JP5041797B2 - Method, classification device, and classification program for classifying user attributes on network - Google Patents

Description

  The present invention relates to a method, a classification device, and a classification program for classifying user attributes on a network.

  In recent years, with the spread of the Internet to the society, in the Web and other network environments, users not only browse information by Web browsers or other technical means, but also purchase products or share information between those with similar attributes. It has come to perform activities similar to those in the real world, such as creating communities by sending and receiving.

  In the Web and other network environments, it is relatively easy to record activities such as user information browsing, product purchase, and information transmission / reception in the community. Specifically, for example, by accepting a user's member registration on the Web, the server stores the basic attributes of the user, information browsing on the Web of the user who performed the member registration, etc., product purchase The server can automatically collect and store the activity as a log (activity record).

  Therefore, in the Web and other network environments, recommend products and services that are considered to be of interest to users based on registered basic attributes, logs (activity records), etc., and send advertisements. Introducing users who seem to have similar attributes (social networking service; SNS).

  For example, according to the technique described in Patent Document 1, the Web server can analyze the user's interest based on the user's purchase record, and can perform product recommendation (recommendation) based on the analysis result. .

  More specifically, the Web server identifies popular products based on the past purchase records of all users, aggregates the purchase frequency of other products purchased by the user who purchased the popular products, The correlation regarding the purchase between the popular product and the other product is analyzed, and therefore, the other product having a strong correlation in the purchase frequency for each popular product is recommended to the user who has purchased the specific product.

  Further, for example, according to the technique of Non-Patent Document 1, the Web server can make a product recommendation based on a user's purchase record or a product evaluation record (rating).

More specifically, the Web server generates a similar product table, samples the users who have purchased popular products, and purchases the popular products based on the purchase records of the users and the evaluation records (ratings) of the products. A product similar to the product purchased or rated by the user is recommended as a highly correlated product.
US Pat. No. 6,912,505 Amazon. com Recommendations, Item-to-Item Collaborative Filtering, Greg Linden, Brent Smith, and Jeremy York, Amazon. com, January, February 2003, IEEE Computer Society

  However, in the inventions described in Patent Document 1 and Non-Patent Document 1, in making the recommendation, not the user, but first taking an approach that focuses on the product, and the user who purchased the popular product together There are several challenges associated with using “correlation” with purchased products.

  The first problem is generally considered that the correlation can take a value between -1 and 1, but there are many distributions that actually take a value in a more limited range. Specifically, for example, it can be considered that only a value of -0.2 to +0.6 is taken. In such a situation, for example, when it is determined that the user attribute is close when the correlation is +0.7, the degree of duplication of the user attribute cannot be determined using the correlation.

  The second problem is that in the analysis using correlation, if the correlation between certain users has a negative value, that value will be discarded, but if the correlation between certain users has a large negative value, Since users can clearly be said to have dependencies, it can be said that there is a limit to capturing the dependencies by correlation.

  A third problem is that the correlation does not include global information regarding the probability distribution representing the user's attributes. Specifically, even when the degree of overlapping of attributes between certain users is relatively large, the correlation cannot always be detected. In other words, if there is no degree of attribute overlap for a certain user, the correlation is zero so that a correct judgment can be made, but conversely, even if the correlation is zero, the degree of attribute overlap may be relatively large. However, an erroneous determination is made as uncorrelated = irrelevant. In particular, since the correlation depends only on the second-order moment, when a probability distribution of behavior characteristics of a plurality of users is given, information on the tail does not appear in the analysis result using the correlation. As an example, if the probability distribution indicating the frequency with which A and B users purchase a certain product has a tail with a power type and B with a Gauss type for the price of the product, for example, There is a decisive difference. That is, even if B recommends an expensive product, there is no prospect of purchasing it, but A may purchase if he likes the product. However, since the probability distribution as a whole is similar, the correlation cannot give information for determining such a difference.

  The fourth problem is that the correlation can express a linear relationship but cannot express a non-linear relationship. Specifically, when two random variables X and Y have a relationship of Y = aX + b (linear relationship), analysis using correlation can provide meaningful information. For example, if two random variables have a relationship of Y = Sin (X) (a periodically dependent relationship) (for example, the relationship between the number of accesses to the network Y and the time X corresponds to this), the correlation is used. Analyzes that have been made cannot give meaningful information about such periodically dependent relationships.

