JP2000315212A - Method, system for sorting information and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an information sorting system capable of performing information sorting on which a user's taste is surely reflected. SOLUTION: Correlation values to express degrees of similarity in tastes between users are held in a correlation value file F4 of each user. And plural pieces of object information to be presented, for example, some attributes such as, manufacturers, functions and prices are set for each piece of merchandise information and the correlation value of the user is updated at any time based on an evaluated result for each attribute by the user by a correlation value managing part 13. The information to be presented to which the same attribute as the one which is relatively highly evaluated by an object person to whom the information is presented is sorted among pieces of the object information to be presented which is applicable to the tastes of other users based on a correlation value after update by taking the opportunity of a request from the object person to whom the information is presented as one of the users by an information managing part 12 and an information presenting part 14.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an information selection technology capable of selecting only information belonging to a range of a user's preference from a variety of information. In this specification, the expression "selection of information" is used as a concept including information retrieval.

[0002]

2. Description of the Related Art As a method for selecting information based on the idea that information of interest to a person having a similar preference to oneself is also information of interest to oneself, for example, US Pat. A collaborative filtering method disclosed in 4996642 is known.

[0003] In this collaborative filtering method, a person who provides information to be used for preference determination (hereinafter referred to as a "user") is evaluated about the information, and based on the evaluation result, the user is evaluated. The preference similarity (correlation value) is calculated. Then, a user who has requested information presentation or a user who presents selected information (hereinafter, referred to as a user)
"Information-presentation target person") and other users whose tastes are similar, that is, the correlation value of the evaluation result is high, and the information presentation target person is determined from the evaluation result by the specified user and the correlation value obtained earlier. The method employs a method of estimating the evaluation of the non-evaluated information and presenting the estimated information with a high evaluation to the information presenting person.

[0004]

SUMMARY OF THE INVENTION In collaborative filtering, the degree of similarity of one user's preference to the preferences of another user is used as a reference. Not been. Therefore, when various kinds of information are mixed, the information presentation target person and the overall preference are somewhat similar, but when looking at a certain genre, there may be a case where they are not very similar. In this case, the filtering did not work. In other words, if the information presenter's tastes vary widely, the correlation value based on the evaluation result of the part where the taste matches is used in the same manner for information that does not match the taste in the process of estimating the evaluation result. As a result, the latter information is also treated as being consistent.

[0005] For example, in the case where the information-presenting person is interested in skiing and tennis and gives high evaluation to the information relating to them, but is not interested in soccer, if the information-presenting person is not interested in skiing and tennis, as in the case of the information presenting person It is assumed that a plurality of users who have given a high evaluation to tennis are identified as having similar preferences to the information presentation target person. It is fully conceivable that the information provided by each of the plurality of specified users has a high evaluation, including soccer information. In this case, if an attempt is made to estimate the evaluation of the information presenter from the correlation value between the information presenter and other users who are similar to the other preferences and the evaluation results of those users, the information of the soccer is also used for skiing and tennis. A high estimation result can be obtained in the same manner as the information of (1). However, since the information presenting person is not interested in the information on soccer, even if it is presented to the information presenting person, it does not match the preference of the information presenting person.

In order to prevent such a situation, it is conceivable to classify the information to be presented in advance for each genre, and to execute collaborative filtering only within the genre. However, since not all users evaluate information from the same viewpoint, the accuracy of information selection is likely to be significantly reduced depending on the classification criteria. For example, one user may give a high rating to a product of a specific maker, while another user may give a high rating to a function or related product information regardless of the manufacturer. As described above, it is not possible to determine from which viewpoint the genre classification is appropriate.

[0007] Therefore, the present invention provides an improved information selection method capable of improving the accuracy of information selection even when a wide variety of information to be presented is mixed and the evaluation criteria of a plurality of users are indeterminate. The task is to provide a method.
Another object of the present invention is to provide an information sorting system to which the above information sorting method is applied. Another object of the present invention is to provide a recording medium suitable for realizing an information selection method using a computer device.

[0008]

According to the information selection method of the present invention for solving the above-mentioned problems, a correlation value indicating the similarity of preference between a plurality of users is held for each user, and the user is one of the users. In a method of selecting presentation target information that matches the preference of another user whose correlation value with the information presentation target is relatively high, several attributes are assigned to each of a plurality of presentation target information Then, the correlation value is updated as needed based on the evaluation result of each user with respect to this attribute, and among the presentation target information that matches the preference of the other user specified based on the updated correlation value, And selecting the presentation target information to which the same attribute as the attribute that the information presentation target person relatively evaluates is given. As a result, information that reliably reflects the preference is presented to the information presentation target person.

