JP2022178713A - Provision device, provision method, and provision program - Google Patents

Abstract

To provide useful information relating to users to become targets.SOLUTION: A provision device includes a receiving part, a generation part, a classification part, a grouping part, and a provision part. The receiving part receives, from a first proprietor, attribute information on users belonging to a prescribed category. The generation part uses user information prestored by a second proprietor to generate a model for classifying the users included in the user information, being the classification model for classifying the users having the attribute information received by the receiving part, into the same category. The classification part uses the classification model generated by the generation part to classify the users belonging to a prescribed category from among the users included in the user information. The grouping part groups the users classified by the classification part. The provision part provides the first proprietor with statistical information relating to the users grouped into the same group by the grouping part.SELECTED DRAWING: Figure 5

Description

The present invention relates to a provision device, a provision method, and a provision program.

2. Description of the Related Art Conventionally, various techniques have been proposed for analyzing a huge amount of data exchanged via the Internet from various aspects for the purpose of utilizing it in business scenes and the like. For example, there is known a technique of analyzing customers by taking into account seasonal trends using purchase history data or the like.

JP 2015-146145 A

However, conventional techniques have room for improvement in providing useful information about target users.

The present application has been made in view of the above, and aims to provide a provision device, a provision method, and a provision program capable of providing useful information regarding target users.

A provision device according to the present application includes a reception unit, a generation unit, a classification unit, a grouping unit, and a provision unit. The reception unit receives attribute information of users belonging to a predetermined category from the first business operator. The generation unit is a model for classifying users included in the user information using user information accumulated in advance by the second business operator, and classifies users having attribute information received by the reception unit into the same category. Generate a model. The classification unit classifies users belonging to a predetermined category from users included in the user information using the classification model generated by the generation unit. The grouping unit groups the users classified by the classifying unit. The providing unit provides the first operator with statistical information about the users grouped into the same set by the grouping unit.

According to one aspect of the embodiments, useful information about the target user can be provided.

FIG. 1 is a diagram illustrating an example of information processing according to an embodiment. FIG. 2 is a diagram illustrating an example of statistical information based on search history according to the embodiment. FIG. 3 is a diagram illustrating an example of statistical information based on purchase history according to the embodiment; FIG. 4 is a diagram showing another example of statistical information based on search history according to the embodiment. FIG. 5 is a diagram illustrating a configuration example of a provision device according to an embodiment; FIG. 6 is a diagram illustrating an example of statistical information regarding search history for each category according to the embodiment. FIG. 7 is a flowchart illustrating an example of a processing procedure by the provision device according to the embodiment; FIG. 8 is a diagram illustrating an example of a statistical information providing method according to a modification. FIG. 9 is a hardware configuration diagram showing an example of a computer that implements the functions of the provision device according to the embodiment or modification.

Embodiments for implementing the provision device, the provision method, and the provision program (hereinafter referred to as "embodiments") according to the present application will be described in detail below with reference to the drawings. Note that the provision device, the provision method, and the provision program according to the present application are not limited to the embodiments described below. Further, the embodiments described below can be appropriately combined within a range that does not contradict the processing contents. Also, in the embodiments described below, the same parts are denoted by the same reference numerals, and overlapping descriptions are omitted.

[1. Overview of information processing]
Hereinafter, an example of information processing according to the embodiment will be described with reference to the drawings. FIG. 1 is a diagram illustrating an example of information processing according to an embodiment. Note that the information processing described below can be executed in parallel for any number of operator terminals, without being limited to the example described below.

As shown in FIG. 1 , the provision system 1 according to the embodiment includes a business operator terminal 10 and a provision device 100 . The operator terminal 10 and the providing device 100 are each connected to a network N (see FIG. 5, for example) by wire or wirelessly. The network N is a communication network such as a LAN (Local Area Network), a WAN (Wide Area Network), a telephone network (mobile telephone network, fixed telephone network, etc.), a local IP (Internet Protocol) network, or the Internet. . The network N may include a wired network or a wireless network. The operator terminal 10 and the providing device 100 can communicate with each other through the network N. FIG.

