JP5108679B2 - User recommendation system, user recommendation method, product information output system, product information output method and program - Google Patents

Description

  The present invention relates to a user recommendation system, a user recommendation method, a product information output system, a product information output method, and a program.

In online shopping using the Internet, there are an increasing number of cases where reviews of product purchasers or users are posted on product introduction pages. By posting reviews, you can shop with reference to the popularity of products, usability, and other people's opinions. Various techniques using user evaluation have also been proposed. For example, Patent Literature 1 identifies users who are familiar with a predetermined specialized field, ranks the identified users, and has a highly reliable user group. An information recommendation device that makes use of the trend of information retrieval behavior by the information recommendation is disclosed. Patent Document 2 discloses a method for recommending highly evaluated products based on the reliability between users.
JP 2004-192542 A JP 2004-326634 A

  The more popular product reviews tend to be in large numbers, and the evaluation often varies. In addition, there is an example in which a review that lowers the evaluation of a product by mischief is performed. In such a case, the user cannot determine which opinion should be trusted.

  On the other hand, as disclosed in Patent Document 1, when the user's reliability is evaluated based on the user's specialized field and the information recognized by the highly reliable user is recommended, It is necessary to make a list of special fields of each user in advance, which is troublesome.

  Further, as disclosed in Patent Document 2, when recommending a product that is highly appreciated by a user trusted by the person based on the reliability between the users, the person himself / herself must be a reliable user. You have to search and register at the site, which takes time.

  The present invention has been made in view of such a background, and provides a user recommendation system, a user recommendation method, a product information output system, a product information output method, and a program capable of automatically evaluating information. For the purpose.

A main invention of the present invention for solving the above-mentioned problems is a system for recommending other users to a user, and an evaluation memory for storing an evaluation value set by the user for an evaluation object for each of the evaluation object and the user A recommendation destination specifying unit that specifies a recommendation destination user who is a user who recommends the other user, and the maximum evaluation target that is set with a maximum value that the recommendation destination user can take the evaluation value. A maximum evaluation target selection unit that selects an evaluation target from the evaluation storage unit, and a minimum evaluation target that is the evaluation target for which the recommended user has set a minimum value that can be taken by the evaluation value is selected from the evaluation storage unit. A minimum evaluation target selection unit that sets the maximum value for the maximum evaluation target, and among the other users different from the recommended user, The candidate user is the user who set the minimum value, the recommendation candidate selecting unit that selects from the evaluation storage unit, for each of the user, and the user information storage unit for storing user information about the user, corresponding to said candidate user A recommendation candidate presenting unit that reads out the user information from the user information storage unit and displays the user information to the recommended user.

  The evaluation target includes, for example, all items that can be evaluated by the user, such as products, services, events, movies, music, and books. According to the user recommendation system of the present invention, the evaluation target given the highest score by the recommendation destination user is given the highest score to the recommendation destination user in the same manner as the recommendation destination user, and the evaluation given by the recommendation destination user is given the lowest score. As for the target, it is possible to recommend candidate users with the lowest score as with the recommended user. It is considered that when a user gives the highest score for a certain evaluation target, the user often likes the evaluation target, and those who like the same thing often agree with each other. Furthermore, it is considered that when a user gives the lowest score for an evaluation object, the user often dislikes the evaluation object, and people who dislike the same object often agree with each other. According to the user recommendation system of the present invention, it is possible to introduce a candidate user who likes a recommendation destination user and also likes a recommendation destination user and also dislikes a recommendation destination user. Thereby, in the user recommendation system of the present invention, it is possible to introduce candidate users who are considered to agree with the recommended user. Therefore, it is possible to create an opportunity for exchange between the recommended user and the candidate user. For example, in a system that provides a place for communication between users, such as a social network system, communication can be activated.

In the user recommendation system of the present invention, the maximum evaluation target selection unit sets the evaluation target set by the recommended user as an evaluation value equal to or higher than a threshold obtained by subtracting a predetermined value from the maximum value as the maximum evaluation target. The recommendation candidate selection unit sets an evaluation value equal to or greater than the threshold for the maximum evaluation target, and a user who sets the minimum value for the minimum evaluation target among users different from the recommended user. You may make it select as said candidate user.
In this case, a candidate user who has also given a high evaluation can be introduced to the recommendation destination user for the evaluation target that has been given a high evaluation by the recommendation destination user. Therefore, for example, even when the user gives a score conservatively while avoiding the highest score at the time of evaluation, a compatible user can be introduced.

Further, in the user recommendation system of the present invention, the minimum evaluation target selection unit sets, as the minimum evaluation target, the evaluation target in which the recommendation destination user sets an evaluation value equal to or less than a threshold obtained by adding a predetermined value to the minimum value. The recommendation candidate selection unit selects a user who sets the maximum value for the maximum evaluation target and sets an evaluation value equal to or less than the threshold for the minimum evaluation target among users different from the recommendation destination user. You may make it select as a candidate user.
In this case, candidate users who have similarly given a low evaluation can be introduced to the recommendation destination user for the evaluation target given a low evaluation by the recommendation destination user. Therefore, for example, even when the user gives a score conservatively while avoiding the lowest score at the time of evaluation, a more compatible user can be introduced.

  Further, in the user recommendation system according to the present invention, the maximum evaluation target selection unit is configured such that the recommendation target user sets an evaluation value equal to or higher than a first threshold value obtained by subtracting a first predetermined value from the maximum value. Is selected as the maximum evaluation target, and the minimum evaluation target selection unit sets the evaluation target for which the recommended user sets an evaluation value equal to or lower than a second threshold value obtained by adding a second predetermined value to the minimum value. Is selected as the minimum evaluation target, and the recommendation candidate selection unit sets the evaluation value equal to or higher than the first threshold for the maximum evaluation target, and among the users different from the recommended user, the minimum evaluation target A user who has set an evaluation value equal to or less than the second threshold value may be selected as the candidate user.

In the user recommendation system of the present invention, the recommendation candidate selection unit refers to the evaluation storage unit and determines whether each user has set the maximum value for each of the maximum evaluation targets. And calculating a maximum evaluation match number that is the number of the maximum evaluation targets for which each user has set the maximum value, and selecting a predetermined number of the users as the candidate users in descending order of the maximum evaluation match number. It may be.
It is considered that the more the two users give the same highest score, the better the compatibility of the two users. In this case, candidate users with better compatibility can be introduced to the recommended user. In addition, since it is possible to recommend candidate users after determining whether or not the evaluation values match for more evaluation targets, it is possible to increase the reliability of the compatibility determination.

  In the user recommendation system of the present invention, the recommendation candidate selection unit further refers to the evaluation storage unit to determine whether each user has set the minimum value for each of the minimum evaluation targets. Determining and calculating a minimum evaluation match number that is the number of the minimum evaluation targets for which each user has set the minimum value, and the recommended candidate selection unit adds the maximum evaluation match number and the minimum evaluation match number A predetermined number of the users may be selected as the candidate users in descending order.

Further, the user recommendation system of the present invention, the user information includes the name of the user, it said include an access information which is information for establishing communication with a user, the recommended candidate presentation unit , the user information corresponding to the candidate user may be listed.
The access information is, for example, a user's e-mail address or a URL (Uniform Resource Locator) of a Web page on which the user provides information. By displaying the user information, the recommended user can easily interact with the candidate user, so that communication can be activated.

  Another aspect of the present invention is a method for recommending another user to a user, wherein the computer stores an evaluation value set by the user for the evaluation target in a memory for each of the evaluation target and the user, A recommended user who is a user who recommends other users is specified, and the maximum evaluation target that is the evaluation target in which the recommended user has set the maximum value that can be taken by the evaluation value is selected from the memory. Selecting the minimum evaluation target, which is the evaluation target for which the recommended user has set the minimum value that can be taken by the evaluation value, from the memory, and setting the maximum value for the maximum evaluation target; A candidate user who is a user who has set the minimum value for the minimum evaluation target is selected from the memory among the other users different from each other, and the candidate user is selected from the memory To be presented to the Como destination user.

