JP7239628B2 - Information processing device, information processing method and information processing program - Google Patents

Description

The present invention relates to an information processing device, an information processing method, and an information processing program for evaluating scores for user attributes.

In recent years, there has been known an information processing apparatus that integrates user attributes estimated by various services on the Internet and manages them so that they can be shared among various services (see, for example, Patent Document 1). This information processing device generates estimated attribute information for each user to which scores have been assigned for various attributes, and provides a user list of users to which scores above a predetermined score have been assigned to the target attributes. provided to operators who operate

Japanese Patent Application Laid-Open No. 2020-35068

However, with the technique disclosed in the above-mentioned Patent Document 1, it is difficult for the business operator to effectively utilize the estimated attribute information because the method of evaluating the score for the attribute is unclear.
For example, if a business operator requests a user list related to the attribute "I have a dog", how much score should be set as a threshold for the user list to determine whether users who actually have a dog in the user list It is not possible to ascertain how likely it is to be included.
For this reason, the business needs to take the time and effort to sample the action history of each user and grasp the relationship between the score for the attribute and the user's action tendency.

An object of the present invention is to provide an information processing device, an information processing method, and an information processing program that enable evaluation of scores for attributes.

The information processing apparatus of the present invention includes a history acquisition unit that acquires action histories of a plurality of users on a network; a high-ranking user group setting unit for setting a high-ranking user group by selecting a first predetermined number of users in descending order of score; a target user group setting unit that sets a target user group by selecting a predetermined number of users; an evaluation information generation unit that calculates an estimated probability that the target attribute is correct in the user group, and generates score evaluation information including the second predetermined number and the estimated probability.

In the present invention, the estimated probability that the target attribute is correct in the target user group is calculated by using the action history of the high-ranking user group as the evaluation criterion. By checking the score evaluation information including the number of users in the target user group (second predetermined number) and the estimated probability, the business operator can select how many users in descending order of the score to determine the target attribute. It is possible to grasp the degree of probability that users who answer correctly are included, and to evaluate the score for the target attribute.

1 is a schematic diagram showing a schematic configuration of an information processing system according to an embodiment of the present invention; FIG. 2 is a block diagram showing a schematic configuration of the information processing apparatus of the embodiment; FIG. 4 is a flowchart for explaining an information processing method in the information processing apparatus of the embodiment; FIG. 4 is a diagram illustrating an analysis graph generated in the embodiment;

An embodiment according to the present invention will be described below with reference to the drawings.
[Overall Configuration of Information Processing System]
FIG. 1 is a schematic diagram showing a schematic configuration of an information processing system 1 of this embodiment.
As shown in FIG. 1, an information processing system 1 of the present embodiment includes an information processing device 10, a plurality of service providing servers 20 that provide various services, and a business operator terminal 30 owned by a business operator that manages services. and a user terminal 40 owned by a user who uses the service, and these are communicably connected via a network such as the Internet.

In this information processing system 1 , the service providing server 20 estimates user attributes based on the user's action history and the like while providing services to the user terminal 40 . The information processing apparatus 10 integrates user attributes estimated by each service providing server 20 to generate estimated attribute information for each user, and manages the information so that it can be shared among various services.
In addition, the information processing device 10 provides score evaluation information obtained by evaluating the score in the estimated attribute information in response to a request from the business operator terminal 30 or the like. By confirming the score evaluation information provided to the operator terminal 30, the operator can effectively utilize the estimated attribute information.
Note that the attributes described here are the user's interest, interest, lifestyle, life stage, consumption tendency, and the like.

[Information processing device 10]
The information processing apparatus 10 is configured by a general computer, and as shown in FIG. 2, includes components such as a communication unit 11, a storage unit 12, a processor 13, etc. that constitute a computer.
Note that the number of computers constituting the information processing apparatus 10 is not particularly limited. In the present embodiment, an example in which the information processing apparatus 10 is configured by one computer is shown for simplification of explanation, but a cloud server constructed by connecting a plurality of computers via a network is used as the information processing apparatus 10. good too.

The communication unit 11 is connected to the Internet and communicates with each device such as the service providing server 20, the business terminal 30, and the user terminal 40 via the Internet.
The storage unit 12 records various information and information processing programs for controlling the information processing apparatus 10 .

The storage unit 12 also includes an action history information storage unit 121, an estimated attribute information storage unit 122, and the like.
In this embodiment, an example in which the action history information storage unit 121 and the estimated attribute information storage unit 122 are provided in the storage unit 12 of the information processing device 10 is shown. Such information may be recorded in another connected data server or cloud storage.

