JP4264884B2 - Server device, communication support method, and introducer selection program - Google Patents

Description

The present invention relates to technology for supporting communication between users, and more particularly to a server device , a communication support method, and an introducer selection program that facilitate communication between users who have similar preferences for items such as content and article information. It is.

  In recent years, with the widespread use of the Internet, communication using means such as e-mail, electronic bulletin boards, and chat has been actively performed. There are also systems for finding new communication partners.

For example, in the following patent documents, a method of introducing other users having a correlation between users by using history data of a website browsed by a user or purchased items and introducing users having similar interests Is disclosed. Since processing is performed using the browsing history and purchase history, it is possible to introduce other users with similar interests without requiring complicated operations from the users.
JP 2002-32492 A JP 2002-24694 A

  In the conventional method, the relationship between users is determined using the number of items commonly used among users.

  For this reason, it is difficult to accurately obtain the relationship between users. For example, when the number of items used in common between user A and user B is the same as the number of items used in common between user A and user C, the conventional method uses user A And the relationship between user B and the relationship between user A and user C are calculated as equal. However, the number of items used usually varies from user to user. In the above example, if the number of items used by the user B is less than the number of items used by the user C, the relationship between the user A and the user B is that of the user A and the user C. It is reasonable to judge that it is higher than relevance. However, in the conventional method, the number of items used by individual users is not considered, and it is difficult to accurately calculate the relationship between users.

  Furthermore, since only users who use items that are common to a certain user can be introduced, there is a problem that the number of people who can be introduced is limited depending on the user.

Therefore, the present invention is to select users with similar interests and preferences with respect to items with high accuracy, and to increase the number of people who can introduce to users who rarely use items that are common to other users. It is an object to provide a server device , a communication support method, and an introducer selection program that can be performed.