  According to another aspect of the present invention, there is provided a program for recommending another user to a user, wherein an evaluation value set by the user for the evaluation target is stored in a memory for each of the evaluation target and the user. A step of identifying a recommendation destination user who is a user who recommends the other user; and a maximum evaluation target that is the evaluation target in which the maximum value that the evaluation destination user can take the evaluation value is set. Selecting from the memory a step of selecting from the memory, a step of selecting from the memory a minimum evaluation target that is the evaluation target for which the recommended user has set a minimum value that can be taken by the evaluation value, and the maximum evaluation target Candidates who are users who set the minimum value for the minimum evaluation target among users different from the recommended user who set the value The chromatography THE, and selecting from said memory, and thereby execute the steps of: presenting the candidate user to the recommended destination user.

  Other problems and solutions to be disclosed by the present application will be made clear by the embodiments of the invention and the drawings. In this application, the product includes a service.

  According to the present invention, it is possible to automatically evaluate information about products.

== First Embodiment ==
Hereinafter, a shopping system for providing a product sales service according to an embodiment of the present invention will be described. In this embodiment, the service is also sold as a product. In the shopping system of this embodiment, it is assumed that information about a product is provided from the user, and when the information about the product is displayed, the information provided from the user about the product is also displayed.

  In the present embodiment, the information provided from the user about the product is the user's evaluation of the product (hereinafter referred to as a review). The review includes a comment from a user who evaluated the product (hereinafter also referred to as a reviewer) and a score (hereinafter referred to as a rating) given by the reviewer to the product. A user who intends to browse product information (hereinafter also referred to as a viewer) can browse reviews of other users along with product information.

  In the following description, a process of outputting information on a product viewed by a viewer will be described, and a description of a process such as a product order process and a process of registering a review posted from a reviewer will be omitted. Product order processing and review registration processing are assumed to be used in a general online shopping system.

  FIG. 1 is a diagram showing an overall configuration of a shopping system according to an embodiment of the present invention. As shown in the figure, the shopping system of this embodiment includes a shopping server 20, and a plurality of client terminals 10 are connected to the shopping server 20 via the Internet 30. The Internet 30 is a communication network constructed using a public telephone line network, a wireless line, and Ethernet (registered trademark), and communication is performed on the Internet 30 according to a TCP / IP protocol. Instead of the Internet 30, a value added network (VAN), a local area network (LAN), or the like may be used.

  The shopping server 20 is a computer such as a personal computer or a workstation that provides an online shopping service. In the present embodiment, the shopping server 20 provides an online shopping service as a Web application. The shopping server 20 provides information about the product as a Web page described in HTML (HyperText Markup Language) in response to an HTTP (HyperText Transfer Protocol) request, and performs sales processing of the product in response to an HTTP request. .

  The client terminal 10 is a computer operated by a user who uses online shopping, such as a personal computer, a workstation, a PDA (Personal Digital Assistance), or a mobile phone. The user operates a Web browser that operates on the client terminal 10 to access the shopping server 20 to browse product information or place an order for the product.

  FIG. 2 is a diagram illustrating a hardware configuration of the shopping server 20. As shown in the figure, the shopping server 20 includes a CPU 201, a memory 202, a storage device 203, a communication interface 204, an input device 205, and an output device 206.

  The storage device 203 stores various programs and data, for example, a hard disk drive, a CD-ROM drive, a flash memory, or the like. The CPU 201 implements various functions by reading the program stored in the storage device 203 into the memory 202 and executing it.

  The communication interface 204 is an interface for connecting to the Internet 30. For example, a modem for connecting to a public telephone line network, an adapter for connecting to Ethernet (registered trademark), a wireless communication device for performing wireless communication, and the like. It is.

  The input device 205 is a keyboard, a mouse, a touch panel, or the like that accepts input of information. The output device 206 is, for example, a display or a printer that outputs information. Note that the input device 205 and the output device 206 can be detached from the shopping server 20 so that the input device 205 and the output device 206 are not always connected.

  FIG. 3 is a functional block diagram of the shopping server 20. As shown in the figure, the shopping server 20 includes a product screen acquisition request reception unit 211, a product information acquisition unit 212, a review information acquisition unit 213, a reliability calculation unit 214, a product screen generation unit 215, a product screen transmission unit 216, A product database 31, a user database 32, a purchase history database 33, and a review database 34 are provided.

  The product database 31 stores information about products (hereinafter referred to as product information). FIG. 4 shows a configuration example of product information stored in the product database 31. As shown in the figure, the product information includes a product ID, a product name, a product price, a classification to which the product belongs, a stock, and a product image. The product ID is identification information that uniquely identifies the product. The product name is the name of the product indicated by the product ID. The stock is the number of products in stock and indicates whether the product can be sold. The product image is a URL (Uniform Resource Locator) indicating a file storing product image data, which is displayed when a screen about the product is displayed.

  The user database 32 stores information about users. A configuration example of user information stored in the user database 32 is shown in FIG. As shown in the figure, the user information stored in the user database 32 includes identification information for uniquely identifying the user (hereinafter referred to as user ID), and the user's name, address, and telephone number. .

  The purchase history database 33 stores information related to products purchased by the user in the past (hereinafter referred to as purchase history information). FIG. 6 is a diagram illustrating a configuration example of purchase history information stored in the purchase history database 33. As shown in the figure, the purchase history information includes a product ID indicating the purchased product, a user ID (purchaser ID) indicating the purchased user, the date and time when the user purchased the product, and the purchased product. Including the number of purchases.

  The review database 34 (corresponding to the provision information storage unit of the present invention) stores information including reviews posted from reviewers (hereinafter referred to as review information. Corresponding to provision information of the present invention). A configuration example of the review information stored in the review database 34 is shown in FIG. As shown in the figure, the review information is associated with the review ID for identifying the review, the product ID indicating the product subject to the review, the user ID indicating the reviewer who posted the review, and the reviewer posted the review. The date and time, as well as the content of the review (comments) and rating are included. In the present embodiment, the content of the review is plain text, and the rating indicates the rank of the review. In the present embodiment, the review is an integer of 0 to 5. Note that the items of review information used in the present embodiment are merely examples, and those used when accepting reviews from users in a general shopping site may be used as they are.

  The product screen acquisition request receiving unit 211 transmits a command (hereinafter referred to as a product screen acquisition request) that is transmitted from the client terminal 10 and that requests screen data (hereinafter referred to as a product screen) for displaying information about the product. Receive. In the present embodiment, the product screen is a Web page described in HTML. The product screen acquisition request is an HTTP request. In the product screen acquisition request, it is assumed that a user ID indicating a viewer who transmits the request and a product ID indicating a product to be displayed are set.

  The product information acquisition unit 212 reads product information corresponding to the product ID set in the product screen acquisition request from the product database 31.

  The review information acquisition unit 213 (corresponding to the provision information acquisition unit of the present invention) reads the review information corresponding to the product ID set in the product screen acquisition request from the review database 34. Further, the review information acquisition unit 213 reads the corresponding user information from the user database 32 using the user ID included in the read review information as a key.

  The reliability calculation unit 214 (corresponding to the evaluation value determination unit of the present invention) calculates the reliability of the review (the evaluation value of the review by the shopping server 20) for each of the review information read by the review information acquisition unit 213. To do. The reliability of the review is calculated so that the reliability of the review from the reviewer whose purchase pattern is similar to the user who browses the product information (that is, the user who transmitted the product screen acquisition request) is high. . Details of the process of calculating the reliability of the review will be described later.