The action history information storage unit 121 stores user identification information (user ID) and action history information for each user.
The behavior history information is history information about user behavior acquired through the Internet. For example, the user action history information includes search history information in search processing performed by the user, browsing history information of content browsed by the user, product trading history information in which the user bought and sold products, and the user's electronic payment method at a physical store. including payment history information using User behavior history information may also include the results of questionnaires given to users via the Internet.
Note that in the present embodiment, search history information is used for the process of generating score evaluation information. This search history information includes items such as a search query and search date and time.

The estimated attribute information storage unit 122 stores user identification information (user ID) and estimated attribute information for each user.
The estimated attribute information is information related to attributes estimated based on the user's action history. For example, estimated attribute information includes one or more attributes and a score assigned to each attribute. Attributes are managed by, for example, attribute identification information (attribute ID).

The processor 13 is composed of an arithmetic circuit such as a CPU and a recording circuit such as a RAM. The processor 13 expands the information processing program recorded in the storage unit 12 into the RAM, and executes various processes in cooperation with the program expanded into the RAM.
By reading and executing the information processing program, the processor 13, as shown in FIG. It functions as an information generation unit 135, a graph generation unit 136, an annotation information generation unit 137, a provision unit 138, and the like.

The history acquisition unit 131 acquires the user's action history from the user terminal 40, the service providing server 20, or the like, and stores it in the action history information storage unit 121. FIG.
The estimated attribute information generation unit 132 generates estimated attribute information for each user based on user attributes and the like acquired from each service providing server 20, and stores the estimated attribute information in the estimated attribute information storage unit 122. FIG. As for the method of generating the estimated attribute information, the method described in the conventional document (for example, Japanese Unexamined Patent Application Publication No. 2020-35068) can be referred to, so detailed description is omitted here.

The high-ranking user group setting unit 133 sets a high-ranking user group related to the target attribute. This high-ranking user group is a user group that serves as a reference for generating score evaluation information corresponding to the target attribute.
The target user group setting unit 134 sets one or more target user groups Cx regarding target attributes. This target user group is a user group used as a comparison target for the top user group in order to generate score evaluation information corresponding to the target attribute.
The evaluation information generation unit 135 generates score evaluation information based on each search history of the top user group S and the target user group Cx. The score evaluation information includes the number of users in the target user group Cx (second predetermined number) and an estimated probability that the target attribute is correct in the target user group Cx (hereinafter referred to as the estimated correctness probability).

The graph generator 136 generates an analysis graph based on score evaluation information.
The annotation information generation unit 137 corresponds to the estimated user number calculation unit and the threshold value calculation unit of the present invention, and generates annotation information (for example, an estimated number of users and a threshold to be described later) of the analysis graph.
The providing unit 138 provides the business terminal 30 with analysis graphs and annotation information.

[Service providing server 20]
The service providing server 20 is a server device that provides services when accessed from the user terminal 40 . For example, the service providing server 20 provides web search service, shopping service, auction service, content providing service, news service, weather forecast service, etc. as services.
The service providing server 20 estimates the attributes of the user based on the user's action history (for example, search history, purchase history, browsing history, etc.) using the service, and provides the information processing device 10 with the attribute.
In the present embodiment, the information processing device 10 and the service providing server 20 are configured separately, but the information processing device 10 may replace some or all of the functions of any of the service providing servers 20. good.

[Business operator terminal 30]
The operator terminal 30 is a terminal device used by an operator who operates a service, and is configured by, for example, a smart phone, a tablet terminal, a personal computer, or the like.
Although illustration of a specific configuration of the business operator terminal 30 is omitted, the business operator terminal 30 has a basic configuration of a general computer. That is, the operator terminal 30 includes an input operation unit that receives operations from an operator, a display that displays image information, a recording device that records various information, and an arithmetic circuit (such as a CPU) that performs arithmetic processing on various information.
The operator terminal 30 accesses the site provided by the information processing device 10 and requests various information from the information processing device 10 by performing various request operations. Further, the business terminal 30 displays various information provided from the information processing device 10 on a display or the like.

[User terminal 40]
The user terminal 40 is a computer managed by a user, and is composed of, for example, a smart phone, a tablet terminal, a personal computer, or the like.
Although illustration of a specific configuration of the user terminal 40 is omitted, the user terminal 40 has a basic configuration of a general computer. That is, the user terminal 40 includes an input operation unit that receives operations from an operator, a display that displays image information, a recording device that records various types of information, and an arithmetic circuit (such as a CPU) that performs arithmetic processing on various types of information.