This information selection method can be implemented, for example, when a computer reads a predetermined program code and executes the following processing. This program code is recorded on a computer-readable recording medium. (1) Several attributes are assigned to each of a plurality of pieces of presentation target information, and an information evaluation degree indicating a level of evaluation of the presentation pair information by each user and an evaluation of the user for each attribute are given. A process of maintaining the attribute evaluation degree indicating the height for all the users and updating the attribute evaluation degree of the user with respect to the attribute given to the presentation target information evaluated relatively high, ( 2) a process of specifying an attribute having a relatively high evaluation of the information presenter based on the attribute evaluation degree of the information presenter after the update; and (3) a process of identifying an attribute that is the same as the specified attribute. A process of identifying all other users who have given high evaluations and updating the correlation value between the identified user and the information presentation target to a higher value; (4) a correlation value after updating Based on the user's information evaluation degree exceeding the predetermined value The information presenter estimates the information evaluation level for the presentation information not yet evaluated by the information presenter, and the information presenter evaluates the information presenter relatively higher among the presentation information corresponding to the estimated information evaluation degree. A process of presenting information including the same attribute to the information presenting person.
This processing is performed by, for example, the above-described collaborative filtering.

[0010] An information selection system according to the present invention for solving the above-mentioned other problems provides a number of attributes to information to be presented and an information evaluation degree indicating a level of evaluation of the information by a user. Information management means for managing, for each user, an attribute evaluation level indicating a level of evaluation of each of the attributes by a user, and a correlation value indicating a similarity of preference with another user; Identify all other users who have given relatively high evaluations to the same attribute as the attribute that the information presentation target person, one of the users, has relatively high evaluation based on their respective attribute evaluation levels And a correlation value management means for updating the correlation value between the identified user and the information presentation target, and information presentation based on the information evaluation degree of the user whose updated correlation value exceeds a predetermined value. Presentation target that the subject has not yet evaluated The information evaluation target for the information is estimated, and only the information including the same attribute as the attribute that the information presenter is relatively highly evaluated from among the presentation target information corresponding to the estimated information evaluation degree is the information presenting target. And information presenting means for presenting to a user.

[0011] Preferably, the information management means is configured to update an attribute evaluation degree of an attribute added to each piece of presentation target information at any time according to a value of the information evaluation degree of the presentation target information. .

[0012] In a more specific embodiment, the information to be presented is merchandise information of a virtual shopping mall formed by a predetermined computer system. The correlation value of each user is updated based on the history of the degree of interest indicated by the user, and the information presenting unit extracts a part of the handled product information based on the updated correlation value. It is configured to be presented to the information presentation target person as recommended product information.

[0013]

Next, an embodiment of an information selection system to which the present invention is applied will be described. FIG. 1 is a configuration diagram of an information selection system 10 according to the present embodiment. The information selection system 10 is realized by a computer device (or system) having a communication mechanism and a storage device in addition to an input device such as a keyboard, an output device such as a display, and a CPU of the computer device having a predetermined program. Functional blocks of an input / output control unit 11, an information management unit 12, a correlation value management unit 13, and an information presentation unit 14 formed by reading and executing a code, an information database DB built in a storage device, and attributes. File F1, information evaluation file F2, attribute evaluation file F3,
It comprises a correlation value file F4.

The presentation target information database DB stores presentation target information (including evaluated information / unevaluated information) to be presented to the user.

The attribute file F1 is a file in which the correspondence between the identifier of the presentation target information in the presentation target information database DB and one or more attributes corresponding to the presentation target information is recorded. If the information to be presented is information on a product, the attribute is a product genre, its brand name or maker, function, price, etc., and can be set arbitrarily according to the type, use, nature of the market, etc. of the product. it can.

The information evaluation file F2 is a file in which information evaluation data as evaluation results for the information to be presented is stored for each user (linked to the identification data of the user). This information evaluation data is an evaluation given to the individual presentation target information by the user according to his / her taste, for example, 5
Numerical information such as “5 points”, “4 points”,... If a user gives a rating of 5 for “camera” and 4 for “TV”,
Is stored in the information evaluation file F2 as "5 points", and the information evaluation data for "TV" is stored as "4 points". Hereinafter, the evaluation result may be referred to as “evaluation point” as being represented by the above-described numerical information.