The product screen generation unit 215 (corresponding to the provision information output unit of the present invention) generates a product screen. The product screen generation unit 215 adds the review information read by the review information acquisition unit 213 to the product information read by the product information acquisition unit 212. Add an icon to indicate that is high. In the present embodiment, an icon indicating that the reliability is high is prepared in advance as an image file, and a URL (Uniform Resource Locator) indicating the image file is described on the product screen.
The product screen transmission unit 216 transmits the product screen generated by the product screen generation unit 215 to the client terminal 10.

  FIG. 8 is a flowchart for explaining the flow of the product screen generation process by the shopping server 20. The process in FIG. 8 is executed when the product screen acquisition request receiving unit 211 receives a product screen acquisition request from the client terminal 10.

  The product information acquisition unit 212 reads product information corresponding to the product ID set in the product screen acquisition request from the product database 31 (S401). The review information acquisition unit 213 reads the review information corresponding to the product ID set in the product acquisition request from the review database 34, and also reads and reads the user information corresponding to the user ID of each review information from the user database 32. User information is attached to each review information (S402).

  For each of the review information read by the review information acquisition unit 213, the reliability calculation unit 214 performs a reliability calculation process illustrated in FIG. 9 (S403).

  As shown in FIG. 9, the reliability calculation unit 214 reads the product ID of the purchase history information corresponding to the user ID set in the product screen acquisition request from the purchase history database 33 and sets it as list 1 (S421). . That is, the list 1 is a list of products purchased in the past by the user who transmitted the request.

Further, the reliability calculation unit 214 reads out the product ID of the purchase history information corresponding to the user ID of the review information from the purchase history database 33 and sets it as a list 2 (S422). That is, the list 2 is a list of products that have been purchased in the past by reviewers who have submitted reviews for which reliability is to be calculated.
The reliability calculation unit 214 calculates the number of product IDs included in both the lists 1 and 2 (corresponding to the number of common product purchases of the present invention) as the review reliability (S423).

  When the reliability of the review is calculated by the process of FIG. 9, the product screen generation unit 215 returns to FIG. 8 and the product screen generation unit 215 has high reliability when the reliability is greater than a predetermined threshold (S404: YES). A URL indicating an icon indicating “” is added to the review information (S405).

  After performing the above processing for each review information, the product screen generation unit 215 sorts the review information in the order of reliability (S406), the product information read by the product information acquisition unit 212, the sorted review information, Is written out as a product screen (S407). Further, the product screen generation unit 215 may output the name of user information attached to the product information in association with the review information.

  A product screen is generated as described above. The generated product screen is transmitted to the client terminal 10 by the product screen transmission unit 216, and product information and reviews are displayed on the client terminal 10. An example of the screen 50 displayed on the client terminal 10 is shown in FIG.

  In the screen 50, product information is displayed in the display column 51, and review information is displayed in the display column 52. In the display column 51, the product image, product name, price, etc. of the product information are displayed. The display column 52 displays the rating of the review information in terms of the number of stars, the posting date of the review information, the content of the review, and the like. In the example of the screen 50, the user name of the user information attached to the review information is also displayed. In the example of the screen 50, the “name” of the user information is output. However, the present invention is not limited to this, and the user ID is output or the nickname is registered in the user information and the nickname is output. It may be.

  Two reviews 521 and 522 are displayed on the screen 50, and an icon 523 is displayed in association with the review 521. The icon 523 indicates the reliability of the review 521, and as described above, the reviewer (Taro Yamada) has purchased the same product as the product previously purchased by the user viewing the screen 50. Indicates.

  As described above, according to the shopping server 20 of the present embodiment, when the number of products that the reviewer has purchased in the past and the products that the viewer of the screen 50 has purchased in the past is equal to or greater than a predetermined number, An icon 523 is displayed for the review by the reviewer. In other words, reviews from reviewers with purchase trends similar to the purchase trends of the viewer can be identified.

  If the viewer and the reviewer have purchased similar products in the past, the viewer and the reviewer will continue to purchase similar products in the future. Here, the product specified in the product screen acquisition request has been reviewed by the reviewer, but the viewer has not yet purchased it, so the viewer and reviewer are similar in the past. If it is a purchase tendency, the viewer is likely to purchase the product, and information from the reviewer is useful to the viewer. Also, in general, people often listen to opinions from people who have bought similar products, that is, people with similar hobbies. Therefore, it is possible to present a review useful for the viewer by displaying the icon 523 on a review from a reviewer with a similar past purchase tendency. Therefore, it is convenient because the viewer can easily determine which review should be referred to.

  In addition, when the number of reviews is large, the content of the reviews tends to be diverse, but the viewer only needs to refer to the review on which the icon 523 is displayed. Useful reviews can be easily identified.

  Further, in the shopping system of the present embodiment, when a plurality of reviews are registered, a review with a high degree of reliability is displayed first, so that a viewer can easily identify a necessary review. be able to. Furthermore, in the shopping system of the present embodiment, the icons 523 are displayed only for those whose reliability exceeds a predetermined threshold value, so that the viewer can easily determine which review should be referred to. be able to.

== Second Embodiment ==
In the first embodiment described above, the reliability of the review for the viewer is determined so that a reviewer who has purchased the same product as the product previously purchased by the viewer has a high evaluation. Instead of this, the reliability of the reviewer who performed the same rating as the rating performed by the viewer in the past may be increased. The flow of the reliability calculation process in this case is shown in FIG.

  In the reliability calculation process illustrated in FIG. 11, the reliability calculation unit 214 initializes the number n of reviews for the same product to “0” (S441), and the same review that is the number of the same rating for the same product. The number is initialized to “0” (S442).

  The reliability calculation unit 214 acquires the review information corresponding to the user ID set in the product screen acquisition request from the review database 34 (S443). In order to distinguish the review information acquired in step S443 from the review information acquired in step S402 of FIG. 8, hereinafter, the review information acquired in step S402 of FIG. 8 is referred to as output review information. The review information acquired in step 443 in FIG. 11 is referred to as past review information.

  The reliability calculation unit 214 performs the following processing for each of the read past review information.

  The reliability calculation unit 214 sets the rating included in the past review information to r1 (S444), and the review information corresponding to the product ID included in the past review information and the user ID included in the output review information. The rating contained in is retrieved from the review database 34 and set as r2 (S445). Thus, the rating (r1) that the viewer has performed for a certain product in the past and the rating (r2) when the reviewer has previously reviewed the product are obtained.

  When r2 exists, that is, when r2 is not a NULL value (S446: YES), the reliability calculation unit 214 increments n (S447), and if r1 and r2 are the same value (S448: YES) ), The number of identical reviews is incremented (S449).

By performing the above processing for each of the past review information, the number of times that the viewer and the reviewer have reviewed the same product in the past (n), and the number of times the same rating was given to the same product (same number of reviews) Is required.
The reliability calculation unit 214 calculates the reliability of the output review information by multiplying the quotient obtained by dividing the number of the same reviews by n by 100 (S450).

  The reliability required by the above processing indicates the degree to which the viewer and the reviewer have made the same review on the same product. Therefore, it is possible to increase the reliability as the viewer and the reviewer perform the same rating on the same product.

  As in the second embodiment, when the reviewer uses the degree of rating similar to that of the viewer as the reliability, the purchase history of the product does not exist, that is, the product is stored in another shopping Even when a reviewer purchased on the site only provides information, the reliability of the reviewer can be calculated. As a result, for example, on sites that provide information on movies, music, games, etc., tickets for movies, music, games, etc. can be widely reviewed by non-buyers without purchasing tickets on the shopping sites. By collecting, it becomes possible for the convenience of the viewer.

  An example of the screen 60 output in this case is shown in FIG. FIG. 12 shows an example of a screen 60 used when a review target (product) is a movie. Also on the screen 60, as with the screen 50 of FIG. 10, movie information (product information) is output to the display column 61, and review information from the reviewer is output to the display column 62. The icon 621 is displayed when the number of reviews performed by the reviewer in the past on the same movie as the viewer exceeds a predetermined number.