  Therefore, it is desired to provide a method for classifying user attributes on a network without using correlation.

  Therefore, an object of the present invention is to provide a method, a classification device, and a classification program for classifying user attributes on a network without using a product and focusing on a user and without using correlation.

  In order to achieve the above-mentioned object, the present inventors have conducted research and have completed the present invention. Specifically, the present invention provides the following.

(1) A classification device (classification device 10) classifies a plurality of user attributes,
Receiving user characteristic data comprising at least the basic attribute data or log data of the plurality of users received and stored by the server (server 20) via the communication network (communication network 30);
Mapping the characteristics of the plurality of users to a probability space based on the received user characteristic data;
Calculating a spherical distance between each of the users in the mapped probability space;
Classifying the plurality of users into a plurality of groups close to the spherical distance based on the calculated spherical distance.

  According to such a configuration of the present invention, the classification device accepts user characteristic data including at least the basic attribute data or log data of the plurality of users received and stored by a server via a communication network. Mapping the characteristics of the plurality of users to a probability space based on the received user characteristic data, calculating a spherical distance between each of the users in the mapped probability space, and based on the calculated spherical distance Thus, the plurality of users can be classified into a plurality of groups having a short spherical distance.

  Accordingly, the classification device can classify the plurality of users into a plurality of groups having a close spherical distance based on the calculated spherical distance.

  As for the spherical distance used here, the greater the degree of overlap with respect to the distribution representing the attribute of the user, the closer to zero, and the smaller the distance, the closer to the maximum value. That is, the spherical distance can take a value from 0 to the maximum value, and the possible range is not limited to a range less than that, as in “correlation”, depending on the distribution of user attributes. As described above, according to the principle of the present invention, even when the “correlation” can take only a value close to zero regardless of the magnitude of the dependency, the spherical distance can take all the values. The problem can be solved.

  Note that the value of the spherical distance reflects all of the user characteristic data as input, and negative values are not discarded as in “correlation”. Thus, according to the principle of the present invention, the second problem can be solved.

  In addition, since the spherical distance represents the degree of duplication of user attributes at a global level, it can be said that basically all global moments are involved and comprehensive information is included. Therefore, as in the above example, the spherical distance is calculated based on the probability distribution indicating the frequency with which A and B users purchase a certain product. For example, A is a power type and B is a Gaussian type with respect to the price of the product. In contrast to having a tail, “correlation” cannot contrast the tails well. Thus, according to the principle of the present invention, the third problem can be solved.

  Further, in the concept of distance, the distance is determined regardless of the dependence of random variables representing the usage characteristics of different users. For this reason, the spherical distance can express the overlapping degree of the user's attributes regardless of whether the dependency is linear or not. Therefore, the spherical distance is freed from the restriction that only a linear dependency can be expressed sufficiently as in “correlation”. Thus, according to the principle of the present invention, the fourth problem can be solved.

  (2) The method according to (1), wherein in the step of calculating the spherical distance, the classification device calculates a spherical distance of a grasshopper as the spherical distance.

  According to such a configuration of the present invention, in the step of calculating the spherical distance, the classification device has an operational effect that the spherical distance of the grasshopper can be calculated as the spherical distance.

  Accordingly, the classification device can calculate the spherical distance of the grasshopper as the spherical distance, and classify the users into a plurality of groups having the spherical distance of the grasshopper based on the spherical distance of the grasshopper.

  The spherical distance of the grasshopper used here is infinitely close to zero as the degree of overlap of the distribution representing the attribute of the user is large, and approaches the maximum value of π / 2 as it is small. That is, the spherical distance can take a value of 0 to π / 2, and the possible range is not limited to a range less than that, as in “correlation”, depending on the distribution of user attributes. As described above, according to the principle of the present invention, even when the “correlation” can take only a value close to zero regardless of the magnitude of the dependency, the spherical distance can take all the values. The problem can be solved.

  Moreover, the said classification | category apparatus can solve the above-mentioned 2nd thru | or 4th subject similarly to the case where the above-mentioned spherical distance is used by classifying the said user based on the spherical distance of a grasshopper.

によって計算する（２）に記載の方法。 (3) The classification device calculates a spherical distance of the grasshopper.

(2) The method according to (2).