A business operator terminal 10 shown in FIG. 1 is an information processing device managed by a business operator X (an example of a “first business operator”). For example, a business operator X repeatedly and continuously independently operates a business of providing predetermined goods or services to consumers. The business terminal 10 is typically a desktop or notebook personal computer. Note that the business terminal 10 may be realized by any information processing terminal such as a smart phone, a tablet, or a PDA (Personal Digital Assistant).

Further, the business terminal 10 can display information provided by the providing device 100 using a web browser or an application. Note that, when receiving control information for realizing information display processing from the providing apparatus 100 or the like, the business terminal 10 realizes display processing according to the control information.

Further, the business terminal 10 collects customer information related to the business run by the business operator X and manages the collected customer information. Further, the business terminal 10 analyzes the customer information collected by the business X to identify the user's label (an example of "attribute information") belonging to a predetermined category. For example, the business operator terminal 10 allows a customer who belongs to the category of "resident in Chuo Ward", age "50's", and sex "male" to "purchase a refrigerator" or "annual income of 10 million yen or more". to identify labels such as . Then, the provider terminal 10 provides the providing device 100 with information on the label specified for each predetermined category at arbitrary timing.

The provision device 100 shown in FIG. 1 is information managed by a business operator Y (an example of a “second business operator”) that repeatedly and continuously independently operates a business that provides users with various services on the Internet. processing equipment. For example, the provision device 100 is implemented by a server device, a cloud system, or the like. Note that the providing device 100 may function as a distribution device that distributes control information to a user terminal (not shown) used by a service user or to the business terminal 10 . Here, the control information is described in, for example, a script language such as JavaScript (registered trademark) or a style sheet language such as CSS (Cascading Style Sheets). Note that the application itself distributed from the providing apparatus 100 may be regarded as control information.

The providing apparatus 100 provides various services on the Internet to service users through online contents such as portal sites. The services provided by the providing device 100 include a search engine site, a news site, a technical explanation site, a shopping site, a finance site (stock price site), a route search site, a map providing site, a travel site, It includes various services provided through web pages related to restaurant introduction sites and web blogs. The providing device 100 also includes various applications (for example, a portal application, a news application, an auction application, a weather forecast application, a shopping application, a finance (stock price) application, a route search application, a map providing application, It can provide information to be displayed in travel apps, restaurant introduction apps, blog browsing apps, etc.).

In addition, the providing apparatus 100 can receive search words input by service users through the provision of various services described above. Further, the providing apparatus 100 can store the received search words as a search history (search log). In addition, the providing apparatus 100 may be configured based on user attributes (for example, demographic attributes such as age, gender, region, etc., usage history of online contents of various services, etc.) based on the operation, browsing, etc. of service users in various services. Estimated psychographic attributes, etc.) and behavior histories of service users in various services are collected and stored in association with user IDs. Further, the providing device 100 may be a device that distributes to a user terminal (not shown) used by a service user information to be displayed in an application related to various services pre-installed in the user terminal. Also, the providing device 100 may be a server device that distributes application data itself.

Further, the providing apparatus 100 creates a classification model for the operator X based on the label provided by the operator X using the service user information accumulated in advance by the operator Y, and creates the classification model Provides statistics based on each user categorized by

As shown in FIG. 1, the provision device 100 first receives label information of a customer (user) belonging to a predetermined category for each predetermined category from the provider terminal 10 (step S1). In the example shown in FIG. 1, the providing apparatus 100, for example, assigns labels LA-1 and LA-2 of users belonging to category CT-A, labels LB of users belonging to category CT-B, and labels to categories CT-C. It accepts the label LC of the user to which it belongs.

Subsequently, the providing apparatus 100 generates a classification model for classifying users having label information received from the business operator terminal 10 into the same category (step S2). That is, the providing apparatus 100 uses service user information (an example of “user information”) accumulated in advance by the operator Y to generate a model for classifying service users included in the service user information.

For example, the providing apparatus 100 uses the learning data to classify (estimate) whether the user corresponding to the input data belongs to a predetermined category corresponding to the label in accordance with the input of the input data. ) to learn a classification model. Specifically, the providing device 100 learns parameters such as connection coefficients (weights) between interconnected nodes that constitute a model (neural network).