  Users who have performed the same rating for the same product in the past are expected to perform the same rating for the same product in the future. Therefore, the rating given by a reviewer who has given a similar rating in the past is highly likely to be a rating that suits the viewer's preference, and is highly reliable for the viewer. Therefore, as described above, the reliability of the review from the reviewer who performed the same rating as the viewer is highly evaluated, and the icon 523 based on the reliability is output, so that the review is trusted for the viewer. It can be shown whether it is a high quality thing. Thereby, the viewer can easily determine whether or not the review is suitable for his / her preference.

  In addition, combining the first and second embodiments, the reliability 1 calculated according to the purchase tendency calculated in the first embodiment and the same rating calculated in the second embodiment were performed. The reliability of the review may be calculated in combination with the reliability 2 of which the degree is evaluated. In this case, a more accurate reliability can be calculated. Further, in this case, the reliability 1 is normalized so as to indicate the degree like the reliability 2 as a quotient obtained by dividing the reliability calculated by the processing of FIG. 9 by the size of the list 1 or the list 2. Alternatively, the reliability level 1 and the reliability level 2 may be weighted and then summed.

  Further, in step S448 of the process of FIG. 11, it is determined whether or not the rating of the viewer matches that of the reviewer. However, whether or not the absolute value of the difference between r1 and r2 exceeds a predetermined threshold value is determined. You may make it determine. In this case, the similarity between the rating of the viewer and the rating of the reviewer can be calculated. Therefore, it is possible to evaluate the reliability of the reviewer who performs the rating close to the viewer's preference, and it is possible to calculate the reliability with higher accuracy. Moreover, when calculating the similarity, it is possible to deal with the case where the rating is ranked more finely than five levels or an arbitrary numerical value.

  The rating of the viewer and the reviewer may be compared only when the rating is either the highest value (“5” in this embodiment) or the lowest value (“0” in this embodiment). . FIG. 13 shows the flow of the reliability calculation process in this case. The process of FIG. 13 is obtained by adding steps S461 and S462 to the process of FIG. 11 described above. In step S446, if r2 is set (S446: YES), the reliability calculation unit 214 increments n (S447), but if r1 is not 5 or 0 (S461: NO), the number of identical reviews Is decremented (S462). That is, the same review number is incremented only when r1 and r2 match at the highest or lowest rating.

  When a user assigns the highest rating, the user often likes the product to be reviewed. Similarly, when the user assigns the lowest rating, the user may dislike the product to be reviewed. Many. Therefore, as described above, the products rated by the viewer with the highest rating are highly evaluated by the reviewer with the highest rating, and the product with the lowest rating given by the viewer. As for the reviewer, the reliability of the reviewer with the lowest rating is also highly evaluated, so that the reliability of the reviewer having the same taste as the viewer can be highly evaluated. This increases the reliability of review by the same reviewer as the viewer likes and dislikes, so it is possible to calculate a highly accurate reliability for the viewer.

  FIG. 14 is a diagram showing an example of a screen 61 used when a review target is a game. In the example of FIG. 14, game information is output to the display field 611, and review information from the reviewer is output to the display field 612. With the process of FIG. 13, for a game in which the viewer has given the highest or lowest rating in the past, if the number of the highest or lowest value exceeds the predetermined number, 614 is displayed. Thus, since the reviewer with the icon 614 displayed with high reliability is considered to have the same preference as the viewer, the viewer can easily refer to the review.

  Further, in the present embodiment, all processing such as management of various databases and generation of product screens is performed by the shopping server 20. However, the present invention is not limited to this. For example, a database server that manages the database is installed separately. Alternatively, a plurality of servers may be installed so that the product screen generation process and the reliability calculation process are performed by different servers.

  FIG. 15 is a diagram illustrating a configuration example of a system in a case where the function provided by the shopping server 20 is realized by three servers, that is, the Web server 21, the commerce server 22, and the community server 23. In this case, the Web server 21 manages only the user database 32 and includes only the product screen acquisition request reception unit 211, the product screen generation unit 215, and the product screen transmission unit 216. The commerce server 22 manages the product database 31 and the purchase history database 33 so as to perform a product sales process. In addition, the community server 23 manages the review database 34, and includes a function of accepting reviews about products and registering them in the review database 34, a review information acquisition unit 213, and the like. Thus, by constructing the shopping server 20 with a plurality of servers, various loads can be distributed, the processing efficiency can be improved, and the response time for the client terminal 10 can be shortened.

  In the shopping system of the present embodiment, the rating is a value from 0 to 5. However, the rating is not limited to this and may be an integer or real number in an arbitrary range. The rating may be a character string or a numerical value indicating a category such as “like” or “dislike”. In this case, instead of the star-shaped figure on the screen 50 in FIG. 10, the rating value can be displayed, and the foreground color, background color, font size, etc. of the review can be changed according to the rating value. .

  In the shopping system of the present embodiment, whether to display the icon 523 on the screen 50 is determined depending on whether the reliability exceeds a predetermined threshold. However, the present invention is not limited to this, and the reliability is a numerical value. It may be configured to output, or a different icon 523 may be displayed depending on the reliability. For example, if the reliability is ranked in three levels and icons of gold, silver, and copper are displayed according to the rank, it is convenient for the viewer to understand.

  In the shopping system of the present embodiment, the screen 50 that displays information about the product indicated by the product ID set in the product screen acquisition request is assumed. However, the present invention is not limited to this. For example, a list of products is displayed. When displaying, it is possible to display the number of reviews with high reliability about the product. In this case, the product ID is not set in the request from the client terminal 10, and only the user ID is set, the product information is read from the product database 31, and the review information corresponding to the product ID of the product information is The reliability described above is calculated.

  Further, when displaying a list of products, the product information may be sorted and displayed in the order of the number of review information whose reliability exceeds the threshold. Thereby, since the viewer can easily discriminate the products with many reviews that can be trusted, it is possible to motivate the user to view the details of the products, and the number of browsing of the Web page can be increased.

  In the shopping system of the present embodiment, the information provided from the user about the product is a review. However, the present invention is not limited to this. Information.

== Third embodiment ==
Next, a social networking service (SNS) system according to the third embodiment of the present invention will be described. The SNS system according to the third embodiment is a computer system that supports communication between users. In the SNS system, each user presents his / her profile, introduces friends to other users, participates in a community where users with the same hobby exchange information, and between other users. To exchange messages such as e-mail.

  The SNS system of the third embodiment recommends a user having the same preference as a communication partner to a certain user.

  The overall configuration of the SNS system of the third embodiment is the same as that of the first embodiment, and an SNS server 70 is connected to the Internet 30 instead of the shopping server 20 in FIG. It is assumed that the hardware configuration of the SNS server 70 is the same as that of the shopping server 20 shown in FIG.

  The software configuration of the SNS server 70 is shown in FIG. As shown in the figure, the SNS server 70 includes a user screen acquisition request reception unit 711, a review information acquisition unit 713, a compatibility degree calculation unit 714, a user screen generation unit 715, a user screen transmission unit 716, an evaluation target database 71, a user A database 72 and a review database 74 are provided. User screen acquisition request reception unit 711, review information acquisition unit 713, compatibility degree calculation unit 714, user screen generation unit 715, user screen transmission unit 716, evaluation target database 71, user database 72, and review database 74 provided in the SNS server 70 Are the product screen acquisition request 211, the review information acquisition unit 213, the reliability calculation unit 214, the product screen generation unit 215, the product screen transmission unit 216, the product database 31 provided in the shopping server 20 of the first embodiment described above. , Corresponding to the user database 32 and the review database 34. Hereinafter, differences from the shopping server 20 will be mainly described.