[How to provide score evaluation information]
Next, a processing procedure for providing the score evaluation information by the information processing device 10 will be described with reference to the flowchart shown in FIG. It should be noted that the history acquisition unit 131 acquires the user's search history at an arbitrary timing before starting the following processing procedure, and stores it in the action history information storage unit 121 (history acquisition step).

First, the operator terminal 30 accesses a predetermined site provided by the information processing device 10 according to the operator's operation. Then, a request for score evaluation information is transmitted to the information processing apparatus 10 by performing an operation of designating an attribute to be evaluated (target attribute). This request includes, for example, the attribute ID corresponding to the target attribute and the business ID corresponding to the business terminal 30 .

When the information processing device 10 receives a request from the operator terminal 30 (step S1; YES), the upper user group setting unit 133 refers to the estimated attribute information storage unit 122 based on the attribute ID included in the request, A high-ranking user group S corresponding to the target attribute is set (step S2; high-ranking user group setting step).
Specifically, the top user group setting unit 133 refers to the estimated attribute information storage unit 122 and sets the top user group S by selecting a first predetermined number of users in descending order of the score for the target attribute. Here, the first predetermined number is, for example, a number that is a predetermined percentage (for example, several percent) of the number of users given a score greater than 0 for the target attribute. Alternatively, the first predetermined number may be a predetermined number, or may be the number of users given a score equal to or higher than a predetermined threshold.

Next, the target user group setting unit 134 refers to the estimated attribute information storage unit 122 based on the attribute ID included in the request, and selects multiple (n) target user groups Cx (C0, C1) corresponding to the target attribute. , C2 . . . Cn) are set (step S3; target user group setting step).
Specifically, the target user group setting unit 134 sets the target user group Cx by referring to the estimated attribute information storage unit 122 and selecting a second predetermined number of users in descending order of score for the target attribute.
Here, the second predetermined number is a number equal to or greater than the first predetermined number, and is a number that differs among the target user groups Cx.
For example, in the present embodiment, the number of users in the target user group C0 (second predetermined number) is the same as the number of users in the top user group S (first predetermined number), and the target user group C0 is substantially It is equal to the top user group S.
In addition, the number of users (second predetermined number) of the target user group C1, C2, . It is a number set to increase.
Also, the number of users in the target user group Cn is the number of users given a score greater than 0 for the target attribute.

Next, the evaluation information generation unit 135 refers to the action history information storage unit 121, and calculates the similarity of the search history between the upper user group S and the target user group Cx for each target user group Cx (step S4).
Specifically, the evaluation information generation unit 135 calculates a query vector corresponding to the high-ranking user group S based on the search history information of each user belonging to the high-ranking user group S. Also, the evaluation information generation unit 135 calculates a query vector corresponding to the target user group Cx based on the search history information of each user belonging to the target user group Cx. Here, as a method of calculating a query vector, a method such as TF-IDF can be used to assign a weight based on the number of searches to a search query included in the search history information, and calculate a vector whose component is the weight assigned to the search query. . Then, the evaluation information generation unit 135 uses a technique called t-SNE to calculate the closeness (similarity) between the query vector corresponding to the top user group S and the query vector corresponding to the target user group Cx. do.
The similarity calculated in step S4 is any value between 0 and 1.0, and is regarded as the estimated probability (estimated correctness probability) that the target attribute is correct in the target user group Cx.
Then, the evaluation information generation unit 135 generates score evaluation information including the number of users (second predetermined number) and the estimated correctness probability for each target user group Cx (step S5). Note that steps S4 and S5 correspond to the evaluation information generation step of the present invention.

Next, the graph generator 136 generates an analysis graph that visualizes the score evaluation information generated in step S5 (step S6).
FIG. 4 is an example of an analysis graph imaging the score evaluation information generated when the target attribute is "resident in Tokyo". In this analysis graph, the horizontal axis (first axis) is the number of users (second predetermined number) of the target user group Cx, and the vertical axis (second axis) is the estimated correct probability corresponding to the target user group Cx. In addition, in this analysis graph, the estimated correctness probabilities for each of the target user groups C0, C1, C2, . In addition, in FIG. 4, the estimated correctness probability is displayed as a percentage.

For example, in the analysis graph of FIG. 4, the estimated correctness probability corresponding to the target user group C0 substantially equal to the top user group S is 100%. Also, the estimated correctness probability corresponding to the target user group C1 having the top 100,000 scores is 98%. Also, the estimated correctness probability corresponding to the target user group C9 whose scores are the top 900,000 is 58%.
According to such an analysis graph, it can be seen that the estimated correctness probability decreases as the number of users (second predetermined number) in the target user group Cx increases.