In order to solve the above problems, the present invention provides the following apparatus, method and program.
(1) A server device that provides information about an item to a user terminal device via a network,
Usage history storage means for associating and storing user identification information for each user and item identification information for each item used by the user;
Based on the information stored in the usage history storage means, the number of uses for each item of each user is calculated, and the frequency of use for each item of each user is calculated using the calculated number of uses. Usage frequency data calculating means for calculating and generating matrix data having the usage frequency values as matrix elements;
A multivariate analysis method that maximizes the correlation between a user and an item in the matrix data and satisfies the condition that each dimension of the number of dimensions smaller than the number of the used items corresponds to a plurality of the items, Applying to the matrix data, score calculating means for calculating eigenvalues corresponding to the respective dimensions and scores that are numerical values having the number of dimensions for each of the users ;
Using the eigenvalue and the score calculated by the score calculation means, a numerical value obtained by multiplying a difference value between the score of one user and the score of another user and the eigenvalue for each dimension, and The one user and the other user using at least one of the numerical values obtained by multiplying the score of one user, the score of the other user, and the eigenvalue for each dimension. An index calculation unit that calculates an index that represents a difference between the index, and another user having a predetermined number or less in order of increasing value of the index , an introduction that is a user with high preference similarity to the one user and the item a selection unit for selecting a user, the introducer selection means having said user identification information of the one user, the introducer storage unit that stores in association with the user identification information of the introducer,
A server device comprising:
(2) A server device that provides information about an item to a user terminal device via a network,
User attribute storage means for associating and storing user identification information for identifying a user, a principal attribute that is the user's own attribute information, and a partner attribute that is the attribute information of the introduction partner desired by the user When,
Usage history storage means for storing the user identification information in association with item identification information for each item used by the user;
Based on the information stored in the usage history storage means, the number of uses for each item of each user is calculated, and the frequency of use for each item of each user is calculated using the calculated number of uses. Usage frequency data calculating means for calculating and generating matrix data having the usage frequency values as matrix elements;
A multivariate analysis method that maximizes the correlation between a user and an item in the matrix data and satisfies the condition that each dimension of the number of dimensions smaller than the number of the used items corresponds to a plurality of the items, Applying to the matrix data, score calculating means for calculating eigenvalues corresponding to the respective dimensions and scores that are numerical values having the number of dimensions for each of the users ;
Using the eigenvalue and the score calculated by the score calculation means, a numerical value obtained by multiplying a difference value between the score of one user and the score of another user and the eigenvalue for each dimension, and The one user and the other user using at least one of the numerical values obtained by multiplying the score of one user, the score of the other user, and the eigenvalue for each dimension. Based on information stored in the user attribute storage means, and an index calculation unit that calculates an index representing a difference between the index value and another user selected in ascending order of the index value. Other users having the principal attribute that matches the partner attribute of the user are selected as introducers who are users with high preference similarity to the one user and the item by a predetermined number or less. and a selection unit for, Serial and the user identification information of one user, the introducer selecting means and a introducer storage unit that stores in association with the user identification information of the introducer,
A server device comprising:
(3) In addition to the user identification information and the item identification information, the usage history storage means associates and stores usage date information indicating a usage date and time when the user uses the item,
The usage frequency data calculation means calculates the usage frequency value by performing a weighting process that increases as the usage date and time is newer, and generates the matrix data.
The server device according to claim 1 or 2, wherein
(4) The introducer selecting means obtains a first value obtained by multiplying a difference value between the score of the one user and the score of the other user and the eigenvalue for each dimension. And a second calculated value obtained by using a numerical value obtained by multiplying the score of the one user, the score of the other user, and the eigenvalue for each dimension. A value obtained by multiplying and adding each by a predetermined coefficient is calculated as the index .
The server device according to any one of claims 1 to 3, wherein
(5) The introducer selection means is a ranking when the introducer is sorted in ascending order of the user identification information of the one user, the user identification information of the introducer, and the index value. Is stored in the introducer storage unit in association with information indicating
The server device according to any one of claims 1 to 4, wherein the server device is characterized in that
( 6 ) A communication support method between users executed by a server device that provides information about an item to a user terminal device via a network,
A user attribute storage means for storing user identification information for identifying a user in association with contact information for making contact with the user;
A step of storing the user identification information in association with the item identification information for each item used by the user;
Based on the information stored in the usage history storage means, the usage frequency data calculation means calculates the usage count for each item of each user, and uses the calculated usage count for each user. Calculating a usage frequency value for each item, and generating matrix data having the usage frequency value as a matrix element;
The score calculation means maximizes the correlation between the user and the item in the matrix data, and satisfies the condition that each dimension having a number of dimensions smaller than the number of the used items corresponds to a plurality of the items. Applying a variable analysis method to the matrix data to calculate eigenvalues corresponding to the respective dimensions and scores that are numerical values having the number of dimensions for each of the users;
The introducer selection means uses the eigenvalue and the score calculated by the score calculation means to obtain a difference value between the score of one user and the score of another user and the eigenvalue for each dimension. The one user using a numerical value multiplied and at least one of the numerical value obtained by multiplying the score of the one user, the score of the other user, and the eigenvalue for each dimension. An index representing the difference between the user and the other user is calculated , and other users having a predetermined number or less in order of increasing index value are designated as users having a high preference similarity to the one user and the item. Selecting as an introducer, and storing the user identification information of the one user in association with the user identification information of the introducer;
The communication management means reads the user identification information of the other user selected as the introducer for the one user from the introducer selection means, and is associated with the read user identification information. Reading the contact information from the user attribute storage means and transmitting it to the terminal device of the one user;
A communication support method characterized by comprising:
( 7 ) A communication support method between users executed by a server device that provides information about an item to a user terminal device via a network,
The user attribute storage means includes user identification information for identifying a user, a personal attribute that is the user's own attribute information, a partner attribute that is the attribute information of the introduction partner desired by the user, and the use Storing in association with contact information for making contact with the person;
A step of storing the user identification information in association with the item identification information for each item used by the user;
Based on the information stored in the usage history storage means, the usage frequency data calculation means calculates the usage count for each item of each user, and uses the calculated usage count for each user. Calculating a usage frequency value for each item, and generating matrix data having the usage frequency value as a matrix element;
The score calculation means maximizes the correlation between the user and the item in the matrix data, and satisfies the condition that each dimension having a number of dimensions smaller than the number of the used items corresponds to a plurality of the items. Applying a variable analysis method to the matrix data to calculate eigenvalues corresponding to the respective dimensions and scores that are numerical values having the number of dimensions for each of the users;
The introducer selection means uses the eigenvalue and the score calculated by the score calculation means to obtain a difference value between the score of one user and the score of another user and the eigenvalue for each dimension. The one user using a numerical value multiplied and at least one of the numerical value obtained by multiplying the score of the one user, the score of the other user, and the eigenvalue for each dimension. Based on the information stored in the user attribute storage means and the other user selected in ascending order of the value of the index Introducing other users having the identity attribute that matches the counterpart attribute of the one user as users having a high preference similarity to the one user and the item by a predetermined number or less. Choose as a person And storing in association with said user identification information of the one user, and the user identification information of the introducer,
The communication management means reads the user identification information of the other user selected as the introducer for the one user from the introducer selection means, and is associated with the read user identification information. Reading the contact information from the user attribute storage means and transmitting it to the terminal device of the one user;
A communication support method characterized by comprising:
( 8 ) An introducer selection program executed by a computer in a server device that provides information about an item to a user terminal device via a network,
User history storage control means for associating user identification information for each user with item identification information for each item used by the user and storing it in the history storage unit;
Based on the information stored in the history storage unit, the number of uses for each item of each user is calculated, and the use frequency value for each item of each user is calculated using the calculated number of uses. Usage frequency data calculation means for generating matrix data having the usage frequency value as a matrix element;
A multivariate analysis method that maximizes the correlation between a user and an item in the matrix data and satisfies the condition that each dimension of the number of dimensions smaller than the number of the used items corresponds to a plurality of the items, Applying to the matrix data, score calculating means for calculating eigenvalues corresponding to the respective dimensions and scores that are numerical values having the number of dimensions for each of the users ;
Using the eigenvalue and the score calculated by the score calculation means, a numerical value obtained by multiplying a difference value between the score of one user and the score of another user and the eigenvalue for each dimension, and The one user and the other user using at least one of the numerical values obtained by multiplying the score of one user, the score of the other user, and the eigenvalue for each dimension. An index calculation unit that calculates an index that represents a difference between the index, and another user having a predetermined number or less in order of increasing value of the index , an introduction that is a user with high preference similarity to the one user and the item introduction having a selection unit for selecting a user, and the user identification information of the one user, the introducer storage control unit for storing the introducer storage unit in association with the user identification information of the introducer Person selection means,
And introducer selection program to make the computer function.
( 9 ) An introducer selection program executed by a computer in a server device that provides information about an item to a user terminal device via a network,
User identification information for identifying a user, a user attribute that is the user's own attribute information, and a partner attribute that is the attribute information of the introduction partner desired by the user are stored in the attribute storage unit in association with each other Person attribute storage control means,
Usage history storage control means for associating the user identification information with item identification information for each item used by the user and storing it in a history storage unit;
Based on the information stored in the history storage unit, the number of uses for each item of each user is calculated, and the use frequency value for each item of each user is calculated using the calculated number of uses. Usage frequency data calculation means for generating matrix data having the usage frequency value as a matrix element;
A multivariate analysis method that maximizes the correlation between a user and an item in the matrix data and satisfies the condition that each dimension of the number of dimensions smaller than the number of the used items corresponds to a plurality of the items, Applying to the matrix data, score calculating means for calculating eigenvalues corresponding to the respective dimensions and scores that are numerical values having the number of dimensions for each of the users ;
Using the eigenvalue and the score calculated by the score calculation means, a numerical value obtained by multiplying a difference value between the score of one user and the score of another user and the eigenvalue for each dimension, and The one user and the other user using at least one of the numerical values obtained by multiplying the score of one user, the score of the other user, and the eigenvalue for each dimension. An index calculation unit that calculates an index representing a difference between the index value and another user selected in ascending order of the index value , and the one usage based on information stored in the attribute storage unit Selection to select other users having the principal attribute that matches the partner attribute of the person as an introducer who is a user having a high preference similarity to the one user and the item by a predetermined number or less and parts, of the one And the user identification information of the use's and introducer selecting means and a introducer storage control unit for storing the introducer storage unit in association with the user identification information of the introducer,
And introducer selection program to make the computer function.