Also, the learning of the classification model is performed using various machine learning techniques. For example, the providing apparatus 100 uses data to be input to a model and a classification label (correct label) indicating an output when the data is input, a learning method, that is, a supervised learning method, to obtain model parameters. learn. Note that the above is just an example, and the providing apparatus 100 may learn the model parameters by any learning method as long as the model parameters can be learned.

Here, the learning of the classification model MA shown in FIG. 1 will be described with a specific example. For example, the business operator Y who manages the providing apparatus 100 corresponds to both label LA-1 (“purchase of refrigerator”) and label LA-2 (“annual income of 10 million yen or more”) from the service user information. Get sample data for Business operator Y assigns category CT-A (“resident in Chuo Ward” + “50s” + “male”) corresponding to labels LA-1 and LA-2 to the obtained sample data. Prepare learning data assigned as (correct label).

Next, the providing apparatus 100 causes the model to learn the prepared learning data, and when inputting data associated with an arbitrary user corresponding to the labels LA-1 and LA-2, the score output by the model is Among them, learning methods such as back propagation so that the score for classifying the corresponding user into category CT-A is equal to or higher than a predetermined threshold (so that users belonging to category CT-A can be correctly classified into category CT-A) is used to learn the parameters described above. In this way, a classification model MA can be generated that classifies service users into category CT-A. Similarly, the providing device 100 can generate a classification model MB, a classification model MC, and the like. The providing apparatus 100 may generate a classification model such that the number of service users classified by the generated classification model is greater than or equal to a predetermined threshold.

The providing apparatus 100 can employ any network such as DNN (Deep Neural Network) or SVM (Support Vector Machine) as a model (network) for generating each classification model. The classification model may also be a decision tree.

Subsequently, the providing apparatus 100 classifies users belonging to a predetermined category from among the service users included in the pre-stored service user information using the classification model (step S3).

Subsequently, the providing device 100 groups the users classified by the classification model (step S4). An example of grouping users classified into category CT-A (“Chuo-ku Resident”+“50s”+“Male”) by the classification model MA will be described below.

For example, the providing apparatus 100 acquires data associated with each user belonging to category CT-A (“Chuo-ku Resident” + “50s” + “Male”) from the service user information, and acquires the acquired data. is used to group (cluster) each user. Note that grouping can be performed, for example, by comparing feature information calculated in advance for each user and based on the degree of similarity of the feature information. Here, the feature information may be information obtained by vectorizing trends in search words entered by the user, action history in various services, user attributes estimated from the action history, and the like. may be information obtained by simply quantifying the attributes of Also, the grouping method may be hierarchical or non-hierarchical. Arbitrary methods such as Ward's method, group average method, shortest distance method, and longest distance method can be used as calculation methods used for hierarchical grouping. Any method such as the k-means method can be used as a calculation method for non-hierarchical grouping.

After the grouping, the providing device 100 provides the business operator X with statistical information about each user grouped into the same set (hereinafter referred to as "same cluster") (step S5).

The providing apparatus 100 individually aggregates data characterizing each cluster (for example, the number of users≧10) into which users belonging to a predetermined category are grouped. Then, the providing apparatus 100 generates statistical information based on the data aggregated individually for each cluster, and provides the business operator X with the statistical information. Note that the providing apparatus 100 may set not only the lower limit but also the upper limit (for example, the number of users≦20) for the number of users belonging to each grouped cluster.

For example, the providing apparatus 100 refers to the search history associated with each user belonging to the same cluster, individually aggregates the search words input by each user for each cluster, and identifies the most frequently searched search word. Then, the providing device 100 can provide the business operator X with the information of the search words specified for each cluster as statistical information. FIG. 2 is a diagram illustrating an example of statistical information based on search history according to the embodiment.

As shown in FIG. 2, the providing apparatus 100 associates, for example, the information of the search word with the largest number of user searches in each cluster with the information of the corresponding category, and creates statistical information in a list format. can be provided to X. In the example shown in FIG. 2, the search word "ballroom dance" that was searched the most in one cluster grouped from the users belonging to the category CT-A, and "Immigration" was the search word with the highest number of searches in one cluster, and "world heritage" was the search word with the highest number of searches in one cluster grouped from users belonging to category CT-A. etc. are shown.