The attribute evaluation file F3 is a file in which attribute evaluation data as evaluation results for individual attributes is stored for each user. The evaluation result is the same as that used for evaluating the selection target information. The attribute evaluation data is updated as needed according to the value of the information evaluation data of the user, that is, the degree of interest in the presentation target information. When one attribute is given by a plurality of pieces of presentation target information, the evaluation result of the user for each of the plurality of pieces of presentation target information is reflected in the evaluation result of the attribute. For example, if the user has a strong interest in the high-quality “camera” of company X, that is, gives a high evaluation, the attributes of the “camera” are “manufacturer: company X” and “function: high function”. , A predetermined score is added to the attribute evaluation data of “price: expensive” as needed. Similarly, when the user gives a high evaluation to the high performance “TV” of Company Y, “Maker: Y” which is the attribute of “TV”
A predetermined score is added to the attribute evaluation data of “company” and “function: high function”. At this time, the score of the attribute “function: high function” is added twice as a result of reflecting the evaluation of both “the camera of company X” and “the television of the company Y”. The attribute will be more highly evaluated.

The correlation value file F4 is a file for storing, for each user, a correlation value indicating the similarity of preference between users.

Next, each of the functional blocks 11 to 14 will be described. The input / output control unit 11 receives an information presenting request and other information input from an information presenting person who is one of the users through an input device or a communication mechanism,
The required information is presented to the information presentation target person.

The information management unit 12 manages the information to be presented and various types of information relating to the preferences of each user. The information management unit 12 accumulates and reads out information from and to the presentation object information database DB, and sets the attributes of each information to be presented , Various files F1 to F of the above-described information evaluation data and attribute evaluation data
3 is configured by one or a plurality of modules that perform processing such as recording, updating, and reading the information.

The correlation value management unit 13 calculates the correlation value for each user by using a known calculation formula (for example, Pearson product moment correlation coefficient calculation formula), and stores the calculation result in the correlation value file F4. A module for storing and updating the correlation value of each user at any time based on the information evaluation data and attribute evaluation data of each user. Specifically, the other users B to D who have given a high evaluation to the same attribute as a certain user A is highly evaluated based on the respective attribute evaluation degrees, and the specified user A process of updating at any time so that the correlation value between B to D and the user A increases.

[0022] The information presenting section 14 is provided for the information to be presented which has not yet been evaluated by the information presenter based on the information evaluation degree of the user whose latest correlation value with the information presenter exceeds a predetermined value. Estimates the information evaluation level and presents only the information including the same attribute as the attribute evaluated by the information presenter relatively high to the information presenter from the information to be presented corresponding to the estimated information evaluation level. This is a module that performs processing for performing

The functional blocks (modules constituting the functional blocks) 11 to 14 and the presentation target information database D
B and the program code for forming each of the files F1 to F4 on a computer device are usually recorded on a fixed recording medium such as a hard disk or a semiconductor memory, and read and executed by the CPU of the computer device as needed. It is supposed to be. However, the present invention can be implemented as long as the functional blocks 11 to 14 and the like are formed in the computer device during operation, so that the recording form may be arbitrary. For example, it is recorded on a portable recording medium such as a CD-ROM, DVD, or FD that can be separated from the computer device, or recorded on a program server connected to a private network, read at the time of use, and installed on the fixed recording medium. It may be provided for execution at any time. The above functional blocks 11 to 14
In addition to the case where the above is realized by the program code alone, the above may be appropriately realized by cooperation with an operating system mounted on a computer device.

Next, the operation of the information selection system 10 will be described with reference to FIG. Here, it is assumed that the attribute information file F1 has already been created. It is also assumed that the information evaluation data of each of the plurality of users (including the information presentation target person) is stored in the information evaluation file F2. In addition, when the information presentation target requests information presentation, the selection and presentation of the presentation target information are performed. However, this premise is for convenience of explanation, and such precondition is not always indispensable.

In FIG. 2, the information management unit 12 refers to the information evaluation file F2 and the attribute evaluation file F3, and the evaluation of each user is relatively high (for example, exceeding the highest evaluation point or exceeding a predetermined evaluation point). 3.) The presentation target information and the attribute corresponding thereto are specified based on the association information in the attribute information file F1 (step S101).
Then, the attribute evaluation data for the specified attribute is updated. Specifically, the evaluation points are added (step S10
2). This is repeated for all applicable users (step S103: No).

When the attribute evaluation data of each user is updated (step S103: Yes), the correlation value management unit 13
The correlation value of the preference between the users is calculated, and the correlation value of each user in the correlation value file F4 is updated based on the calculation result (step S104). Upon input of the information presentation request from the information presenting person (step S105: Yes), the latest attribute evaluation data of the information presenting person in the attribute evaluation file F3 is referred to, and the attribute having a relatively high evaluation among them is referred to. The evaluation data and its attributes are specified (step S1
06) Then, another user having the same high evaluation for the specified attribute is specified (step S107). This process is performed by the information management unit 12. Whether an attribute has a high evaluation is determined by, for example, whether or not an evaluation point having a certain value or more is given, or whether or not the ratio of the entire attribute to the evaluation points is a certain value or more.