  As described above, according to the present invention, processing is performed using a user's usage history for an item, so that a service provider does not perform item classification work in advance or a user does not perform evaluation work on an item. It is possible to search for and introduce other users that match the user's interests and preferences. Therefore, according to the present invention, it is possible for a user to easily communicate with a user who has similar interests and tastes with himself / herself without troublesome work.

  Further, according to the present invention, in the process of selecting users with similar preferences, each user is placed in an information space reflecting the preference similarity for the items, not the number of items commonly used among users. A score is used. Therefore, if the present invention is used, it is possible to select a similar user with higher accuracy than in the past.

  Furthermore, in the score calculation, if the information space is configured by coordinate axes related to a plurality of items, introducers including users who do not use items common with the target users are included. Can be introduced.

[First embodiment]
FIG. 1 shows the overall configuration of a communication support system to which a first embodiment of the server device and communication support method of the present invention is applied. As can be seen from FIG. 1, the communication support system is configured by connecting a server 1 that provides information about an item and a plurality of terminal devices 2 used by a user via a network 3 such as the Internet. ing. An item is content composed of text, music, video, etc., and / or goods handled in online shopping.

  The server 1 identifies each user by a transmission / reception unit 11 that controls the network 3, an item ID for identifying each item, an item attribute storage unit 13 that stores attribute data such as the name of each item, and the like. User ID storage unit 14 for storing attributes such as user ID and password, which are user identification information, and user attributes of each user, such as gender, age, introduction partner desired conditions, and items of items used It has a usage history storage unit 15 for recording and storing the ID and the user ID of the user who uses the ID.

  The server 1 also includes a usage frequency data creation unit 16 that creates data representing the usage frequency for each item of each user based on the data in the usage history storage unit 15, and each usage frequency data in the information space. A score calculation unit 17 that calculates a score for arranging the user and each item based on the similarity, and another user having a score that is small in difference from the score of the user who uses the terminal device 2 is selected. An introducer selection unit 18, a communication management unit 12 that performs management so that information can be exchanged between the introducer selected by the introducer selection unit 18 and the user, and the entire server 1 having a time measuring function. And a control unit 19 for controlling.

  In FIG. 1, the units 11 to 19 of the server 1 are illustrated as hardware. However, the units 11 to 19 may be implemented as software as the built-in program processing. As a result, the server 1 can be realized by a personal computer, a workstation, or another computer.

  Further, the processing unit of the server 1 can be realized by being distributed to a plurality of computers.

  The terminal device 2 includes a CPU, a RAM, a ROM, a network control circuit, an input device such as a keyboard and a mouse, and a display device such as a display, and performs a processing operation using a built-in program. As the terminal device 2, a general personal computer can be used.

[Operation of the embodiment]
Next, the operation of the communication support system according to the embodiment having such a configuration will be described.

[User registration operation]
In order to receive a service in the communication support system of the embodiment, the user performs user registration on the server 1 side. This user registration is performed according to the flowchart shown in FIG. 2, and when performing user registration, the user operates the terminal device 2 in step S1, for example, the usage desired by the user. Others such as person ID, desired password, user disclosure information (gender, age group, etc.), desired conditions of the referral partner (sex, age group desired to be introduced, etc.), email or phone number Input user attributes such as contact information to be used occasionally. The user attribute information indicating the user attribute input by the user is transmitted to the server 1 side via the network 3.

  The server 1 receives the user attribute information transmitted from the user via the transmission / reception unit 11 under the control of the control unit 19, and the user ID input by the user is already registered in the user attribute storage unit 14. It is determined whether it has been performed (step S2).

  The user attribute storage unit 14 has a user ID and password, which are user identification information that can uniquely identify a user, in the format shown in FIG. 3, and the gender and age that are user attributes of each user. A user attribute table is provided that includes attributes such as a group, desired conditions of an introduction partner (gender desired for introduction, age group desired for introduction, etc.), contact information, and the like. When the control unit 19 confirms in step S2 that the user ID transmitted from the user is not already stored in the user attribute table, the control unit 19 creates a new entry in the user attribute storage unit 14 in step S3. And the received user attribute information is stored in the user attribute table.

  If the control unit 19 determines in step S2 that the user ID transmitted from the user has already been stored in the user attribute table, the control unit 19 sends the terminal device 2 via the transmission / reception unit 11. Then, a message prompting for input of a new user ID is transmitted (step S4).

  Depending on the usage environment, it is possible for the user to select an introducer without specifying the desired conditions of the referral partner. In this case, it is not requested to input the attribute information of the desired condition of the introduction partner in step S1 or both the desired condition of the introduction partner and the user disclosure information. In step S3, the desired condition attribute of the introduction partner or Both the desired condition attribute of the introduction partner and the user disclosure information are not stored in the user attribute storage unit 14. By not requiring the user to input the desired conditions of the referral partner or the user disclosure information, the user can use the information more easily.

[Item information provision operation]
Next, when the user attribute is registered on the server 1 side as described above, it is possible to receive the item information providing service in the communication support system. This information providing service is performed according to the flowchart shown in FIG. 4, and when receiving the information providing service, in step S11, the user operates the terminal device 2 to obtain the registered user ID and password. Make input. The terminal device 2 transmits the user ID and password input by the user to the server 1.