Further, for example, the providing apparatus 100 refers to the purchase history associated with each user belonging to the same cluster, individually aggregates the number of purchases of products purchased by each user for each cluster, and finds the most purchased product. Identify. Then, the providing apparatus 100 can provide the business operator X with the information of the products specified for each cluster as statistical information. FIG. 3 is a diagram illustrating an example of statistical information based on purchase history according to the embodiment;

As shown in FIG. 3, the providing apparatus 100, for example, associates the information of the product with the largest number of purchases by the user in each cluster with the information of the corresponding category, and creates the statistical information in list format. can be provided to In the example shown in FIG. 3, in one cluster grouped from the users belonging to the category CT-A, the item "camera" which is the product with the largest number of purchases, and "Vacuum cleaner", which is the product with the largest number of purchases in one cluster, and "Bedding", which is the product with the largest number of purchases in one cluster grouped from users belonging to category CT-A, etc. It is shown.

Further, the provision device 100 is not limited to providing list-format statistical information. FIG. 4 is a diagram showing another example of statistical information based on search history according to the embodiment. As shown in FIG. 4, the provision device 100 associates the information of the search word with the largest number of user searches in each cluster with the corresponding classification model MA, and creates statistical information in the form of a keyword map. may be provided to

In addition, the providing device 100 can provide information based on the score of each user output from the classification model as statistical information. In the examples shown in FIGS. 2 and 4, the average value VA1-1 of the output scores output from the classification model MA for each user belonging to a set (cluster) associated with the search word: "ballroom dance" is the search word: "ballroom dance". provided in connection with the Dance. Similarly, average values VA1-2 of output scores output from the classification model MA for each user belonging to a set (cluster) associated with the search word: "migration" are provided in association with the search word: "migration." . Similarly, average values VA1-3 of output scores output from the classification model MA for each user belonging to a set (cluster) associated with the search word: "world heritage" are provided in association with the search word: "world heritage". ing. As a result, the operator X, who has received the statistical information, can specify a search word that has a stronger correlation with the users belonging to the category CT-A.

In this way, the providing device 100 can provide the business operator X with useful information regarding users belonging to a predetermined category. In other words, the providing apparatus 100 can provide the business operator X with useful information regarding users who are targets of the business operator X, for example. Note that the providing apparatus 100 can provide the business operator X with statistical information related to user attributes and the like, in addition to search histories and purchase histories.

Further, the providing apparatus 100 provides statistical information on users belonging to the same cluster on condition that the number of users grouped into the same cluster is equal to or greater than a predetermined threshold. This makes it possible to provide statistical information that guarantees anonymity.

Further, when the number of users grouped into the same set is less than a predetermined threshold, the providing apparatus 100 selects a predetermined category (for example, category CT-A) so that the number of users is equal to or greater than the predetermined threshold. : “Chuo-ku resident” + “50s” + “male”) may be relaxed. That is, the providing apparatus 100 may relax the threshold for determining the score for classifying the user into a predetermined category among the scores output from the classification model. Thereby, the number of users belonging to a predetermined category can be expanded, and the number of grouped users can be adjusted so as to satisfy a predetermined condition. Note that the providing apparatus 100 may relax the conditions for determining similar users when performing clustering.

[2. Configuration of provision device]
The configuration of the provision device 100 according to the embodiment will be described with reference to FIG. FIG. 5 is a diagram illustrating a configuration example of a provision device according to an embodiment;

As shown in FIG. 5, the provision device 100 includes a communication unit 110, a storage unit 120, and a control unit . Note that FIG. 5 shows a configuration example of the providing apparatus 100, and is not limited to the form shown in FIG.