  The evaluation target database 71 stores information (hereinafter referred to as evaluation target information) about what is subject to user reviews (hereinafter simply referred to as evaluation target). In the third embodiment, the evaluation targets include all items that can be evaluated by the user, such as products, services, events, movies, music, games, books, entertainers, and topics. A configuration example of the evaluation target information is shown in FIG. Evaluation object information includes evaluation object identification information (hereinafter referred to as evaluation object ID), evaluation object name (evaluation object name), evaluation object category (classification), description of evaluation object, and image of evaluation object. URL etc. are included.

  The user database 72 has the same configuration as the user database 32 of the first embodiment. In the third embodiment, the SNS server 70 uses the URL of a Web page (hereinafter referred to as a user screen) for the user to display his / her profile (user information) based on the user ID, and the user ID. It is possible to generate a Web page (hereinafter referred to as a message screen) for sending a message to the user indicated by the user information. An e-mail address may be included.

  The review database 74 has the same configuration as the review database 34 of the first embodiment. However, as shown in FIG. 18, the review object of the third embodiment includes an evaluation target ID instead of the product ID. Shall be included.

  In the first embodiment, the shopping server 20 receives the product screen acquisition request from the client terminal 10 and responds to the product screen. However, in the third embodiment, the SNS server 70 is connected to the client terminal 10. Then, a command for acquiring the user screen (hereinafter referred to as a user screen acquisition request) is received, and the user screen is responded. An example of a user screen 90 generally used in the SNS system is shown in FIG. On the user screen 90, an image display field 901 displays an image representing the user. The URL to the image displayed in the display field 901 can be set in advance as user information, for example. In the friend display column 902, a list of images representing other users registered as user friends is displayed. The user's friend displayed in the display field 902 has, for example, a database for managing the friend's user, and registers the user ID indicating the user's friend user indicated by the user ID in the database in association with the user ID. Can be kept. In the new arrival information display field 903, a message from the SNS system operator to the user is displayed. The latest diary display column 904 displays a list of diaries of other users registered as user friends and diaries of other users commented by the user. The community display field 905 displays a list of communities in which the user is participating.

  The user screen acquisition request receiving unit 711 (corresponding to the recommendation destination specifying unit of the present invention) receives a user screen acquisition request from the client terminal 10. In the user screen acquisition request, a user ID (hereinafter referred to as a login user ID) indicating a user operating the client terminal 10 (hereinafter referred to as a login user, which corresponds to a recommended user of the present invention). It shall be included.

  The review information acquisition unit 713 (corresponding to the maximum evaluation target selection unit and the minimum evaluation target selection unit of the present invention) corresponds to the review information acquisition unit 213 of the first embodiment and corresponds to the login user ID. Information is read from the review database 74.

  The degree of compatibility calculation unit 714 (corresponding to the recommendation candidate selection unit of the present invention) corresponds to the degree of reliability calculation unit 214 of the first embodiment, and indicates the degree of compatibility between the login user and other users ( Hereinafter, the degree of compatibility is simply calculated), and another user (hereinafter referred to as a candidate user) having good compatibility with the login user is selected according to the degree of compatibility.

  FIG. 20 is a diagram showing a flow of processing for calculating the degree of compatibility between each user and the logged-in user. FIG. 21 is a diagram illustrating a configuration example of a list (candidate list) used for candidate user selection processing. In the candidate list, a maximum value count, a minimum value count, and a degree of compatibility are registered in association with the user ID. The maximum value count and the minimum value count are the number of times that the user indicated by the user ID assigns the highest or lowest rating to the same evaluation target as the login user. The degree of compatibility is a numerical value calculated based on the maximum value count and the minimum value count. In the present embodiment, the compatibility degree is obtained by summing the maximum value count and the minimum value count.

  The review information acquisition unit 713 reads out the review information corresponding to the login user ID that has a maximum rating value (in this embodiment, “5”) (S801).

  The compatibility degree calculation unit 714 creates an empty candidate list (S802), and performs the following processing using each of the review information read by the review information acquisition unit 713 as process review information.

  The compatibility degree calculation unit 714 reads from the review database 34 the evaluation target ID included in the process review information and the user ID corresponding to the rating that are different from the login user ID (referred to as temporary candidate ID) (S803). . The compatibility level calculation unit 714 determines whether or not the temporary candidate ID is registered in the candidate list for each of the read temporary candidate IDs (S804), and if registered (S804: YES), the candidate In the list, 1 is added to the maximum value count corresponding to the temporary candidate ID (S805). If no temporary candidate ID is registered (S804: NO), the user ID is set as the temporary candidate ID and the maximum value count is “1”. Then, a record with the lowest value count “0” is added to the candidate list (S806).

  When the compatibility level calculation unit 714 performs the above processing for each process review information, the review information acquisition unit 713 determines the minimum value for rating (“0” in the present embodiment) among the review information corresponding to the login user ID. Are read) (S807). The compatibility degree calculation unit 714 performs the following processing using each of the review information read by the review information acquisition unit 713 as process review information.

  The compatibility level calculation unit 714 reads, as a temporary candidate ID, from the review database 34, the evaluation target ID included in the process review information and the user ID corresponding to the rating, which is different from the login user ID (S808). The compatibility degree calculation unit 714 determines whether or not the temporary candidate ID is registered in the candidate list for each temporary candidate ID (S809), and if it is registered (S809: YES), in the candidate list 1 is added to the minimum count corresponding to the temporary candidate ID (S810), and if the temporary candidate ID is not registered in the candidate list (S809: NO), the user ID is set as the temporary candidate ID and the maximum value count is set to “ A record having “0” and a minimum value count of “1” is added to the candidate list (S811).

  After performing the above process for each process review information, the compatibility degree calculation unit 714 sets the sum of the highest count and the lowest count as the affinity degree for each record in the candidate list (S812).

  As described above, the degree of compatibility between each user and the logged-in user is calculated. The compatibility level calculation unit 714 extracts a predetermined number of user IDs from the candidate list in descending order of the compatibility level, and selects a user indicated by the extracted user ID as a candidate user.

  The user screen generation unit 715 generates a user screen including a list of candidate users. For example, the user screen generation unit 715 reads a predetermined number of user IDs in descending order of compatibility of the candidate list, reads user information corresponding to the read user IDs from the user database 72, and selects candidates based on the read user information. A user screen for displaying a list of users is generated.

  The user screen transmission unit 716 transmits the user screen generated by the user screen generation unit 715 to the client terminal 10. The user screen generation unit 715 and the user screen transmission unit 716 correspond to the recommendation destination presentation unit of the present invention.

  On the user screen 90, a list of candidate users having high compatibility with the login user can be displayed in any of the display columns or in a new display column.

  FIG. 22 is a diagram illustrating an example of a user screen 91 that displays a message indicating that a candidate user exists in the new arrival information display field 903. Only when the candidate user ID is included in the final candidate list, the user screen generation unit 715 displays a message 911 indicating that there is a candidate user such as “There is a person who has a hobby with you” in the display field. 903 is added. Also, the user screen generation unit 715 sets a link to a Web page for displaying a list of candidate users in the message 911.

  FIG. 23 is a diagram illustrating an example of a user screen 92 that displays a list of candidate users in a new display field. In the example of FIG. 23, a new friendly friend display field 921 is provided, and a list 922 of candidate users is displayed in this display field 921. The candidate user list 922 is, for example, a list of images and names representing users.

  FIG. 24 is a diagram showing an example of a user screen 94 that displays a list of candidate users in the friend display field 902. The friend display field 903 displays the image and name of the user set as the friend of the logged-in user, and the user screen generation unit 715 provides a separator after this to display the image and name of the candidate user. A user screen 93 that displays the above is generated. The list of candidate users can be, for example, a list of items such as names included in user information corresponding to candidate user IDs included in the final candidate list.

  As described above, according to the SNS system according to the third embodiment, other users who have similarly given the highest price to the login user who have given the highest rating to the login user are assigned to the login user. Can be recommended. When a user assigns the highest rating, the user often likes the evaluation target, and those who like the same thing often agree with each other, and communication between those who agree Become active. Therefore, according to the SNS system according to the third embodiment, by introducing a candidate user who likes the same thing as the logged-in user, it is possible to create an opportunity for exchange and activate user communication in the SNS.