  The control unit 19 of the server 1 receives the user ID and password via the transmission / reception unit 11, and compares them with the user ID and password registered in the user attribute table of the user attribute storage unit 14 in step S12. To do. The information providing service described below is performed only when a match between the two is detected. If the two do not match, an error code is returned to the terminal device 2 side. As a result, the user notices an input error of the user ID or password, and inputs the correct user ID or password again.

  Next, the control unit 19 of the server 1 detects a match between the user ID and password transmitted from the user and the user ID and password registered in the user attribute table of the user attribute storage unit 14. Then, in step S <b> 13, information related to an item which is content composed of text, music, video, and / or item information handled in online shopping is transmitted to the terminal device 2. In step S <b> 14, when the user selects an item that the user wants to use, the item ID is transmitted from the terminal device 2 to the server 1. Next, in step S15, the item ID of the item designated by the user, the user ID of the user, and the use date and time measured by the control unit 19 of the server 1 are in the format shown in FIG. Stored in the usage history table of one usage history storage unit 15.

  Next, the server 1 transmits the item main body data to the terminal 2, and the user displays and reproduces the item on the terminal 2 in step S16. When the item is an article, this step is a step for sending the article from the business operator to the user.

An item attribute table is formed in the item attribute storage unit 13 in a format as shown in FIG. 6, and attributes such as the name and the storage location of the item body data are stored for each item ID that uniquely identifies the item. ing.
[Introducer data creation operation]
The server 1 creates introducer data using data in the usage history storage means 15 according to the flowchart shown in FIG. In the first embodiment, this operation is performed periodically according to a predetermined schedule, asynchronously with the item information providing operation and the introducer presenting operation. This method is suitable when the number of users and the number of items are large.

(Use frequency data creation operation)
In step S21, the usage frequency data creation unit 16 shown in FIG. 1 creates usage frequency data A in a matrix format as described below. Hereinafter, it is assumed that the number of users stored in the usage history storage unit 15 is M and the number of items is N.

  The usage frequency data creation unit 16 first creates correspondence data between user IDs and user index values i (i = 1 to M) and correspondence data between item IDs and item index values j (j = 1 to N). To do. Then, the number of times the user i (i = 1 to M) has used the item j (j = 1 to N) is calculated. This is because the user ID of user i is U [i], the item ID of item j is C [j], and the user ID is U [i] from the table of the usage history storage unit 15. It is obtained by counting the number of records whose ID is C [j].

  It is assumed that user i has used item j L times, the usage date and time is T [i] [j] [k] (k = 1 to L), and the current date and time is Tc. The unit 16 calculates the usage frequency value a [i] [j] (i = 1 to M, j = 1 to N) for the item j of the user i using the following formula (1).

ここで、この数式（１）に示す関数ｆ（ｘ）は、図８に示すように入力ｘが大きくなるに従って出力が減少する特性を持つ重み関数である。従って、例えばアイテムｊを１日前に利用した場合は、１ヶ月前に同じアイテムｊを利用した場合に比べ、そのアイテムｊの利用頻度が高い値となる。利用頻度データ作成部１６は、このようにして利用頻度ａ[i][j]を要素とするＭ行Ｎ列の行列の利用頻度データＡを作成する。そして、この利用頻度データＡが作成されると、サーバ１はステップＳ２２に進む。

Here, the function f (x) shown in the equation (1) is a weighting function having a characteristic that the output decreases as the input x increases as shown in FIG. Therefore, for example, when the item j is used one day ago, the use frequency of the item j is higher than that when the same item j is used one month ago. The usage frequency data creation unit 16 creates the usage frequency data A of a matrix of M rows and N columns having the usage frequency a [i] [j] as elements. When the usage frequency data A is created, the server 1 proceeds to step S22.

  In this way, if the user records the use date and time when the item is used, and creates the use frequency data using a larger weighting factor as the use date is newer, the latest interest of the user will be included in the introducer selection process later. And introducer selection reflecting the taste.

(Score calculation operation)
Next, in step S22, the score calculation part 17 shown in FIG. 1 calculates the score which arrange | positions each user based on the preference similarity with respect to an item in information space. By placing each user in the information space and expressing the similarity between users using the distance or angle in the information space, it is possible to select similar users with higher accuracy than the conventional method become.

  For example, in the conventional method, regardless of the number of items used by the user, if the number of items commonly used by the user A and the user B, and the user A and the user C is the same, The relevance of user B and user C to person A is equal. By placing each user in the information space, not only the number of items used in common among users, but also the number of other items used by each user is taken into account to determine the relationship between users. be able to.

  Many methods can be applied to the method of configuring the information space. As the simplest method, an information space in which one item corresponds to one coordinate axis can be used. However, with this method, it is difficult to increase the number of introducers for users who have few items in common with other users. Therefore, by using the relationship between multiple items and multiple users, each coordinate axis corresponds to multiple items, so that it is also an introducer to users who have few items in common with other users An information space that can increase the number of times is calculated.

Specifically, the score calculation unit 17 is obtained by applying, for example, quantification type 3, which is one method of multivariate analysis, to the usage frequency data A. The quantification type 3 itself is described in detail, for example, in “Quantification” (Tomoo Hayashi, Asakura Shoten), and is omitted here. ] [q] (i = 1 to M, q = 1 to Q) and eigenvalue λ [q] (q = 1 to Q) are obtained. Here, Q is a predetermined integer that satisfies 1 ≦ Q <min (M, N), and represents the dimension of the information space. The eigenvalue λ [q] has a maximum value of λ [1] in the first dimension, sequentially decreases with the number of dimensions, and has a minimum value of λ [Q] in the Q dimension.
The sample score X [i] [q] is a quantity calculated on the condition that the correlation between the user and the item is maximized in the usage frequency data A. Since items that are frequently used by users are considered items that users prefer, this information space can be said to be a space that reflects the user's preference for items.

