as keywords and product description information associated with the acquired user information, and receives the information from the user terminal 2. A keyword included in the acquired text information may be selected from keywords associated with the image ID indicating the selected image, and the selected keyword may be used as a search keyword.

In this case, search results personalized for the user can be provided to the user.

In addition, for each selected image indicated by the image ID received from the user terminal 2, the system control unit 20 includes all keywords included in the text information associated with the user information among the keywords associated with the selected image. Is generated as a condition that is included in the product description information, and a search condition such as a keyword or keyword is generated, and the final search condition is generated by combining the generated search conditions between the selected images. Product information may be searched based on typical search conditions.

In this case, a personalized search condition can be generated as a search condition in which a plurality of keywords are combined simply by selecting an image by the user.

In the above-described embodiment, the system control unit 20 extracts a keyword that is attribute information of the representative image as a keyword related to the representative image of the product from the product review text. However, the system control unit 20 may extract a keyword that is attribute information of the representative image as a keyword related to the representative image of the product from, for example, at least one of the product name or the product description information.

The product name and product description information are registered in the product DB 103 together with other information related to the product. In order to register a product to be sold, a store employee or the like inputs information about the product including the product name and product description information by operating a terminal device that can be connected to the network NW, for example. The terminal device transmits the input information to the shopping server 1. The system control unit 20 registers the received information regarding the product in the product DB 103.

The system control unit 20 may extract a keyword when information related to a product is registered in the product DB 103. In this case, the system control unit 20 extracts keywords from the product name and product description information included in the information registered in the product DB 103. Then, the system control unit 20 registers each extracted keyword as attribute information of the representative image of the product. Note that these processes are basically the same as the processes in steps S3 and S4 in FIG.

Moreover, in the said embodiment, the shopping server 1 determines whether the same word as the attribute information of a selection image is contained in description information of goods, and the same word as attribute information is contained in description information. I was searching for a product. That is, in order to search for a product using the attribute information of the selected image, product description information has been used. However, information about the product other than the description information may be used. For example, a product name may be used, and both a product name and description information may be used.

In the above embodiment, the object of the commercial transaction in the present invention is applied to a product, and the present invention is applied to a shopping system for buying and selling the product. However, the object of application is not limited to this. For example, the object of the commercial transaction in the present invention may be applied to a service that can be traded, and the present invention may be applied to a system for trading the service. Examples of services that can be traded include provision of accommodation facilities, transportation by transportation, and provision of competition facilities such as golf courses. As a system for trading services, there are, for example, a reservation system for accommodation facilities, a reservation system for boarding passes and boarding tickets for transportation facilities, a reservation system for use of competition facilities, and the like.

In addition, when the information to be searched is applied to the information of the commercial transaction, the system to which the present invention is applied may not be configured to actually trade the commercial transaction. The present invention can be applied to any system that searches for at least a commercial transaction target and provides the user with information related to the retrieved commercial transaction target.

In addition, the information to be searched in the present invention is not limited to information on products and services that are targets of commercial transactions. For example, information to be searched may be applied to information such as a web page, a still image, a moving image, sound, text, and an electronic document. When the search target is a Web page, the image that can be selected on the image selection page as an image associated with the Web page may be, for example, a thumbnail of an image posted on the Web page. In this case, for example, the identification information of the Web page (for example, the URL of the HTML document of the Web page) and the thumbnail may be associated and registered in the database. In addition, when the search target is a still image or a moving image, the image that is selectably displayed on the image selection page as an image associated with the still image or the moving image may be, for example, a thumbnail of a still image or a moving image. . In this case, for example, identification information of still images or moving images (for example, URLs of still images or moving images) and thumbnails may be associated with each other and registered in the database. The keyword associated with the image as the keyword associated with the image associated with the information to be searched may be, for example, the title of the information to be searched. The keyword associated with the image as a keyword associated with the image associated with the search target information is, for example, a keyword extracted from an HTML document, an electronic document, text, or the like of the Web page to be searched, It may be a keyword extracted from a sentence obtained by speech recognition from the target speech.

In addition, for example, information to be searched may be applied to information such as news, events, and events. Such information may be conveyed by a web page, a still image, a moving image, sound, text, an electronic document, or the like. The image displayed so as to be selectable on the image selection page as an image associated with news, an event, an event, or the like may be, for example, an image that shows a situation where an event or an event is occurring. The images associated with news, events, events, etc. are, for example, thumbnails of images posted on web pages that convey news, events, events, etc., and thumbnails of still images and videos that convey news, events, events, etc. Etc.

Further, in the above-described embodiment, the images displayed on the image selection page so as to be selectable are images of the products themselves registered in the product DB 103 or images related to the products. In other words, images that can be selected on the image selection page correspond to information to be searched as images of things themselves indicated by information to be searched or images related to things indicated by information to be searched. It was applied to the attached image. However, the image displayed on the image selection page so as to be selectable may not be an image associated with the information to be searched. For example, the image displayed so as to be selectable on the image selection page may be an arbitrary image. For example, the image displayed on the image selection page so as to be selectable may be an image registered by an administrator or the like, an image posted by a user, an image collected from the WWW, or the like. In this case, for example, a keyword input by an administrator, a user, or the like may be associated with the image as a keyword related to the image. An administrator, a user, or the like may input a word related to the image, such as a word indicated by the image or a word associated with the image, as a keyword related to the image. In addition, a keyword extracted from an HTML document of a Web page on which an image is posted may be associated with the image as a keyword related to the image.