Next, the annotation information generation unit 137 generates annotation information for each target user group Cx based on the score evaluation information generated in step S5 (step S7).
Specifically, the annotation information generating unit 137 calculates the estimated number of users whose target attribute is correct within the target user group Cx based on the estimated correct probability corresponding to the target user group Cx. For example, the estimated number of users is calculated by multiplying the number of users in the target user group Cx by the corresponding estimated correct probability.
Further, the annotation information generating unit 137 calculates the smallest score among the scores corresponding to the users belonging to the target user group Cx as the threshold score of the target user group Cx with respect to the target attribute. For example, when the score band of users belonging to the target user group C9 is 0.346 to 0.350, the minimum value of 0.346 in the score band is calculated as the threshold score.
Then, the annotation information generation unit 137 generates annotation information including the estimated number of users and the threshold score for each target user group Cx.
Note that the order of steps S6 and S7 may be changed.

Next, the providing unit 138 provides the score evaluation information, the analysis graph, and the annotation information to the provider terminal 30, which is the source of the request (step S8).
With the above, the flow of FIG. 3 ends.

The operator terminal 30 displays the analysis graph transmitted from the information processing device 10 on the display. In addition, when the operator performs an operation to select an arbitrary target user group Cx in the analysis graph (for example, aligning the cursor with an area corresponding to an arbitrary target user group Cx), the operator terminal 30 selects The number of users of the target user group Cx (displayed as "acquired number"), the estimated number of users (displayed as "estimated number"), the threshold score (displayed as "threshold"), etc. are displayed in a pop-up (see FIG. 4).

For example, in FIG. 4, when the target user group C9 is selected, the number of users corresponding to this target user group C9 (acquisition number: 536,744), the estimated number of users (estimated number: 1,000,000), and A pop-up display of a threshold score (threshold: 0.3460) is shown.

By confirming the analysis graphs and pop-ups displayed on the operator terminal 30, the operator determines how many users with the highest scores are acquired, and with what probability the number of users whose target attribute is correct is included. It is possible to assess whether the target attribute is

[Effect of this embodiment]
The information processing apparatus 10 of the present embodiment includes a storage unit 12 and a processor 13. The processor 13 reads an information processing program stored in the storage unit 12 to obtain a history acquisition unit 131 and a high-level user group setting unit. 133, a target user group setting unit 134, and an evaluation information generation unit 135, and perform the above-described processing.
In the present embodiment as described above, the estimated correctness probability corresponding to the target user group Cx is calculated by using the search history information of the high-ranking user group S with a high probability that the target attribute is correct as an evaluation criterion. By confirming the number of users in the target user group Cx and the score evaluation information including the estimated correctness probability, the business operator determines how many users with the correct target attribute can be obtained by obtaining the number of users from the top scores. It is possible to grasp whether it is included with a probability of , and evaluate the score corresponding to the target attribute. As a result, the estimated attribute information can be actively utilized in the marketing of services provided by businesses.
In particular, in the present embodiment, the estimated attribute information is generated by integrating the attributes estimated by various service providing servers 20, and the business operator, regarding the attribute provision information provided from the information processing device 10, It is difficult to have an image of the score. Therefore, it is considered that the score evaluation information of the present embodiment is of great importance.

In this embodiment, the target user group setting unit 134 sets a plurality of target user groups Cx (C0, C1, C2, . . . Cn) having different numbers of users.
In this embodiment, by confirming score evaluation information about a plurality of target user groups Cx, the operator can grasp changes in the estimated correctness probability with respect to changes in the number of users in the target user group Cx.

In the present embodiment, the processor 13 also functions as a graph generator 136 that generates an analysis graph in which score evaluation information is graphed.
In this embodiment, by checking the analysis graph, the operator can visually grasp the change in the estimated correctness probability with respect to the change in the number of users of the target user group Cx.

In this embodiment, the processor 13 functions as an annotation information generator 137 that generates annotation information (estimated number of users, threshold score) of score evaluation information.
In this embodiment, the operator can more accurately perform score evaluation based on the score evaluation information by checking the annotation information.

[Modification]
It should be noted that the present invention is not limited to the above-described embodiments, and includes modifications shown below within the scope of achieving the object of the present invention.

(Modification 1)
In the above-described embodiment, the evaluation information generator 135 calculates the estimated correctness probability using the user's search history information, but the present invention is not limited to this. For example, the evaluation information generation unit 135 may calculate the estimated correctness probability using various action histories such as browsing history and purchase history instead of the search history or together with the search history.