によって計算することができる、という作用効果を有する。 According to such a configuration of the present invention, the classification device calculates the spherical distance of the grasshopper.

It has the effect that it can be calculated by.

  (4) The method according to (3), wherein the classification device displays the calculated spherical distance of the grasshopper on a display unit (display device 1022) as a map.

  According to such a configuration of the present invention, the classification device has an effect of being able to display the calculated spherical distance of the grasshopper on the display unit as a map.

  Accordingly, the classification device can display the calculated spherical distance of the grasshopper on the display unit as a map.

  As a result, the user of the classification device can easily visually refer to the classification results of the plurality of users.

(5) The classification device further includes a step of calculating a relative distance from the other users around the user based on the calculated spherical distance of the grasshopper.
The method according to (3) or (4), wherein, in the classifying step, the classification device classifies the plurality of users into a plurality of groups having close relative distances based on the calculated relative distance.

  According to such a configuration of the present invention, the classification device calculates a relative distance from the other users around the respective users based on the calculated spherical distance of the grasshopper, and performs the classification. In the step, the classification device has an operational effect that the plurality of users can be classified into a plurality of groups having a close relative distance based on the calculated relative distance.

  Accordingly, the classification device can classify the plurality of users into a plurality of groups having a close relative distance based on the calculated relative distance.

  Here, in the probability space representing the characteristics of the user, the relative distance is calculated individually around each of the users. Therefore, the attribute of the attribute between each other user and the user at the center is determined. The degree of overlap can be expressed.

  Therefore, the classification device can classify users having a large degree of attribute duplication into a plurality of groups based on the relative distance representing the degree of duplication of the user attributes.

によって計算する（５）に記載の方法。 (6) The classification device determines the relative distance.

(5) The method according to (5).

によって計算することができる、という作用効果を有する。
（７） 前記分類装置は、計算した前記相対距離をヒストグラムとして表示部（表示装置１０２２）に表示させる（６）に記載の方法。 According to such a configuration of the present invention, the classification device calculates the relative distance.

It has the effect that it can be calculated by.
(7) The method according to (6), wherein the classification device displays the calculated relative distance on a display unit (display device 1022) as a histogram.

  According to such a configuration of the present invention, the classification device has an effect that the calculated relative distance can be displayed as a histogram on the display unit.

  Thus, the classification device can display the calculated relative distance on the display unit as a histogram.

  As a result, the user of the classification device can easily visually refer to the classification results of the plurality of users.

  (8) The method according to any one of (1) to (5), wherein the classification device further includes a step of receiving a setting of a threshold value when the user is classified in the classification step.

  According to such a configuration of the present invention, the classification device has an effect of being able to accept setting of a threshold value when classifying users in the classification step.

  Thereby, the classification device can accept setting of a threshold value when classifying the user.

  As a result, the user of the classification device can adjust the number of classifications of the user according to the situation by checking the classification result of the user while changing the setting of the threshold.

  (9) The method according to any one of (5) to (7), wherein the classification device further includes a step of accepting a setting of accuracy of aggregation of the relative distance in the classification step.

  According to such a configuration of the present invention, the classification device has an operational effect that it can accept the setting of the accuracy of aggregation of the relative distance in the classification step.

  Thereby, the classification device can accept the setting of the accuracy of aggregation of the relative distance.

  As a result, the user of the classification device can adjust the granularity for classifying the user according to the situation by checking the classification result of the user while adjusting the accuracy of the aggregation.

  (10) A recommendation list for making a recommendation or solicitation for a specific user among the plurality of users classified into the group is classified into a group into which the specific user is classified The method according to any one of (1) to (9), further including a step of generating based on user characteristic data of other users.

  According to such a configuration of the present invention, the classification device includes a recommendation list for making a recommendation or a solicitation to the community for a specific user among the plurality of users classified into the group, The specific user can be generated based on user characteristic data of other users classified into the classified group.

  Accordingly, the classification device can generate a recommendation list for a specific user based on user characteristic data of other users classified into the same group as the specific user.

  As a result, since the classification device performs recommendation or solicitation to the community within the users classified into the same group, high satisfaction of the user can be expected for the recommendation or solicitation to the community.