  In addition, each coordinate axis does not correspond to one item, but corresponds to a plurality of items, and when there are items with similar properties, users who do not use the same item Can be related.

  For example, when user A uses only item 1 and user B uses only item 2, in the conventional method, it is determined that user A and user B are not related at all. In other words, it is impossible to introduce each other. However, if many other users use both item 1 and item 2, item 1 and item 2 are considered to have similar properties, and user A and user B have some degree of It is reasonable to determine that these are related. In such a case, according to the present invention, the score of the user A and the score of the user B are not irrelevant at all and are close to some extent. For this reason, even users who do not use a common item can find relevance.

  Also, if the distance or angle of any two sample scores X [i1] [q], X [i2] [q] (i1 ≠ i2) are close, the preferences for the two items are similar. When the distance or angle is far, it can be determined that the preference is different.

  In this embodiment, the quantification type 3 is applied to the score calculation. However, other statistical methods that can obtain the same result may be used.

(Introducer selection action)
Next, in step S23, the introducer selection unit 18 shown in FIG. 1 selects an introducer for each user. In this process, the process of step S23 will be described in detail using the flowchart of FIG.

  In step S31, a control variable i representing the user is set to 1.

  In step S32, the control variable p representing the introduction candidate is set to 1.

  In step S33, the buffer memory B for temporarily storing the calculation result is initialized.

  In step S34, it is determined whether or not the control variable i and the control variable p are different. If they are different, the process proceeds to step S35, and if they are the same, the process proceeds to step S37.

  In step S <b> 35, it is determined whether or not the introduction condition (gender, age group, etc.) of the user i matches the attribute (gender, age group, etc.) of the introduction candidate p. If they match, the process proceeds to step S36, and if they do not match, the process proceeds to step S37.

  In step S36, an index D [i] [p] representing the difference between the user i and the introduction candidate p is calculated according to the formula (2), the formula (3), or the formula (4). Equation (2) is the difference in distance between the vectors formed by the score, Equation (3) is the difference in angle between the vectors formed by the score, and Equation (4) is the sum of the distance and angle between the vectors. It is an indicator. In Equation (4), α and β are predetermined constants satisfying α + β = 1. Then, the calculation result D [i] [p] and the control variable p are stored in the buffer memory B in association with each other.

ステップＳ３７においては、紹介候補者を表わす制御変数pを１増やす。

In step S37, the control variable p representing the introduction candidate is incremented by one.

  In step S38, it is determined whether or not the control variable p is equal to or less than the number of users M stored in the usage history storage unit 15. If it is less than or equal to M, the process proceeds to step 34;

  In step S39, the data stored in the buffer memory B is sorted in ascending order of the distance D [i] [p]. Then, a predetermined number R of introduction candidates are selected in ascending order of D [i] [p], and are introduced to the user i.

  In step S40, the user ID of the introducer selected in step S40 is written in the introducer table of the user attribute storage means 14 in correspondence with the user ID of the user i. The introducer table has a data format as shown in FIG. 10, and stores the user ID of the user and the user IDs of up to R introducers in association with each other. When the number of introducers is less than R, a NULL value that means not applicable is set in the remaining fields.

In step S41, the control variable i representing the user is incremented by one. In step S 42, it is determined whether or not the control variable i is equal to or less than the number of users M stored in the usage history storage unit 15. If it is less than or equal to M, the process proceeds to step 32. If it is greater than M, the introducer selection operation is terminated.
[Introducer action]
Users can receive information about the item, use the item, and receive the referral service of other users with similar preferences for the user and the item after completing the initial introducer data creation operation. become.

  The introducer presenting operation is performed according to the flowchart shown in FIG.

  In step S51 of FIG. 11, the user operates the terminal device 2 to input the registered user ID and password. The terminal device 2 transmits the user ID and password input by the user to the server 1.

  The control unit 19 of the server 1 receives the user ID and password via the transmission / reception unit 11 and compares them with the user ID and password registered in the user attribute table of the user attribute storage unit 14 in step S52. To do. The information providing service described below is performed only when a match between the two is detected. If the two do not match, an error code is returned to the terminal device 2 side. As a result, the user notices an input error of the user ID or password, and inputs the correct user ID or password again.

  Next, the control unit 19 of the server 1 detects a match between the user ID and password transmitted from the user and the user ID and password registered in the user attribute table of the user attribute storage unit 14. Then, in step S53, the communication management unit 12 of the server 1 sequentially reads the user IDs from the introducer 1 to the introducer R corresponding to the user from the introducer table of the user attribute storage means 14, and further introduces the introduction. The user attribute table of the user attribute storage means 14 is read using the user ID of the user as a key, and the user attribute information such as the rank from 1 to R, the sex of the introducer, the age group, and the message is read. Transmit to device 2.

  In step S54, the terminal device 2 displays user attributes such as rank, introducer's gender, age group, and contact information on the screen in a format as shown in FIG. A user with a lower rank number indicates a similar preference for an item with the user. The user can make contact with an introducer using an e-mail address or a telephone number as contact information.

  As described above, in the user registration operation, when the introducer does not input personal disclosure information, the gender, age group, etc. of the introducer are not displayed on the screen of the terminal device 2.

The above is the description of the first embodiment of the present invention.
[Second Embodiment]
The second embodiment is different from the first embodiment in [introducer data creation operation] and [introducer presentation operation]. Since the introducer data creation operation is performed every time the user requests introduction of the introducer, the introducer can be presented based on the latest usage information of the item.
[Introducer action]
After the user receives information related to the item and uses the item, the user can receive an introduction service of the user and another user who has similar preferences for the item.

  The introducer presenting operation is performed according to the flowchart shown in FIG.

  In step S61 of FIG. 13, the user operates the terminal device 2 to input the registered user ID and password. The terminal device 2 transmits the user ID and password input by the user to the server 1.