(Communication unit 110)
The communication unit 110 is connected to the network N by wire or wirelessly, for example, and transmits and receives information to and from other devices such as the operator terminal 10 via the network N. The communication unit 110 is implemented by, for example, a NIC (Network Interface Card), an antenna, or the like. The network N is a communication network such as a LAN (Local Area Network), a WAN (Wide Area Network), a telephone network (mobile telephone network, fixed telephone network, etc.), a regional IP (Internet Protocol) network, or the Internet. . The network N may include wired networks or wireless networks.

The communication unit 110 receives label information of customers (users) belonging to a predetermined category for each predetermined category from the business operator terminal 10 . Also, the communication unit 110 transmits the statistical information generated by the control unit 130 to the operator terminal 10 .

(storage unit 120)
The storage unit 120 is realized by, for example, a RAM (Random Access Memory), a semiconductor memory device such as a flash memory, or a storage device such as a hard disk or an optical disk. Storage unit 120 stores programs and data for realizing information processing executed by each unit of control unit 130 .

As shown in FIG. 5 , the storage unit 120 has a service user information storage unit 121 and a model storage unit 122 . The service user information storage unit 121 stores service user information related to the service user in association with identification information (for example, user ID) for specifying the service user. Service user information includes service user's search history (search log), user attributes based on service user's operation and browsing in various services (for example, demographic attributes such as age, gender, region, etc., including psychographic attributes estimated based on usage history of online content of various services, etc., and behavior history of service users in various services, etc. The model storage unit 122 stores model data generated for each predetermined category by the control unit 130 .

(control unit 130)
The control unit 130 is a controller that controls the providing device 100 . The control unit 130 executes various programs (for example, an information processing program) stored in a storage device inside the providing apparatus 100 using a RAM as a work area by a CPU (Central Processing Unit), MPU (Micro Processing Unit), or the like. It is realized by Also, the control unit 130 may be implemented by an integrated circuit such as an ASIC (Application Specific Integrated Circuit) or an FPGA (Field Programmable Gate Array).

As shown in FIG. 5 , the control unit 130 has a reception unit 131 , a generation unit 132 , a classification unit 133 , a grouping unit 134 and a provision unit 135 . Control unit 130 implements or executes the functions and actions of information processing described below by means of these units. Note that the internal configuration of the control unit 130 is not limited to the configuration shown in FIG. 4, and may be another configuration as long as it performs information processing to be described later. Moreover, the connection relationship of each part of the control unit 130 is not limited to the connection relationship shown in FIG. 4, and may be another connection relationship.

(Reception unit 131)
Receiving unit 131 receives user belonging to a predetermined category (e.g., “resident in Chuo-ku” + “50s” + “male”) from a first business operator (e.g., “business operator X”) through communication unit 110 . label information (for example, "purchasing a refrigerator" or "annual income of 10 million yen or more") is accepted. The receiving unit 131 sends information about the received label to the generating unit 132 .

(Generating unit 132)
The generating unit 132 generates, for each predetermined category, a classification model for classifying users having label information acquired from the receiving unit 131 into the same category. The generating unit 132 stores the generated classification model data in the model storage unit 122 .

(Classification unit 133)
The classification unit 133 uses the classification model stored in the model storage unit 122 to classify service users included in the service user information stored in the service user information storage unit 121 into predetermined categories. Categorize the users to which they belong. Classification unit 133 sends identification information (for example, user ID) for identifying each classified user to grouping unit 134 .

(Grouping unit 134)
The grouping unit 134 groups the users classified by the classifying unit 133 using an arbitrary grouping method, as described above. The grouping unit 134 sends identification information (for example, user ID) for specifying each grouped user to the providing unit 135 .

(Providing unit 135)
The providing unit 135 provides the operator X with statistical information on each user grouped into the same cluster, as described above. Specifically, for each cluster in which the number of users grouped into the same cluster is 10 or more, the providing unit 135 individually stores data (search words, purchased products, user attributes, etc.) that characterize users belonging to each cluster. Aggregate to Then, the providing unit 135 generates statistical information based on the data specified for each cluster, and provides the business operator X with the statistical information through the communication unit 110 .