(11) A classification device for classifying attributes of a plurality of users,
Means for receiving user characteristic data comprising at least the basic attribute data or log data of the plurality of users received and stored by a server via a communication network;
Means for mapping to a probability space containing the characteristics of the plurality of users based on the received user characteristic data;
Means for calculating a spherical distance between each of the users in the mapped probability space;
Classifying apparatus comprising: means for classifying the plurality of users into a plurality of groups close to the spherical distance based on the calculated spherical distance.

  According to such a configuration of the present invention, the classification device accepts user characteristic data including at least the basic attribute data or log data of the plurality of users received and stored by a server via a communication network. , Based on the received user characteristic data, mapping to a probability space including the characteristics of the plurality of users, calculating a spherical distance between the users in the mapped probability space, to the calculated spherical distance Based on this, it is possible to classify the plurality of users into a plurality of groups having a short spherical distance.

  Thus, the same effect as (1) is obtained by operating the classification device.

(12) A program for causing a classification device to classify attributes of a plurality of users, wherein the classification device includes:
Accepting user characteristic data comprising at least the basic attribute data or log data of the plurality of users received and stored by a server via a communication network; and
Mapping the characteristics of the plurality of users to a probability space based on the received user characteristic data;
Calculating a spherical distance between each of the users in the mapped probability space;
Classifying the plurality of users into a plurality of groups close to the spherical distance based on the calculated spherical distance.

  According to such a configuration of the present invention, the classification device accepts user characteristic data including at least the basic attribute data or log data of the plurality of users received and stored by a server via a communication network. , Based on the received user characteristic data, mapping to a probability space including the characteristics of the plurality of users, calculating a spherical distance between the users in the mapped probability space, to the calculated spherical distance Based on this, it is possible to classify the plurality of users into a plurality of groups having a short spherical distance.

  Thus, the same effect as (1) can be obtained by introducing and operating the program in the classification device.

  According to the present invention, the classification device can classify the plurality of users into a plurality of groups having a close spherical distance based on the calculated spherical distance.

  Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a diagram showing an overall configuration of a system according to an example of a preferred embodiment of the present invention. FIG. 2 is a diagram illustrating the configuration of the classification device 10 and the server 20 according to an example of the preferred embodiment of the present invention. FIG. 3 is a flowchart showing user attribute classification processing 1 by the classification device 10 according to an example of the preferred embodiment of the present invention. FIG. 4 is a flowchart showing user attribute classification processing 2 by the classification device 10 according to an example of the preferred embodiment of the present invention. FIG. 5 is a diagram illustrating the spherical distance and relative distance of the user calculated by the classification device 10 according to an example of the preferred embodiment of the present invention. FIG. 6 is a diagram illustrating an example of a relative distance histogram displayed by the classification device 10 according to an example of the preferred embodiment of the present invention.
[System overall configuration]

  FIG. 1 is a diagram showing an overall configuration of a system 1 according to an example of a preferred embodiment of the present invention.

  The classification device 10 can be connected to the server 20 and the user terminal 40 via the communication network 30.

The form of connection between the classification device 10 and the server 20 may be wired or wireless. In the present invention, it is not a requirement that these devices be connected. The server 20 may receive user characteristic data that includes at least the basic attribute data or log data of the plurality of users received from the terminals 40 of the plurality of users via the communication network 30 and stored offline.
[Hardware Configuration of Classification Device 10]

  FIG. 2 is a diagram illustrating an example of a hardware configuration of the classification device 10 according to an example of the preferred embodiment of the present invention described in FIG. The classification device 10 includes a central processing unit (CPU) 1010 (a plurality of CPUs such as a CPU 1012 may be added in a multiprocessor configuration), a bus line 1005, a communication I / F 1040, a main memory 1050, and a BIOS. (Basic Input Output System) 1060, a USB port 1090, an I / O controller 1070, input means such as a keyboard and mouse 1100, and a display device 1022.

  Storage means such as a tape drive 1072, a hard disk 1074, an optical disk drive 1076, and a semiconductor memory 1078 can be connected to the I / O controller 1070.

  The BIOS 1060 stores a boot program executed by the CPU 1010 when the classification device 10 is started up, a program depending on the hardware of the classification device 10, and the like.

  The hard disk 1074 constituting the storage unit 107 stores various programs for the classification device 10 to function as a server and programs for executing the functions of the present invention, and can further configure various databases as necessary.