  In addition, in the SNS system according to the third embodiment, it is also possible to recommend other users who gave the same lowest value to the login user for the evaluation target to which the logged-in user gave the lowest rating. it can. When a user assigns the lowest rating, the user often dislikes the evaluation target, and those who dislike the same object often feel the same as those who like the same object . Therefore, according to the SNS system according to the third embodiment, by introducing candidate users who dislike the same user as the logged-in user, it is possible to create an opportunity for exchange and activate user communication in the SNS.

  In the SNS system according to the third embodiment, since a predetermined number of users can be introduced to the login users in descending order of compatibility, too many users can be introduced even when the number of users is large. Can be recommended, and only users who seem to be better compatible with the logged-in user can be recommended.

  In the third embodiment, the number of times that the logged-in user gave the highest value to the evaluation target that gave the highest value to the rating, and the number of times that the logged-in user gave the same lowest value to the evaluation target that gave the lowest value to the rating, However, the present invention is not limited to this, and only the highest value count may be used as the compatibility degree, or only the lowest value count may be used as the compatibility degree. Further, the compatibility degree may be calculated by weighting each of the highest value count and the lowest value count and then adding them together.

  Further, when selecting a candidate user, a record in which either the maximum value count or the minimum value count is 0 may be excluded.

  In the third embodiment, it is determined whether or not the highest value or the lowest value is assigned to the rating. For example, the highest value and the lowest value may have a range. In this case, it is determined whether or not a rating equal to or higher than the first threshold obtained by subtracting a predetermined value from the maximum value is applied, and whether or not a rating equal to or lower than the second threshold obtained by adding the predetermined value to the minimum value is applied. be able to. For example, in step S803 in FIG. 20, the compatibility degree calculation unit 714 corresponds to the evaluation target ID, the user ID does not match the login user ID, and the rating is equal to or higher than a first threshold (for example, 4). The review information is read out. In step S808, the review information corresponding to the evaluation target ID, the user ID does not match the login user ID, and the rating is equal to or lower than a second threshold (for example, 2) is read. Like that. By providing a range for the maximum and minimum values in this way, for example, it is possible to introduce users who are compatible with each other even when the user gives a conservative rating for a favorite evaluation target. Become. For example, if the rating is based on a 100-point scale, the rating of 100 points or 0 points will be reduced, but it is recommended that the maximum value and the minimum value have a range. It can be avoided that the number of possible users is extremely reduced. Note that when the maximum value and the minimum value have a range, the maximum value count and the minimum value count are not simply incremented, but a larger value is added as the difference from the logged-in user's rating is smaller. Alternatively, a larger value may be added as the value approaches the maximum value or the minimum value.

  In addition, the list of candidate users may be notified to the login user by a message or an e-mail without being displayed on the user screen.

  In addition to the image and name, various items included in the user information can be arbitrarily included in the list of candidate users. Further, the user screen generation unit 715 may display a part of the review information of each candidate user on the user screen. In addition, the user screen generation unit 715 may acquire the content spoken by the candidate user in the community and display it together with the list of candidate users.

  The user screen generation unit 715 may identify a community to which the candidate user belongs, and introduce a predetermined number of communities (hereinafter referred to as recommended communities) in descending order of the number of candidate users to which the user user belongs. . In this case, for example, a message indicating that there is a recommended community that is considered to be compatible with the logged-in user, such as “There is a recommended community”, is displayed in the new information display field 903 of the user screen 91. Alternatively, a new display field may be provided in the user screen 92, and a list of recommended communities may be displayed in the new display field. In the user screen 93, the login user belongs to the community display field 905. A list of recommended communities may be displayed in addition to the list of existing communities.