Further, when the keyword is associated with the image in such a form, the information search apparatus does not have to extract the keyword from the text information input by the user.

DESCRIPTION OF SYMBOLS 1 Shopping server 2 User terminal 11 Operation part 12 Display part 13 Communication part 14 Drive part 15 Storage part 16 Input / output interface part 17 CPU
18 ROM
19 RAM
20 System control unit 21 System bus 101 Member DB
102 Store DB
103 Product DB
104 Member use DB
NW Network S Shopping System

Claims

An information search device that searches for information in response to a request from a terminal device connected via a network,
Image display information transmitting means for transmitting, to the terminal device, image display information for causing the terminal device to display a plurality of images stored in a keyword storage means for storing an image and a keyword related to the image in association with each other;
Selection image information receiving means for receiving, from the terminal device that has received the image display information, selected image information indicating an image selected by a user among images displayed by the terminal device;
Search means for searching for information using a keyword associated with an image indicated by the received selected image information as a search keyword;
An information retrieval apparatus comprising:
The information search device according to claim 1,
The keyword storage means stores an image associated with information to be searched,
Text information receiving means for receiving text information input by the user about information to be searched or things indicated by the information from the terminal device;
A keyword is extracted from the received sentence information, and the extracted keyword is stored in the keyword storage unit in association with an image associated with information on which the received sentence information is input or information indicating things. Keyword extraction means for
An information retrieval apparatus further comprising:
The information search device according to claim 2,
Count means for counting the number of extractions by the keyword extraction means for each keyword stored in the keyword storage means;
Determination means for determining the level of association between an image stored in the keyword storage means and a keyword based on the number of extractions;
A first selection means for selecting a keyword determined to have a high association with the image from keywords associated with the image indicated by the received selected image information;
Further comprising
The information search apparatus, wherein the search means uses a keyword selected by the first selection means as a search keyword.
In the information search device according to claim 3,
Further comprising selection information receiving means for receiving selection information indicating the selection by the related high and low users from the terminal device;
The information selecting apparatus according to claim 1, wherein the first selection unit selects a keyword that is determined to have a low association with an image when the selection information indicates that a lower association is selected.
The information search device according to any one of claims 1 to 4,
A second selection means for selecting a keyword used by the search means from keywords associated with the image indicated by the received selection image information;
The search means uses the keyword selected by the second selection means as a search keyword,
The information search apparatus according to claim 2, wherein the second selection unit changes at least one of the keywords to be selected every time the search unit searches.
In the information search device according to any one of claims 1 to 5,
User information acquisition means for acquiring information of a user who has selected an image indicated by the received selection image information;
Text information acquisition means for acquiring text information associated with the content of the acquired user information;
Third selection means for selecting a keyword included in the acquired text information from keywords associated with an image indicated by the received selection image information;
Further comprising
The information search apparatus according to claim 1, wherein the search means uses the keyword selected by the third selection means as a search keyword.
The information search device according to claim 6,
For each image indicated by the received selected image information, all the keywords selected by the third selection means among the keywords associated with the image are included in the information to be searched. First generation means for generating first condition information shown as:
Second generation means for combining the generated first condition information between images indicated by the received selection image information to generate second condition information indicating a search condition used by the search means;
Further comprising
The information search device, wherein the search means searches for information based on the generated second condition information.
The information search device according to any one of claims 1 to 7,
An information search apparatus characterized in that the information to be searched is information for a commercial transaction.
An information search method for searching for information in response to a request from a terminal device connected via a network,
An image display information transmitting step for transmitting, to the terminal device, image display information for causing the terminal device to display a plurality of images stored in a keyword storage unit that associates and stores an image and a keyword related to the image;
A selected image information receiving step for receiving selected image information indicating an image selected by a user from among the images displayed by the terminal device from the terminal device that has received the image display information;
A search step of searching for information using a keyword associated with an image indicated by the received selected image information as a search keyword;
A method for retrieving information, comprising:
A computer included in an information search device that searches for information in response to a request from a terminal device connected via a network,
Image display information transmitting means for transmitting, to the terminal device, image display information for causing the terminal device to display a plurality of images stored in a keyword storage means for storing an image and a keyword related to the image in association with each other;
Selected image information receiving means for receiving selected image information indicating an image selected by a user among images displayed by the terminal device from the terminal device that has received the image display information; and
Search means for searching for information using the related keyword associated with the image indicated by the received selected image information as a search keyword;
Information search program characterized by functioning as
A computer included in an information search device that searches for information in response to a request from a terminal device connected via a network,
Image display information transmitting means for transmitting, to the terminal device, image display information for causing the terminal device to display a plurality of images stored in a keyword storage means for storing an image and a keyword related to the image in association with each other;
Selected image information receiving means for receiving selected image information indicating an image selected by a user among images displayed by the terminal device from the terminal device that has received the image display information; and
Search means for searching for information using the related keyword associated with the image indicated by the received selected image information as a search keyword;
A recording medium in which an information retrieval program that functions as a computer is recorded so as to be readable by a computer.