  The control unit 19 of the server 1 receives the user ID and password via the transmission / reception unit 11, and compares them with the user ID and password registered in the user attribute table of the user attribute storage unit 14 in step S62. To do. The information providing service described below is performed only when a match between the two is detected. If the two do not match, an error code is returned to the terminal device 2 side. As a result, the user notices an input error of the user ID or password, and inputs the correct user ID or password again.

  Next, the control unit 19 of the server 1 detects a match between the user ID and password transmitted from the user and the user ID and password registered in the user attribute table of the user attribute storage unit 14. Then, in step S63, the server 1 performs an introducer data creation operation described later.

  Next, in step S64, the communication management unit 12 of the server 1 sequentially reads the user IDs from the introducer 1 to the introducer R corresponding to the user from the introducer table of the user attribute storage means 14, and further introduces the introduction. The user attribute table of the user attribute storage means 14 is read using the user ID of the user as a key, and the user attribute information such as the rank from 1 to R, the sex of the introducer, the age group, and the message is read. Transmit to device 2.

In step S65, the terminal device 2 displays user attributes such as rank, introducer's gender, age group, and contact information on the screen in a format as shown in FIG. A user with a lower rank number indicates a similar preference for an item with the user. The user can make contact with an introducer using an e-mail address or a telephone number as contact information.
[Introducer data creation operation]
Details of the processing step of step S63 described above are shown in the flowchart of FIG.

(Use frequency data creation operation)
Step S71 is the same as step S21 in the first embodiment.

(Score calculation operation)
Next, in step S72, the score calculation unit 17 shown in FIG. 1 calculates the score for arranging each user in the information space based on the preference similarity to the item in the same process as step S22 in the first embodiment. To do.

(Introducer selection action)
Next, in step S73, the introducer selection unit 18 shown in FIG. 1 selects an introducer for each user. In this process, the process of step S73 will be described in detail using the flowchart of FIG.

  In step S81, using the correspondence data between the user ID and the user index value i created by the usage frequency data creation unit 16, the index of the user is obtained from the user ID of the user requesting the introducer. A value is obtained and this value is set in a control variable i representing the user.

  In step S82, the control variable p representing the introduction candidate is set to 1.

  In step S83, the buffer memory B for temporarily storing the calculation result is initialized.

  In step S84, it is determined whether or not the control variable i and the control variable p are different. If they are different, the process proceeds to step S85, and if they are the same, the process proceeds to step S87.

  In step S85, it is determined whether or not the introduction request condition (gender, age group, etc.) of user i matches the attribute (gender, age group, etc.) of introduction candidate p. If they match, the process proceeds to step S86, and if they do not match, the process proceeds to step S87.

  In step S86, an index D [i] [p] representing a difference between the user i and the introduction candidate p is calculated according to the formula (2), the formula (3), or the formula (4). Equation (2) is a difference in distance between vectors formed by the score, Equation (3) is an angle difference between vectors formed by the score, and Equation (4) is a comprehensive index using the distance and angle. There are, but you can use any. Then, D [i] [p] as a calculation result is associated with the control variable p and stored in the buffer memory B.

  In step S87, the control variable p representing the introduction candidate is incremented by one.

  In step S88, it is determined whether or not the control variable p is equal to or less than the number of users M stored in the usage history storage unit 15. If it is less than or equal to M, the process proceeds to step S84, and if greater than M, the process proceeds to step S89.

  In step S89, the data stored in the buffer memory B is sorted in ascending order of the distance D [i] [p]. Then, a predetermined number R of introduction candidates are selected in ascending order of D [i] [p], and are introduced to the user i.

  In step S90, the user ID of the introducer selected in step S89 is written in the introducer table of the user attribute storage means 14 in association with the user ID of the user i. The introducer table has a data format as shown in FIG. 10, and stores the user ID of the user and the user IDs of up to R introducers in association with each other. When the number of introducers is less than R, a NULL value that means not applicable is set in the remaining fields. Then, the introducer selection operation for the user requesting the introducer is terminated.

  The above is the description of the second embodiment of the present invention.

  In the first and second embodiments, the introducer selection operation for each user is performed using the score calculated based on the usage frequency information and the desired condition of the introduction partner as the user attribute information. However, depending on the usage environment, it may be possible to perform an introducer selection operation for each user using only the score calculated based on the usage frequency information without requesting input of attribute information such as desired conditions of the introduction partner. That is, without performing step S35 shown in FIG. 9 and step S85 shown in FIG. 15, another user having a small difference from the score of the target user may be selected as an introduction partner. In this case, since each user does not need to input attribute information such as desired conditions of the referral partner, each user does not need to disclose his detailed personal information, and the psychological resistance is reduced, and more It is possible to gather many users.

  Finally, the description of the above embodiment is an example of the present invention. For this reason, the present invention is not limited to this embodiment, and various modifications can be made without departing from the technical idea according to the present invention, even if other than this embodiment. Of course.