Although the present embodiment has been described above, the above embodiment is intended to facilitate understanding of the present invention and is not intended to limit the present invention. The present invention can be changed and improved without departing from the gist thereof, and the present invention includes equivalents thereof.
For example, another aspect is a system that outputs information provided by a user for a product, the user ID for specifying the user, the product ID for specifying the product, and the product provided by the user for the product. A provision information storage unit that stores the provision information that is information in association with each other, a purchase history storage unit that stores purchase history information including the user ID and the product ID of the product purchased by the user in the past, and the provision A user ID receiving unit that receives a first user ID that is a user ID that identifies a user who browses information, and a provision information acquisition that reads the provision information corresponding to the product ID that identifies the product from the provision information storage unit And the product I included in the purchase history information including a second user ID that is a user ID corresponding to the read provision information An evaluation value determination unit that determines an evaluation value of the provision information according to the number of items included in the purchase history information including the first user ID, and the evaluation value corresponds to the provision information A provision information output unit that outputs the attached information.
When the first and second users have purchased many of the same products in the past, they tend to purchase similar products in the future. Here, although the 1st user has not purchased about a certain goods, when the 2nd user purchases and provides information, the 1st user has a high possibility of purchasing the goods. For the first user, the information provided by the second user is useful.
According to the product information output system, provided information depends on how much the second user who provided the offer information purchased the same product as the product previously purchased by the first user who browses the offer information. Can be determined and output in association with the provided information. Therefore, the first user can easily determine whether the provided information is useful by himself or herself, which is convenient. Even when the number of pieces of provision information is large, necessary provision information can be easily narrowed down by referring to the evaluation value.
In the product information output system, the provision information output unit may output the provision information in descending order of the evaluation value.
In the product information output system, the evaluation value determination unit is included in the purchase history information including the first user ID among the product IDs included in the purchase history information including the second user ID. The provided information may be ranked according to the number of items provided, and the rank assigned to the provided information may be determined as the evaluation value.
In this case, since the evaluation values are ranked and output, the first user who browses the provided information can more easily grasp the evaluation of the provided information.
In the product information output system, the provision information output unit may output an icon corresponding to the rank of the provision information in association with the provision information.
In this case, since the icon corresponding to the rank is output, the first user can grasp the evaluation of the provided information more easily.
In the product information output system, the provision information output unit is included in the purchase history information including the first user ID among the product IDs included in the purchase history information including the second user ID. Only in a case where the number of items provided exceeds a predetermined number, the fact that the provision information is reliable may be output in association with the provision information.
In this case, since it is output whether or not the information is reliable according to the evaluation value, the first user can more easily identify the necessary provision information.
Further, in the product information output system, the provision information is a score given by the user for the product, and the evaluation value determination unit is configured to use the product ID among the product IDs corresponding to the second user ID. And the purchase history information including the number corresponding to the product ID and the score corresponding to the second user ID, and the purchase history information including the second user ID. The evaluation value may be determined according to the number of items included in the purchase history information including the first user ID among the included product IDs.
In this case, the provision information of the second user can be evaluated depending on whether or not the same product as that of the first user is purchased and the same score as the score given by the first user is given in the past. The first and second users who have given the same score for the same product in the past are expected to give the same score for other products, and are provided by the second user who gives the same score as the first user. The information becomes highly reliable for the first user. Therefore, the first user can determine whether the provided information is reliable by referring to the evaluation value. Moreover, in addition to whether or not the same product has been purchased, by evaluating whether or not the same score is given, the usefulness for the first user can be evaluated more accurately.
In the product information output system, the evaluation value determination unit includes the product ID and the score corresponding to the first user ID among the product IDs corresponding to the second user ID, and the product ID. The absolute value of the difference between the score corresponding to the second user ID and the product ID included in the purchase history information including the second user ID includes the first user ID. The evaluation value may be determined according to the number of items included in the purchase history information.
In this case, the evaluation value can be determined depending on whether the second user has given the same score as the first user in the past. Therefore, the evaluation value can be determined depending on whether or not a closer score is given even if it does not completely match, so that the evaluation value can be determined with high accuracy even when the range that the score can take is wide.
Another aspect is a system for outputting information provided by a user about a product, the user ID for specifying the user, the product ID for specifying the product, and the points given to the product by the user A providing information storage unit that stores the providing information in association with each other, a user ID receiving unit that receives a first user ID that is a user ID that identifies a user who browses the providing information, and the product that specifies the product An offer information acquisition unit that reads output offer information that is the offer information corresponding to the product ID from the offer information storage unit, and the product that corresponds to the second user ID that is the user ID corresponding to the output offer information Among the IDs, provided information corresponding to the product ID and the first user ID, and provided information corresponding to the product ID and the second user ID An evaluation value determination unit that determines an evaluation value of the output provision information according to the number of matches, and a provision information output unit that outputs the evaluation value in association with the output provision information; To do.
In the product information output system, the evaluation value determination unit includes the product ID and the provided information corresponding to the first user ID among the product IDs corresponding to the second user ID, and the product ID. The evaluation value of the output provision information may be determined according to an absolute value of a difference from the provision information corresponding to the second user ID.
Another aspect is a method of outputting information provided by a user about a product, wherein the computer identifies a user ID for identifying the user, a product ID for identifying the product, and the user from the product. Storing the provided information, which is provided information, in association with the memory; storing the purchase ID information including the user ID and the product ID of the product purchased by the user in the past; Receiving a first user ID which is a user ID for identifying a user who browses the provision information, reading the provision information corresponding to the commodity ID for identifying the commodity from the memory, and the read provision The product ID included in the purchase history information including the second user ID that is the user ID corresponding to the information. That is, according to the number of items included in the purchase history information including the first user ID, a step of determining an evaluation value of the provision information, and outputting the evaluation value in association with the provision information Steps are executed.
Another aspect is a method of outputting information provided by a user about a product, wherein a computer identifies a user ID for identifying the user, a product ID for identifying the product, and the user identifies the product. Storing the provided information corresponding to the given score in the memory, receiving a first user ID that is a user ID for identifying a user who browses the provided information, and the computer The step of reading out the provision information for output corresponding to the product ID to be identified from the memory, and the computer corresponding to the second user ID corresponding to the user ID corresponding to the provision information for output Among the product IDs, provided information corresponding to the product ID and the first user ID, the product ID and the second user ID Determining the evaluation value of the output provision information according to the number of matches with the provision information corresponding to the user ID, and the computer outputs the evaluation value in association with the output provision information And the step of performing.
In the product information output method, in the step of determining the evaluation value, the computer provides the product ID and the first user ID among the product IDs corresponding to the second user ID. The evaluation value of the output provision information may be determined according to the absolute value of the difference between the information and the provision information corresponding to the product ID and the second user ID.
Further, another aspect is a program for outputting information provided by a user about a product, the computer including a user ID for identifying the user, a product ID for identifying the product, and the user A step of storing, in a memory, provision information that is information provided for a product; and a step of storing in the memory purchase history information including the user ID and the product ID of a product purchased by the user in the past Receiving a first user ID that is a user ID for identifying a user who browses the provision information, reading the provision information corresponding to the commodity ID for identifying the commodity from the memory, and the reading Before being included in the purchase history information including the second user ID that is the user ID corresponding to the provided information A step of determining an evaluation value of the provision information according to the number of items included in the purchase history information including the first user ID among the product IDs, and associating the evaluation value with the provision information And outputting the output step.
Further, another aspect is a program for outputting information provided by a user about a product, and a user ID for identifying the user, a product ID for identifying the product, and the user A step of associating the provided information, which is the score given to the product, with the memory, a step of receiving a first user ID that is a user ID for specifying a user who browses the provided information, and specifying the product The step of reading out the provision information for output corresponding to the commodity ID from the memory, and the commodity ID corresponding to the second user ID corresponding to the user ID corresponding to the provision information for output, Provided information corresponding to the product ID and the first user ID, and provided information corresponding to the product ID and the second user ID There, depending on the number of a match, to determining the evaluation value of the output informational and outputting in association with the evaluation value in the output informational, and thereby the execution.
In the program, in the step of determining the evaluation value in the computer, provided information corresponding to the product ID and the first user ID among the product IDs corresponding to the second user ID, The evaluation value of the output provision information may be determined according to the absolute value of the difference between the product ID and the provision information corresponding to the second user ID.

It is a figure showing the whole shopping system composition concerning one embodiment of the present invention. It is a figure which shows the hardware constitutions of the shopping server 20 which concerns on one Embodiment of this invention. It is a functional block diagram of shopping server 20 concerning one embodiment of the present invention. It is a figure which shows the structural example of the merchandise information memorize | stored in the merchandise database 31. FIG. It is a figure which shows the structural example of the user information memorize | stored in the user database. It is a figure which shows the structural example of the purchase history information memorize | stored in the purchase history database. It is a figure which shows the structural example of the review information memorize | stored in the review database. It is a flowchart explaining the flow of the production | generation process of the goods screen by the shopping server. It is a figure which shows the flow of the calculation process of reliability. 4 is a diagram illustrating an example of a screen 50 displayed on the client terminal 10. FIG. It is a figure which shows the flow of the calculation process of the reliability in the case of evaluating the review of the other goods by a reviewer. It is a figure which shows an example of the screen 60 used when the object of review is a movie. It is a figure which shows the flow of the calculation process of the reliability which compares the rating of the highest value or the lowest value. It is a figure which shows an example of the screen 61 used when the object of review is a game. It is a figure showing an example of composition of a system at the time of realizing shopping server 20 by three servers. It is a figure which shows the software structure of the SNS server. It is a figure which shows the structural example of evaluation object information. It is a figure which shows the structural example of the review information which concerns on 3rd Embodiment. It is a figure which shows an example of the user screen. It is a figure which shows the flow of the process which calculates the compatibility degree of each user and a login user. It is a figure which shows the structural example of the candidate list | wrist used for a candidate user's selection process. It is a figure which shows an example of the user screen. It is a figure which shows an example of the user screen. It is a figure which shows an example of the user screen.

Explanation of symbols

10 Client Terminal 20 Shopping Server 21 Web Server 22 Commerce Server 23 Community Server 30 Internet 31 Product Database 32 User Database 33 Purchase History Database 34 Review Database 50 Screen 201 CPU
202 Memory 203 Storage Device 204 Communication Interface 205 Input Device 206 Output Device 211 Product Screen Acquisition Request Receiving Unit 212 Product Information Acquisition Unit 213 Review Information Acquisition Unit 214 Reliability Calculation Unit 215 Product Screen Generation Unit 216 Product Screen Transmission Unit 521 Review 522 Review 523 Icon 70 SNS server 71 Evaluation target database 72 User database 73 Review database 711 User screen acquisition request reception unit 713 Review information acquisition unit 714 Compatibility degree calculation unit 715 User screen generation unit 716 User screen transmission unit

Claims (18)