Thereafter, the correlation value management unit 13 updates the correlation value file F4 by adding the α point to the specified correlation value of each user (step S108). If the α point is large, only the presentation target information having a high evaluation point for the same attribute is extracted. Conversely, if the α point is small, the evaluation point for the attribute is not much reflected. The value of the α point is a tuning item in the system.

The information presentation unit 14 refers to the correlation value file F4 and the information evaluation file F2, and presents information by using the information evaluation data of the user whose updated correlation value exceeds a predetermined value and the correlation value. The evaluation is performed on the presentation target information that has not yet been evaluated by the subject (step S109), and some presentation target information with a high estimated evaluation is specified (step S110). Then, these presentation target information,
The information presentation target is narrowed down to those corresponding to the highly evaluated attributes (step S111), and the presentation target information is presented to the information presentation target through the input / output control unit 11 (step S112).

By limiting the range of the information to be presented in this way, even in the case where various types of information to be presented are mixed, the information selection and the information that reliably reflect the preference of the information presented person can be achieved. Presentation is possible. Therefore, by applying the method of the present embodiment to collaborative filtering, the accuracy can be significantly improved as compared with the conventional method.

[0030]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The information selection system 10 according to the present embodiment can present information that matches each user's preference to a plurality of users in a situation where various kinds of information to be presented are mixed. . Here, as an example, an example in which the present invention is applied to a virtual shopping mall (shopping mall) realized on the Internet will be described.

In this case, the information selection system 10 stores the above-mentioned presentation target information database DB, each file F1 to F4, and each function in one Web server (a plurality of servers when performing distributed processing) connected to the Internet. Blocks 11 to 14 will be formed. In the presentation target information database DB, various kinds of handled product information in the shopping mall are stored. In each of the files F1 to F4,
Information evaluation data, attribute evaluation data, and correlation values of registered users are stored.

The information management unit 12 is configured to automatically update the correlation value of each user based on the history of the degree of interest indicated by a plurality of users with respect to various kinds of product information handled in the shopping mall. More specifically, the degree of interest of the user, that is, the degree of evaluation of the information, is quantified based on the reference history of each user and the purchase history of the product for the products offered at each store in the shopping mall. , And generates and updates the above-described information evaluation data and attribute evaluation data, and updates evaluation values based on these. The information presenting unit 14 presents the handled product information selected based on the updated correlation value to the information presentation target person as recommended product information, for example.

In actual operation of the shopping mall, in addition to the information selection system 10, an operation management server 20 for exchanging information with a store server (agent) that constructs the mall, and a computer provided in a credit company A payment server 30 or the like for performing a payment process when a user purchases a product in cooperation with the system is provided.

As in this embodiment, by applying the information selection technology of the present invention to a shopping mall, information on the preferences of a plurality of users for the products handled is automatically collected, and based on the collected information. Since recommended product information narrowed down to the range of preference can be presented to each user, it is possible to increase the user's awareness of using the shopping mall.

[0035]

As is apparent from the above description, in the present invention, several attributes are added to the information to be presented,
Information selection is performed after reflecting the evaluation result of the assigned attribute on the correlation value of the preference between users, so that various types of information to be presented are mixed, and the evaluation criteria of multiple users are undefined. Even in such a case, a unique effect that the accuracy of information selection can be improved can be obtained.

[Brief description of the drawings]

FIG. 1 is a functional block diagram showing an embodiment of an information selection system according to the present invention.

FIG. 2 is a procedure explanatory diagram showing an example of an operation (information sorting method) of the information sorting system of the embodiment.

FIG. 3 is an explanatory diagram showing one embodiment of the present invention.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Information selection system 11 Input / output control part 12 Information management part 13 Correlation value management part 14 Information presentation part F1 Attribute information file F2 Information evaluation file F3 Attribute evaluation file F4 Correlation value file 20 Operation management server 30 Payment server

Claims (10)