It is a figure which shows the structure of the whole communication assistance system to which the 1st and 2nd Example of this invention is applied. It is a flowchart which shows the processing step of the user registration operation | movement in the communication assistance system to which the 1st and 2nd Example of this invention is applied. It is a figure which shows the data format of the user attribute table of the user attribute storage part 14 in the communication assistance system to which the 1st and 2nd Example of this invention is applied. It is a flowchart which shows the process step of the item information provision operation | movement in the communication assistance system to which the 1st and 2nd Example of this invention is applied. It is a figure which shows the data format of the usage history table of the usage history storage part 15 in the communication assistance system to which the 1st and 2nd Example of this invention is applied. It is a figure which shows the data format of the item attribute table of the item attribute storage part 13 in the communication assistance system to which the 1st and 2nd Example of this invention is applied. It is a flowchart which shows the processing step of the introducer data creation operation | movement in the communication assistance system to which the 1st Example of this invention is applied. It is a figure which shows the characteristic of the function f (x) used by the utilization frequency data preparation part 16 in the communication assistance system to which the 1st and 2nd Example of this invention is applied. It is a flowchart which shows the processing step of the introducer selection operation | movement of the introducer selection part 18 in the communication assistance system to which the 1st Example of this invention is applied. It is a figure which shows the data format of the introducer table of the user attribute storage means 14 in the communication assistance system to which the 1st and 2nd Example of this invention is applied. It is a flowchart which shows the process step of the introducer presentation operation | movement in the communication assistance system to which the 1st Example of this invention is applied. It is a figure which shows an example of the introducer display screen of the terminal device 2 in the communication assistance system to which the 1st and 2nd Example of this invention is applied. It is a flowchart which shows the process step of the introducer presentation operation | movement in the communication assistance system to which the 2nd Example of this invention is applied. It is a flowchart which shows the processing step of the introducer data creation operation | movement in the communication assistance system to which the 2nd Example of this invention is applied. It is a flowchart which shows the processing step of the introducer selection operation | movement of the introducer selection part 18 in the communication assistance system to which the 2nd Example of this invention is applied.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Server 2 Terminal device 3 Network 11 Server 1 transmission / reception part 12 Server 1 communication management part 13 Server 1 item attribute storage part 14 Server 1 user attribute storage part 15 Server 1 usage history storage part 16 Use of server 1 Frequency data creation unit 17 Server 1 score calculation unit 18 Server 1 introducer selection unit 19 Server 1 control unit

Claims (9)