  As the optical disc drive 1076, for example, a DVD-ROM drive, a CD-ROM drive, a DVD-RAM drive, or a CD-RAM drive can be used. In this case, the optical disk 1077 corresponding to each drive is used. A program or data may be read from the optical disk 1077 by the optical disk drive 1076 and provided to the main memory 1050 or the hard disk 1074 via the I / O controller 1070. Similarly, the tape medium 1071 corresponding to the tape drive 1072 can be used mainly for backup.

  The program provided to the classification device 10 is provided by being stored in a recording medium such as a hard disk 1074, an optical disk 1077, or a memory card. The program may be installed in the classification device 10 and executed by being read from a recording medium via the I / O controller 1070 or downloaded via the communication I / F 1040.

  The aforementioned program may be stored in an internal or external storage medium. Here, as a storage medium constituting the storage unit 107, a magneto-optical recording medium such as an MD or a tape medium can be used in addition to the hard disk 1074, the optical disk 1077, or the memory card. Further, a storage device such as a hard disk 1074 or an optical disk library provided in a server system connected to a dedicated communication line or the Internet may be used as a recording medium, and the program may be provided to the classification device 10 via the communication line.

  Here, the display device 1022 displays a screen for accepting data input to the user or displays a screen of a calculation processing result by the classification device 10. The display device 1022 displays a cathode ray tube display device (CRT), a liquid crystal display device ( LCD).

  Here, the input means accepts input by the user, and may be configured by a keyboard, a mouse 1100, and the like.

  The communication I / F 1040 is a network adapter for enabling the classification device 10 to be connected to a terminal via a dedicated network or a public network. The communication I / F 1040 may include a modem, a cable modem, and an Ethernet (registered trademark) adapter.

In the above example, the classification apparatus 10 has been mainly described. However, the functions described above can also be realized by installing a program in a computer and operating the computer as a server apparatus. Therefore, the functions realized by the server described as an embodiment in the present invention can be realized by executing the above-described method by the computer, or by introducing the above-mentioned program into the computer and executing it. It is.
[Hardware Configuration of Server 20]

Here, the server 20 may have the same configuration as the above-described classification device 10.
[Hardware Configuration of User's Terminal 40]

The user terminal 40 may have the same configuration as the above-described classification device 10.
[User attribute classification process 1]

  The classification device 10 performs user attribute classification processing as shown in FIG.

  First, the control unit 101 receives user characteristic data including at least the basic attribute data or log data of the plurality of users stored by the server 20 received from the terminals 40 of the plurality of users via the communication network 30 and stored. (Step S101).

  Specifically, for example, basic attribute data including the user's gender, age, occupation, field of interest, or the like, data such as purchase history of products / services, evaluation records (ratings) for the products / services, and the like are received.

  Next, the control unit 101 maps the characteristics of the plurality of users to a probability space based on the received user characteristic data (step S102).

  Next, based on the received user characteristic data, the control unit 101 calculates the spherical distance of the batcher between the users in a probability space including the characteristics of the plurality of users (step S103).

  Next, based on the calculated relative distance, the control unit 101 receives from the user a threshold specifying operation and an accuracy adjusting operation for classifying the plurality of users into a plurality of groups having a spherical distance close to the grasshopper. (Step S104).

  Next, based on the calculated spherical distance of the grasshopper, the control unit 101 classifies the plurality of users into a plurality of groups having close relative distances and displays a map (step S105). Specifically, a map as shown in FIG. 5 is displayed. In this example, eight user attributes of users 1 to 8 are mapped to the probability space.

  The spherical distance of the batcher between the users 1 to 5 is a very small value (ε). Similarly, the spherical distance of the batcher between the users 6 and 7 is also a very small value (ε). Furthermore, the spherical distance of the grasshopper between the users 1 to 5 and the users 6 and 7 is X. Furthermore, the spherical distance of the batcher between the users 1 to 5 and the user 8 is √3X / 2, and the spherical distance of the batcher between the users 6 and 7 and the user 8 is X / 2.

  Thus, since the control unit 101 of the classification device 10 displays the user attributes mapped in the probability space, the user can easily visually recognize the degree of such user attribute overlap.

  In FIG. 5, the number of users is eight, which is relatively small for the sake of explanation, but in reality, the control unit 101 is similarly applied to users with a scale of tens of thousands or millions. Can map these user attributes to the probability space and display them.