[Claims] 1. A correlation value indicating a similarity of preference between a plurality of users is stored for each user, and the correlation value between the user and an information presenting person who is one of the users is relatively high. In the method of selecting presentation target information that matches the user's preference, several attributes are assigned to each of the plurality of presentation target information, and the correlation is performed based on the evaluation result of each user with respect to this attribute. While updating the value at any time, the same as the attribute that the information presenter evaluates relatively higher among the presenter information that matches the preference of the other user specified based on the updated correlation value An information selection method, which comprises selecting presentation target information to which an attribute is assigned. 2. The information selection method according to claim 1, wherein the evaluation result of the user with respect to the attribute assigned to the presentation target information highly evaluated by the user is cumulatively increased. 3. When one attribute is given by a plurality of pieces of presentation target information, the evaluation result of the user for each of the plurality of pieces of presentation target information is reflected in the evaluation result of the attribute. The information selection method according to claim 1. 4. A correlation value indicating a similarity of a preference between a plurality of users is stored for each user, and the correlation value between the information presentation target and one of the users is relatively high. In the method of selecting the presentation target information that matches the user's preference, several attributes are given to each of the plurality of presentation target information, and the height of the evaluation of the presentation pair information by each user is increased. The information evaluation level and the attribute evaluation level indicating the level of the user's evaluation for each attribute are held for all users, and the information evaluation level for the attribute assigned to the relatively highly evaluated presentation target information is retained. A process of updating the attribute evaluation level of the user to be higher; a process of specifying an attribute having a relatively high evaluation of the information presenter based on the updated attribute evaluation level of the information presenter; the identified attribute Relatively high for the same attribute as A process of identifying all other users who are giving evaluations and updating the correlation value between the identified user and the information presentation target to be higher, and using the updated correlation value exceeding a predetermined value. Based on the information evaluation level of the presenter, the information presenter estimates the information evaluation level for the presentation target information that the information presenter has not yet evaluated, and selects the information presenter from the presentation target information corresponding to the estimated information evaluation level. An information selection method comprising: executing, on a computer device, a process of presenting information including the same attribute as an attribute relatively highly evaluated to the information presenting person. 5. The information selection method according to claim 4, wherein the information that has not been evaluated is selected by collaborative filtering. 6. A method for assigning several attributes to information to be presented, an information evaluation level indicating a level of evaluation of the information by a user, and a level of evaluation of each of the attributes by a user. Information management means for managing, for each user, an attribute evaluation degree indicating a similarity, and a correlation value indicating a similarity of preference with another user, and an information presenting person who is one of the users is relatively All other users who gave a relatively high evaluation to the same attribute as the highly evaluated attribute are specified based on each attribute evaluation level, and the identified user and the information presentation target Correlation value management means for updating the correlation value between, and information on presentation target information which has not yet been evaluated by the information presenter based on the information evaluation degree of the user whose updated correlation value exceeds a predetermined value. Estimate the degree of evaluation and evaluate the estimated information Information presenting means for presenting only information including the same attribute as the attribute that the information presenter is relatively highly evaluated from the presenter information corresponding to the information presenter to the information presenter. Sorting system. 7. The information management unit is configured to update an attribute evaluation level of an attribute assigned to each piece of presentation target information according to a value of the information evaluation degree of the presentation target information. The information selection system according to claim 6, characterized in that: 8. The information to be presented is merchandise information of a virtual shopping mall formed by a predetermined computer system, and the information management means includes a degree of interest indicated by a plurality of users with respect to various merchandise information. The information presenting unit is configured to update the correlation value of each user based on the history of the information, and the information presenting unit presents a part of the handled product information as recommended product information based on the updated correlation value. The information selection system according to claim 6, wherein the information is presented to a target person. 9. The information management means is configured to quantify a degree of interest of the user based on a reference history and a purchase history of each user with respect to a product handled in the virtual shopping mall. The information sorting system according to claim 8, wherein 10. A correlation value indicating a similarity of a preference between a plurality of users is stored for each user, and the correlation value between the information presentation target and one of the users is relatively high. A computer device having a function of selecting presentation target information that matches the user's preference, by assigning several attributes to each of a plurality of presentation target information, The information evaluation level indicating the evaluation level and the attribute evaluation level indicating the evaluation level of the user for each attribute are held for all users, and are given to the relatively highly evaluated presentation target information. A process of updating the attribute evaluation level of the user with respect to the attribute that is higher, a process of specifying an attribute whose evaluation of the information presenter is relatively high based on the updated attribute evaluation level of the information presenter , The same attribute as the identified attribute A process of identifying all other users who have given a relatively high evaluation to the gender, and updating the correlation value between the identified user and the information presentation target to be higher, after the update Based on the information evaluation level of the user whose correlation value exceeds a predetermined value, the information evaluation target estimates the information evaluation level for the presentation target information that has not yet been evaluated, and the presentation target information corresponding to the estimated information evaluation level Computer readable program code for executing a process of presenting information including the same attribute as the information presentation target person to the information presentation target person to the information presentation target person. recoding media.