A system for recommending other users to a user,
An evaluation storage unit that stores an evaluation value set by the user for the evaluation target for each of the evaluation target and the user;
A recommendation destination identifying unit that identifies a recommendation destination user who is a user who recommends the other user;
A maximum evaluation target selection unit that selects, from the evaluation storage unit, a maximum evaluation target that is the evaluation target in which the recommended user sets the maximum value that the evaluation value can take;
A minimum evaluation target selection unit that selects, from the evaluation storage unit, a minimum evaluation target that is the evaluation target for which the recommended user has set a minimum value that can be taken by the evaluation value;
Among the other users different from the recommended user who set the maximum value for the maximum evaluation target, select candidate users who are users who set the minimum value for the minimum evaluation target from the evaluation storage unit A recommendation candidate selection unit to
For each user, a user information storage unit that stores user information related to the user;
A recommendation candidate presenting unit that reads out the user information corresponding to the candidate user from the user information storage unit and displays the user information to the recommendation destination user;
A user recommendation system comprising: The user recommendation system according to claim 1,
The maximum evaluation target selection unit selects, as the maximum evaluation target, the evaluation target set by the recommended user as an evaluation value equal to or higher than a threshold obtained by subtracting a predetermined value from the maximum value,
The recommendation candidate selection unit sets, as the candidate user, a user who sets an evaluation value equal to or higher than the threshold value for the maximum evaluation target and who sets the minimum value for the minimum evaluation target among users different from the recommendation destination user. To choose as,
User recommendation system characterized by The user recommendation system according to claim 1,
The minimum evaluation target selection unit selects, as the minimum evaluation target, the evaluation target set by the recommended user as an evaluation value equal to or lower than a threshold obtained by adding a predetermined value to the minimum value,
The recommendation candidate selection unit sets, as the candidate user, a user who sets the maximum value for the maximum evaluation target and sets an evaluation value equal to or less than the threshold for the minimum evaluation target among users different from the recommendation destination user. To choose,
User recommendation system characterized by The user recommendation system according to claim 1,
The maximum evaluation target selection unit selects, as the maximum evaluation target, the evaluation target in which the recommended user sets an evaluation value equal to or higher than a first threshold value obtained by subtracting a first predetermined value from the maximum value,
The minimum evaluation target selection unit selects, as the minimum evaluation target, the evaluation target in which the recommended user sets an evaluation value equal to or lower than a second threshold value obtained by adding the second predetermined value to the minimum value. ,
The recommendation candidate selection unit sets an evaluation value equal to or higher than the first threshold for the maximum evaluation target, and is an evaluation value equal to or lower than the second threshold for the minimum evaluation target among users different from the recommended user. Selecting as a candidate user,
User recommendation system characterized by The user recommendation system according to claim 1,
The recommendation candidate selection unit includes:
Referring to the evaluation storage unit, for each of the maximum evaluation objects, it is determined whether each user has set the maximum value,
Calculating a maximum evaluation match number that is the number of the maximum evaluation targets for which each user has set the maximum value;
Selecting a predetermined number of the users as the candidate users in descending order of the maximum evaluation match number;
User recommendation system characterized by The user recommendation system according to claim 5,
The recommended candidate selection unit further includes:
Referring to the evaluation storage unit, for each of the minimum evaluation objects, determine whether each user has set the minimum value,
Calculating a minimum evaluation match number that is the number of the minimum evaluation targets for which each user has set the minimum value;
The recommendation candidate selection unit includes:
Selecting a predetermined number of the users as the candidate users in descending order of the total value of the maximum evaluation match number and the minimum evaluation match number;
User recommendation system characterized by The user recommendation system according to claim 1,
The maximum evaluation target selection unit selects, as the maximum evaluation target, the evaluation target set by the recommended user as the evaluation value equal to or higher than a first threshold obtained by subtracting a first predetermined value from the maximum value,
The recommendation candidate selection unit includes:
Referring to the evaluation storage unit, for each of the maximum evaluation objects, determine whether each user has set the evaluation value equal to or higher than the first threshold,
Calculating a maximum evaluation coincidence number that is the number of the maximum evaluation objects for which each user sets the evaluation value equal to or greater than the first threshold;
Selecting a predetermined number of the users as the candidate users in descending order of the maximum evaluation match number;
User recommendation system characterized by The user recommendation system according to claim 7,
The recommendation candidate selection unit is configured to determine the evaluation value set by the user and the evaluation set by the recommended user for each of the maximum evaluation targets for which the user sets the evaluation value equal to or greater than the first threshold. Calculating a number obtained by adding a larger value as the difference from the value is smaller as the maximum evaluation match number,
User recommendation system characterized by The user recommendation system according to claim 7,
For each of the maximum evaluation targets for which each user has set the evaluation value equal to or greater than the first threshold, the recommendation candidate selection unit increases a value as the evaluation value set by the user is closer to the maximum value. Calculating the added number as the maximum evaluation match number;
User recommendation system characterized by The user recommendation system according to claim 7,
The minimum evaluation target selection unit selects, as the minimum evaluation target, the evaluation target set by the recommended user with the evaluation value equal to or lower than a second threshold obtained by adding a second predetermined value to the minimum value,
The recommended candidate selection unit further includes:
With reference to the evaluation storage unit, for each of the minimum evaluation targets, it is determined whether each user has set the evaluation value equal to or lower than the second threshold,
Calculating a minimum evaluation coincidence number that is the number of the minimum evaluation objects for which each user sets the evaluation value equal to or less than the second threshold;
The recommendation candidate selection unit includes:
Selecting a predetermined number of the users as the candidate users in descending order of the total value of the maximum evaluation match number and the minimum evaluation match number;
User recommendation system characterized by The user recommendation system according to claim 10,
For each of the minimum evaluation targets for which each user has set the evaluation value equal to or less than the second threshold, the recommendation candidate selection unit and the evaluation value set by the user and the evaluation target user set Calculating a number obtained by adding a larger value as the difference from the value is smaller, as the minimum evaluation matching number,
User recommendation system characterized by The user recommendation system according to claim 10,
For each of the minimum evaluation targets for which each user has set the evaluation value equal to or lower than the second threshold, the recommendation candidate selection unit increases the value as the evaluation value set by the user is closer to the minimum value. Calculating the added number as the minimum evaluation match number;
User recommendation system characterized by The user recommendation system according to claim 1,
Wherein the user information includes the name of the user, and the access information is information for establishing communication between said user is included,
The recommended candidate presentation section, that displays a list of the user information corresponding to the candidate user,
User recommendation system characterized by The user recommendation system according to any one of claims 1 to 13,
The evaluation object is a product,
The evaluation storage unit stores, for each of the product and the user, the evaluation value and provided information provided from the user for the product,
A provision information acquisition unit that reads the provision information from the evaluation storage unit;
A reliability determination unit that determines the reliability of the provided information so that the one corresponding to the candidate user is high;
A provision information output unit for outputting the provision information in association with the reliability;
A user recommendation system, further comprising: 15. The user recommendation system according to any one of claims 1 to 14,
The recommendation candidate presenting unit identifies a community to which the candidate user belongs, and outputs the identified community;
User recommendation system characterized by The user recommendation system according to claim 15,
The recommendation candidate presenting unit outputs a predetermined number of the communities in order of the number of candidate users belonging to the recommendation candidate;
User recommendation system characterized by A method for recommending other users to a user,
Computer
The evaluation value set by the user for the evaluation target is stored in a memory for each of the evaluation target and the user,
For each user, store user information about the user in the memory;
Identify a recommended user that is a user who recommends the other user,
The recommendation target user selects from the memory the maximum evaluation target that is the evaluation target that has set the maximum value that the evaluation value can take,
Select the minimum evaluation target, which is the evaluation target in which the recommended user sets the minimum value that can be taken by the evaluation value, from the memory,
Among the other users different from the recommended user who set the maximum value for the maximum evaluation target, select candidate users who are users who set the minimum value for the minimum evaluation target from the memory,
Reading the user information corresponding to the candidate user from the memory and displaying the user information to the recommended user;
The user recommendation method characterized by this. A program for recommending other users to a user,
On the computer,
Storing the evaluation value set by the user for the evaluation object in a memory for each of the evaluation object and the user;
Storing user information about the user in the memory for each user;
Identifying a recommended user who is a user who recommends the other user;
Selecting from the memory a maximum evaluation target that is the evaluation target for which the recommended user has set a maximum value that can be taken by the evaluation value;
Selecting from the memory a minimum evaluation target that is the evaluation target for which the recommended user has set a minimum value that can be taken by the evaluation value;
A step of selecting, from the memory, a candidate user who is a user who has set the minimum value for the minimum evaluation target among the other users different from the recommended user who has set the maximum value for the maximum evaluation target. When,
Reading the user information corresponding to the candidate user from the memory and displaying the user information to the recommended user;
A program for running

Images