(Modification 2)
In the above embodiment, the annotation information such as the estimated number of users and the threshold score corresponding to the target user group Cx is popped up and displayed in response to the selection operation on the target user group Cx, but the present invention is not limited to this, and other It may be displayed in a format (eg, a table, etc.).

(Modification 3)
In the above embodiment, the score evaluation information is displayed as an analysis graph with the number of users in the target user group Cx as the first axis and the estimated correctness probability corresponding to the target user group Cx as the second axis. It is not limited to this, and may be displayed in other forms (for example, a table, etc.).

(Modification 4)
In the above-described embodiment, the graph generation unit 136 may generate a graph using an index different from indexes such as the number of users in the target user group Cx and the estimated correct probability corresponding to the target user group Cx. For example, the graph generating unit 136 may generate another graph having the number of users in the target user group Cx as the first axis and the estimated number of users corresponding to the target user group Cx or the threshold score as the second axis. In this case, the information processing device 10 may provide the operator terminal 30 with another graph instead of the analysis graph of the above embodiment or together with the analysis graph.

(Modification 5)
In the above embodiment, the number of users in the target user group Cx (C1, C2, . If the provider's request includes specification of a range of the number of users, the interval of the number of users between the target user groups Cx may differ between this range and another range.

(Modification 6)
In the above embodiment, at least one target user group Cx should be set, and score evaluation information corresponding to this target user group Cx should be generated. If the provider's request includes designation of the number of users, a target user group corresponding to this number of users may be set, and score evaluation information corresponding to this target user group may be generated.

(Modification 7)
In the above-described embodiment, the information processing apparatus 10 generates estimated attribute information for each user by integrating user attributes estimated by each service providing server 20, but the present invention is not limited to this. For example, the information processing device 10 acquires user action history information (for example, search history, browsing history, purchase history, etc.) from each service providing server 20 or user terminal 40, and based on the acquired action history information, the user Estimated attribute information may be generated by estimating the attribute of . As a method of estimating user attributes, a known technique using a model or the like can be used.

DESCRIPTION OF SYMBOLS 1... Information processing system 10... Information processing apparatus 11... Communication part 12... Storage part 121... Action history information storage part 122... Presumed attribute information storage part 13... Processor 131... History acquisition part 132... Estimated attribute information generation unit 133 Top user group setting unit 134 Target user group setting unit 135 Evaluation information generation unit 136 Graph generation unit 137 Annotation information generation unit 138 Provider 20 Service Provided server 30...Business operator terminal 40...User terminal Cx...Target user group S...Upper user group.

Claims (7)

a history acquisition unit that acquires the behavior history of a plurality of users on the network;
A high-ranking user group is set by selecting a first predetermined number of the users in descending order of the score based on estimated attribute information of the users including a score estimated for a target attribute that is an attribute to be evaluated. a setting unit;
a target user group setting unit that sets a target user group by selecting a second predetermined number of users equal to or greater than the first predetermined number in descending order of score based on the estimated attribute information;
calculating the degree of similarity between the action history of the top user group and the action history of the target user group as an estimated probability that the target attribute is correct in the target user group; and an evaluation information generation unit that generates score evaluation information including the estimated probability. The information processing apparatus according to claim 1, wherein said target user group setting unit sets a plurality of said target user groups in which said second predetermined number is a number different from each other. 3. The graph generating unit according to claim 2, further comprising a graph generation unit that generates a graph having the second predetermined number corresponding to the target user group as a first axis and the estimated probability corresponding to the target user group as a second axis. information processing equipment. 4. The method according to any one of claims 1 to 3, further comprising an estimated number of users calculating an estimated number of users in the target user group whose target attribute is correct based on the estimated probability corresponding to the target user group. or the information processing device according to claim 1. Any one of claims 1 to 4, further comprising a threshold calculation unit that calculates the smallest score among the scores estimated for the users belonging to the target user group as a threshold score of the target user group. or the information processing device according to claim 1. An information processing method for processing information by a computer,
a history acquisition step of acquiring behavior histories of a plurality of users on the network;
A high-ranking user group is set by selecting a first predetermined number of the users in descending order of the score based on estimated attribute information of the users including a score estimated for a target attribute that is an attribute to be evaluated. a configuration step;
a target user group setting step of setting a target user group by selecting a second predetermined number of users equal to or greater than the first predetermined number in descending order of the score based on the estimated attribute information;
calculating the degree of similarity between the action history of the top user group and the action history of the target user group as an estimated probability that the target attribute is correct in the target user group; and an evaluation information generation step of generating score evaluation information including the estimated probability. A computer readable and executable information processing program,
An information processing program causing the computer to function as the information processing apparatus according to any one of claims 1 to 5.

Images