FIG. 6 is a diagram illustrating an example of statistical information regarding search history for each category according to the embodiment. For example, as shown in FIG. 6 , the providing unit 135 creates list-format statistical information that associates the information of the search words most searched by each user grouped into the same cluster with the corresponding category. can be provided to

[3. Processing procedure]
A processing procedure performed by the providing apparatus 100 according to the embodiment will be described below with reference to FIG. FIG. 7 is a flowchart illustrating an example of a processing procedure by the provision device according to the embodiment; The processing procedure shown in FIG. 7 is executed by the control unit 130 of the providing apparatus 100 . The processing procedure shown in FIG. 7 is repeatedly executed while the providing apparatus 100 is in operation.

As shown in FIG. 7, the accepting unit 131 accepts label information of a user belonging to a predetermined category from the operator terminal 10 through the communication unit 110 (step S101).

In addition, the generation unit 132 generates a classification model for classifying users having label information acquired from the reception unit 131 into the same category for each predetermined category (step S102).

Further, the classification unit 133 uses the classification model stored in the model storage unit 122 to select a predetermined service user from among the service users included in the service user information stored in the service user information storage unit 121. The users belonging to the category are classified (step S103).

Moreover, as described above, the grouping unit 134 groups the users classified by the classifying unit 133 using an arbitrary grouping method (step S104).

Also, the providing unit 135 provides the operator X with statistical information on the users grouped into the same cluster by the grouping unit 134 through the communication unit 110 (step S105), and ends the processing procedure shown in FIG.

[4. Modification]
(4-1. Provision of statistical information based on multiple classification models)
In the above-described embodiment, the providing apparatus 100 may provide a combination of statistical information of users classified into each category. FIG. 8 is a diagram illustrating an example of a statistical information providing method according to a modification. FIG. 8 shows an example of providing statistical information on users belonging to the category of "Chuo-ku Resident"+"50's".

As shown in FIG. 8, when providing statistical information about users belonging to the category of “Chuo-ku Resident”+“50s”, the providing apparatus 100 provides classification corresponding to each of a plurality of categories with more limited conditions. The model is used to classify users and group the classified users. For example, the providing apparatus 100 uses the classification model MA to classify the users belonging to the category of "resident of Chuo Ward"+"50's"+"male". Further, the providing apparatus 100 classifies users belonging to the category of “resident of Chuo Ward”+“50s”+“female” using the classification model MX.

Subsequently, the providing apparatus 100 generates statistical information about the users grouped into the same cluster by grouping the users classified into the categories of “Chuo-ku Resident” + “Fifties” + “Male”. . In addition, the providing apparatus 100 generates statistical information about the users grouped into the same cluster by grouping the users classified into the categories of “Chuo-ku Resident” + “50s” + “Female”.

Then, the providing apparatus 100 collects the statistical information about the users belonging to the category of “resident in Chuo-ku” + “50s” + “male” and Provide statistical information in combination with As a result, the providing apparatus 100 can provide statistical information about users belonging to a desired category without reducing the classification accuracy of users belonging to a predetermined category.

(4-2. Use of classification model)
In the above-described embodiment, the provision device 100 uses the classification model stored in the model storage unit 122 when providing statistical information to other companies that are in a competitive relationship with the business run by the operator X. good too.

[5. Hardware configuration]
A provision device 100 according to an embodiment or a modified example is implemented by a computer 1000 configured as shown in FIG. 9, for example. FIG. 9 is a hardware configuration diagram showing an example of a computer that implements the functions of the provision device according to the embodiment or modification.

Computer 1000 has CPU 1100 , RAM 1200 , ROM 1300 , HDD 1400 , communication interface (I/F) 1500 , input/output interface (I/F) 1600 and media interface (I/F) 1700 .

The CPU 1100 operates based on programs stored in the ROM 1300 or HDD 1400 and controls each section. The ROM 1300 stores a boot program executed by the CPU 1100 when the computer 1000 is started, a program depending on the hardware of the computer 1000, and the like.

HDD 1400 stores programs executed by CPU 1100 and data used by these programs. Communication interface 1500 receives data from other devices via network (communication network) N, sends the data to CPU 1100, and transmits data generated by CPU 1100 to other devices via network (communication network) N.