A server device that provides information about an item to a user terminal device via a network,
Usage history storage means for associating and storing user identification information for each user and item identification information for each item used by the user;
Based on the information stored in the usage history storage means, the number of uses for each item of each user is calculated, and the frequency of use for each item of each user is calculated using the calculated number of uses. Usage frequency data calculating means for calculating and generating matrix data having the usage frequency values as matrix elements;
A multivariate analysis method that maximizes the correlation between a user and an item in the matrix data and satisfies the condition that each dimension of the number of dimensions smaller than the number of the used items corresponds to a plurality of the items, Applying to the matrix data, score calculating means for calculating eigenvalues corresponding to the respective dimensions and scores that are numerical values having the number of dimensions for each of the users ;
Using the eigenvalue and the score calculated by the score calculation means, a numerical value obtained by multiplying a difference value between the score of one user and the score of another user and the eigenvalue for each dimension, and The one user and the other user using at least one of the numerical values obtained by multiplying the score of one user, the score of the other user, and the eigenvalue for each dimension. An index calculation unit that calculates an index that represents a difference between the index, and another user having a predetermined number or less in order of increasing value of the index , an introduction that is a user with high preference similarity to the one user and the item a selection unit for selecting a user, the introducer selection means having said user identification information of the one user, the introducer storage unit that stores in association with the user identification information of the introducer,
A server device comprising: A server device that provides information about an item to a user terminal device via a network,
User attribute storage means for associating and storing user identification information for identifying a user, a principal attribute that is the user's own attribute information, and a partner attribute that is the attribute information of the introduction partner desired by the user When,
Usage history storage means for storing the user identification information in association with item identification information for each item used by the user;
Based on the information stored in the usage history storage means, the number of uses for each item of each user is calculated, and the frequency of use for each item of each user is calculated using the calculated number of uses. Usage frequency data calculating means for calculating and generating matrix data having the usage frequency values as matrix elements;
A multivariate analysis method that maximizes the correlation between a user and an item in the matrix data and satisfies the condition that each dimension of the number of dimensions smaller than the number of the used items corresponds to a plurality of the items, Applying to the matrix data, score calculating means for calculating eigenvalues corresponding to the respective dimensions and scores that are numerical values having the number of dimensions for each of the users ;
Using the eigenvalue and the score calculated by the score calculation means, a numerical value obtained by multiplying a difference value between the score of one user and the score of another user and the eigenvalue for each dimension, and The one user and the other user using at least one of the numerical values obtained by multiplying the score of one user, the score of the other user, and the eigenvalue for each dimension. Based on information stored in the user attribute storage means, and an index calculation unit that calculates an index representing a difference between the index value and another user selected in ascending order of the index value. Other users having the principal attribute that matches the partner attribute of the user are selected as introducers who are users with high preference similarity to the one user and the item by a predetermined number or less. and a selection unit for, Serial and the user identification information of one user, the introducer selecting means and a introducer storage unit that stores in association with the user identification information of the introducer,
A server device comprising: In addition to the user identification information and the item identification information, the usage history storage means associates and stores usage date information indicating the usage date and time when the user used the item,
The usage frequency data calculation means calculates the usage frequency value by performing a weighting process that increases as the usage date and time is newer, and generates the matrix data.
The server device according to claim 1 or 2, wherein The introducer selection means is a first calculated value obtained by using a numerical value obtained by multiplying a difference value between the score of the one user and the score of the other user and the eigenvalue for each dimension. And two calculated values obtained by using a numerical value obtained by multiplying the score of the one user, the score of the other user, and the eigenvalue for each dimension. A value obtained by multiplying and adding the coefficient is calculated as the index .
The server device according to any one of claims 1 to 3, wherein The introducer selecting means is information indicating the user identification information of the one user, the user identification information of the introducer, and the ranking when the introducers are sorted in ascending order of the index values. Is stored in the introducer storage unit in association with
The server device according to any one of claims 1 to 4, wherein the server device is characterized in that A communication support method between users executed by a server device that provides information about an item to a user terminal device via a network,
A user attribute storage means for storing user identification information for identifying a user in association with contact information for making contact with the user;
A step of storing the user identification information in association with the item identification information for each item used by the user;
Based on the information stored in the usage history storage means, the usage frequency data calculation means calculates the usage count for each item of each user, and uses the calculated usage count for each user. Calculating a usage frequency value for each item, and generating matrix data having the usage frequency value as a matrix element;
The score calculation means maximizes the correlation between the user and the item in the matrix data, and satisfies the condition that each dimension having a number of dimensions smaller than the number of the used items corresponds to a plurality of the items. Applying a variable analysis method to the matrix data to calculate eigenvalues corresponding to the respective dimensions and scores that are numerical values having the number of dimensions for each of the users;
The introducer selection means uses the eigenvalue and the score calculated by the score calculation means to obtain a difference value between the score of one user and the score of another user and the eigenvalue for each dimension. The one user using a numerical value multiplied and at least one of the numerical value obtained by multiplying the score of the one user, the score of the other user, and the eigenvalue for each dimension. An index representing the difference between the user and the other user is calculated , and other users having a predetermined number or less in order of increasing index value are designated as users having a high preference similarity to the one user and the item. Selecting as an introducer, and storing the user identification information of the one user in association with the user identification information of the introducer;
The communication management means reads the user identification information of the other user selected as the introducer for the one user from the introducer selection means, and is associated with the read user identification information. Reading the contact information from the user attribute storage means and transmitting it to the terminal device of the one user;
A communication support method characterized by comprising: A communication support method between users executed by a server device that provides information about an item to a user terminal device via a network,
The user attribute storage means includes user identification information for identifying a user, a personal attribute that is the user's own attribute information, a partner attribute that is the attribute information of the introduction partner desired by the user, and the use Storing in association with contact information for making contact with the person;
A step of storing the user identification information in association with the item identification information for each item used by the user;
Based on the information stored in the usage history storage means, the usage frequency data calculation means calculates the usage count for each item of each user, and uses the calculated usage count for each user. Calculating a usage frequency value for each item, and generating matrix data having the usage frequency value as a matrix element;
The score calculation means maximizes the correlation between the user and the item in the matrix data, and satisfies the condition that each dimension having a number of dimensions smaller than the number of the used items corresponds to a plurality of the items. Applying a variable analysis method to the matrix data to calculate eigenvalues corresponding to the respective dimensions and scores that are numerical values having the number of dimensions for each of the users;
The introducer selection means uses the eigenvalue and the score calculated by the score calculation means to obtain a difference value between the score of one user and the score of another user and the eigenvalue for each dimension. The one user using a numerical value multiplied and at least one of the numerical value obtained by multiplying the score of the one user, the score of the other user, and the eigenvalue for each dimension. Based on the information stored in the user attribute storage means and the other user selected in ascending order of the value of the index Introducing other users having the identity attribute that matches the counterpart attribute of the one user as users having a high preference similarity to the one user and the item by a predetermined number or less. Choose as a person And storing in association with said user identification information of the one user, and the user identification information of the introducer,
The communication management means reads the user identification information of the other user selected as the introducer for the one user from the introducer selection means, and is associated with the read user identification information. Reading the contact information from the user attribute storage means and transmitting it to the terminal device of the one user;
A communication support method characterized by comprising: An introducer selection program executed by a computer in a server device that provides information about an item to a user terminal device via a network,
User history storage control means for associating user identification information for each user with item identification information for each item used by the user and storing it in the history storage unit;
Based on the information stored in the history storage unit, the number of uses for each item of each user is calculated, and the use frequency value for each item of each user is calculated using the calculated number of uses. Usage frequency data calculation means for generating matrix data having the usage frequency value as a matrix element;
A multivariate analysis method that maximizes the correlation between a user and an item in the matrix data and satisfies the condition that each dimension of the number of dimensions smaller than the number of the used items corresponds to a plurality of the items, Applying to the matrix data, score calculating means for calculating eigenvalues corresponding to the respective dimensions and scores that are numerical values having the number of dimensions for each of the users ;
Using the eigenvalue and the score calculated by the score calculation means, a numerical value obtained by multiplying a difference value between the score of one user and the score of another user and the eigenvalue for each dimension, and The one user and the other user using at least one of the numerical values obtained by multiplying the score of one user, the score of the other user, and the eigenvalue for each dimension. An index calculation unit that calculates an index that represents a difference between the index, and another user having a predetermined number or less in order of increasing value of the index , an introduction that is a user with high preference similarity to the one user and the item introduction having a selection unit for selecting a user, and the user identification information of the one user, the introducer storage control unit for storing the introducer storage unit in association with the user identification information of the introducer Person selection means,
And introducer selection program to make the computer function. An introducer selection program executed by a computer in a server device that provides information about an item to a user terminal device via a network,
User identification information for identifying a user, a user attribute that is the user's own attribute information, and a partner attribute that is the attribute information of the introduction partner desired by the user are stored in the attribute storage unit in association with each other Person attribute storage control means,
Usage history storage control means for associating the user identification information with item identification information for each item used by the user and storing it in a history storage unit;
Based on the information stored in the history storage unit, the number of uses for each item of each user is calculated, and the use frequency value for each item of each user is calculated using the calculated number of uses. Usage frequency data calculation means for generating matrix data having the usage frequency value as a matrix element;
A multivariate analysis method that maximizes the correlation between a user and an item in the matrix data and satisfies the condition that each dimension of the number of dimensions smaller than the number of the used items corresponds to a plurality of the items, Applying to the matrix data, score calculating means for calculating eigenvalues corresponding to the respective dimensions and scores that are numerical values having the number of dimensions for each of the users ;
Using the eigenvalue and the score calculated by the score calculation means, a numerical value obtained by multiplying a difference value between the score of one user and the score of another user and the eigenvalue for each dimension, and The one user and the other user using at least one of the numerical values obtained by multiplying the score of one user, the score of the other user, and the eigenvalue for each dimension. An index calculation unit that calculates an index representing a difference between the index value and another user selected in ascending order of the index value , and the one usage based on information stored in the attribute storage unit Selection to select other users having the principal attribute that matches the partner attribute of the person as an introducer who is a user having a high preference similarity to the one user and the item by a predetermined number or less and parts, of the one And the user identification information of the use's and introducer selecting means and a introducer storage control unit for storing the introducer storage unit in association with the user identification information of the introducer,
And introducer selection program to make the computer function.

Images