  Furthermore, in the example of FIG. 5, when it is desired to investigate the degree of overlapping of user attributes among the users 1 to 5 in more detail, the control unit 101 designates a small value as the threshold value or the accuracy of the spherical distance of the grasshopper Can be executed by accepting from the user an operation for finely adjusting.

  Next, although not illustrated, the control unit 101 specifies the recommendation list for making a recommendation or solicitation to the community for a specific user among the users classified into the group as described above. Can be generated based on user characteristic data of other users classified into the classified group.

Next, the control unit 101 repeats steps S104 and S105 described above until an end instruction is received from the user (step S106).
[User attribute classification process 2]

  As an example of a preferred embodiment different from the above-described user attribute classification process 1, the classification device 10 performs a user attribute classification process 2 as illustrated in FIG. 4.

  First, the control unit 101 receives user characteristic data including at least the basic attribute data or log data of the plurality of users stored by the server 20 received from the terminals 40 of the plurality of users via the communication network 30 and stored. (Step S201).

  Specifically, for example, basic attribute data including the user's gender, age, occupation, field of interest, or the like, data such as purchase history of products / services, evaluation records (ratings) for the products / services, and the like are received.

  Next, the control unit 101 maps the characteristics of the plurality of users to a probability space based on the received user characteristic data (step S202).

  Next, based on the received user characteristic data, the control unit 101 calculates the spherical distance of the batcher between the users in a probability space including the characteristics of the plurality of users (step S203).

  Next, based on the calculated spherical distance of the grasshopper, the control unit 101 calculates relative distances from all the other users around the respective users (step S204).

  Next, based on the calculated relative distance, the control unit 101 receives, from the user, a threshold designation operation and an accuracy adjustment operation when classifying the plurality of users into a plurality of groups having a relatively close relative distance (Step S101). S205).

  Next, based on the calculated relative distance, the control unit 101 classifies the plurality of users into a plurality of groups close to the relative distance, and displays a histogram (step S206).

  Next, the control unit 101 repeats step S205 and step S206 described above until an end instruction is received from the user (step S207).

  A series of processes from the above-described calculation of the spherical distance of the grasshopper to the classification of user attributes based on the relative distance will be described with a specific example.

  Consider a case in which the probability space including the characteristics of each user (user 1 to user 8) is as shown in FIG. 5 based on the calculated spherical distance of the grasshopper.

  The spherical distance of the batcher between the group of users 1 to 5 and the group of users 6 and 7 is X, and the spherical distance of the batcher between the group of users 1 to 5 and user 8 is √3X / 2, Assume that the spherical distance of the grasshopper between the group of users 6 and 7 and the user 8 is X / 2.

となる。 Here, when calculating the relative distance,

It becomes.

  The calculation result of the relative distance is represented in the histogram as shown in FIG.

  According to FIG. 5, a total of six histograms of near zero, near 0.09, near 0.16, near 0.18, near 0.3, and near 0.35 can be observed.

  Alternatively, if the scale near 0.09 is further expanded, it is not a single histogram near 0.09 (height 4), but a total of two (height), one near 0.091 and one near 0.094. The histogram of 2) can be observed.

  The number of histograms indicates a unit for classifying the user.

  That is, according to the scale of FIG. 6, the plurality of users can be classified into groups having six common relative distances. Alternatively, using a finer scale, a group around 0.09 can be divided into two groups and divided into two groups.

  As described above, the control unit 101 can automatically classify the plurality of users according to the overlapping degree of the attributes by using the relative distance calculated based on the spherical distance of the grasshopper.

  Furthermore, the fineness of the classification can be adjusted by adjusting the accuracy of summing up the histograms of the relative distances of the plurality of users. That is, as the histogram is aggregated with finer accuracy, the group classification becomes finer, and as the histogram is aggregated with coarser accuracy, the group classification becomes coarser.

  Furthermore, in FIG. 6, if only histograms with a threshold value of 6 or more are aggregated, users can be classified into three groups (near zero, near 0.18, and near 0.35), and the remaining users do not belong to any classification. . However, if the threshold is lowered and aggregation is performed with a histogram of 5 or more, users near 0.16 and users near 0.3 are newly added to the grouping. it can.

  In this way, by adjusting the threshold value, the control unit 101 can adjust the number of groups into which users are classified. That is, the higher the threshold value, the smaller the number of groups, and the lower the threshold value, the larger the number of groups.

  As described above with reference to the histogram with reference to FIG. 6, the control unit 101 actually sets the classification of the plurality of users for automatic data processing by setting the threshold and the accuracy of aggregation. It can be performed.

  As described above, according to the principle of the present invention, it is possible to grasp a rough situation in a short period of time by performing the classification process according to the degree of duplication of user attributes.

  In the above example, the example using the spherical distance of the grasshopper has been described. However, the technical scope of the present invention is not limited to this, and the principle of the present invention can be applied to an embodiment using a general spherical distance. It is.

  As mentioned above, although embodiment of this invention was described, this invention is not restricted to embodiment mentioned above. The effects described in the embodiments of the present invention are only the most preferable effects resulting from the present invention, and the effects of the present invention are limited to those described in the embodiments of the present invention. is not.

It is a figure which shows the whole structure of the system which concerns on this invention. It is a figure which shows the structure of the classification | category apparatus 10 and the server 20 which concern on this invention. It is a flowchart which shows the user attribute classification | category process 1 by the classification device 10 which concerns on this invention. It is a flowchart which shows the user attribute classification | category process 2 by the classification device 10 which concerns on this invention. It is a figure explaining a user's spherical distance and relative distance which classification device 10 concerning the present invention calculates. It is a figure which shows the example of the histogram of the relative distance which the classification device 10 concerning the present invention displays.

Explanation of symbols

1 system 10 classification device 20 server 30 communication network 40 terminal 101 control unit 107 storage unit 1022 display device

Claims (11)

A classification device is a method for classifying attributes of a plurality of users,
Receiving user characteristic data comprising at least the basic attribute data or log data of the plurality of users received and stored by a server via a communication network;
Mapping the characteristics of the plurality of users to a probability space based on the received user characteristic data;
Calculating a spherical distance between each of the users by calculating a spherical distance of a grasshopper in the mapped probability space;
Classifying the plurality of users into a plurality of groups close to the spherical distance based on the calculated spherical distance. The classification device calculates the spherical distance of the grasshopper.

The method of claim 1, which is calculated by: The method according to claim 2 , wherein the classification device displays the calculated spherical distance of the grasshopper on a display unit as a map. The classification apparatus further includes a step of calculating a relative distance from each of the users around the user based on the calculated spherical distance of the grasshopper.
In the step of classification, the classification unit on the basis of the calculated the relative distance A method according to claim 2 or claim 3 for classifying the plurality of users into a plurality of groups close in the relative distance. The classification device calculates the relative distance.

The method according to claim 4, which is calculated by: The method according to claim 5 , wherein the classification device displays the calculated relative distance on a display unit as a histogram. The classifier A method according to any one of claims 1 to 4, further comprising a step of accepting the setting of the threshold of the time of classifying the user in the step of classifying. The method according to any one of claims 4 to 6 , wherein the classification device further includes a step of accepting a setting of accuracy of aggregation of the relative distance in the classification step. Other users classified into the group into which the specific user is classified, with a recommendation list for making a recommendation or solicitation for a specific user among the plurality of users classified into the group the method according to any of claims 1 to step further comprising a generated based on the user characteristic data according to claim 8. A classification device for classifying attributes of a plurality of users,
Means for receiving user characteristic data comprising at least the basic attribute data or log data of the plurality of users received and stored by a server via a communication network;
Means for mapping to a probability space containing the characteristics of the plurality of users based on the received user characteristic data;
Means for calculating a spherical distance between each of the users by calculating a spherical distance of a grasshopper in the mapped probability space;
Classifying apparatus comprising: means for classifying the plurality of users into a plurality of groups close to the spherical distance based on the calculated spherical distance. A program that causes a classification device to classify the attributes of a plurality of users.
Accepting user characteristic data comprising at least the basic attribute data or log data of the plurality of users received and stored by a server via a communication network; and
Mapping the characteristics of the plurality of users to a probability space based on the received user characteristic data;
Calculating a spherical distance between the respective users by calculating a spherical distance of the grasshopper in the mapped probability space;
Classifying the plurality of users into a plurality of groups close to the spherical distance based on the calculated spherical distance.

Images