The CPU 1100 controls output devices such as displays and printers, and input devices such as keyboards and mice, via an input/output interface 1600 . CPU 1100 acquires data from an input device via input/output interface 1600 . CPU 1100 also outputs the generated data to an output device via input/output interface 1600 .

Media interface 1700 reads programs or data stored in recording medium 1800 and provides them to CPU 1100 via RAM 1200 . CPU 1100 loads such a program from recording medium 1800 onto RAM 1200 via media interface 1700, and executes the loaded program. The recording medium 1800 is, for example, an optical recording medium such as a DVD (Digital Versatile Disc) or a PD (Phase change rewritable disc), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory. and so on.

For example, when the computer 1000 functions as the provision device 100 according to the embodiment, the CPU 1100 of the computer 1000 implements the functions of the control unit 130 by executing programs loaded on the RAM 1200 . In addition, data in storage unit 120 is stored in HDD 1400 . CPU 1100 of computer 1000 reads these programs from recording medium 1800 and executes them, but as another example, these programs may be obtained from another device via network (communication network) N.

[6. others]
Of the processes described in the above-described embodiment or modification, all or part of the processes described as being performed automatically can be performed manually, or the processes described as being performed manually can be performed manually. All or part of the processing can also be performed automatically by known methods. In addition, information including processing procedures, specific names, various data and parameters shown in the above documents and drawings can be arbitrarily changed unless otherwise specified.

In the above-described embodiment, in order to realize the provision method (for example, see FIG. 7) by the provision apparatus 100, each section (the reception section 131, the generation section 132, the classification section 133, the grouping section) of the control section 130 of the provision apparatus 100 134 and the providing unit 135) may be realized as an add-on to a providing program pre-installed in the providing apparatus 100, or may be flexibly implemented as a dedicated providing program using a lightweight programming language or the like. It may be realized by describing in

In addition, in the above-described embodiment and modifications, the provision device 100 is physically dispersed into a device that provides various services to service users and a device that provides statistical information to business operators. may have been

Also, each component of each device illustrated is functionally conceptual, and does not necessarily need to be physically configured as illustrated. In other words, the specific form of distribution and integration of each device is not limited to the one shown in the figure, and all or part of them can be functionally or physically distributed and integrated in arbitrary units according to various loads and usage conditions. Can be integrated and configured. For example, reception unit 131 and generation unit 132 of control unit 130 may be functionally integrated.

Further, the above-described embodiment and modifications can be appropriately combined within a range that does not contradict the processing content.

[7. effect]
The provision device 100 according to the above embodiment or modification includes a reception unit 131 , a generation unit 132 , a classification unit 133 , a grouping unit 134 and a provision unit 135 . The reception unit 131 receives label information (an example of “attribute information”) of a user belonging to a predetermined category, for example, from an operator X (an example of a “first operator”). The generation unit 132, for example, uses service user information (an example of "user information") accumulated in advance by the operator Y (an example of a "second operator") to determine users included in the service user information. A classification model for classifying users having label information received by the receiving unit 131 into the same category is generated. The classification unit 133 uses the classification model generated by the generation unit 132 to classify users belonging to a predetermined category from users included in the service user information. The grouping unit 134 groups the users classified by the classifying unit 133 . The providing unit 135 provides the business operator X with statistical information about the users grouped into the same set by the grouping unit 134 .

In this way, the provision device 100 according to the embodiment or the modified example can provide useful information regarding users belonging to a predetermined category to the business operator X, for example. In other words, the providing apparatus 100 can provide the business operator X with useful information about the target user, in other words, a wide variety of information that cannot be known within the business scope of the business operator X.

In addition, in the provision device 100 according to the embodiment or the modified example, the provision unit 135 puts users into the same set on condition that the number of users grouped into the same set by the grouping unit 134 is equal to or greater than a predetermined threshold. Provides statistical information about the users to whom they belong. As a result, the provision device 100 according to the embodiment or the modified example can provide users with anonymity.

Further, in the providing apparatus 100 according to the embodiment or the modification, if the number of users grouped into the same set by the grouping unit 134 is less than a predetermined threshold, the providing unit 135 sets the number of users to a predetermined threshold. As described above, the threshold conditions for classifying whether a user belongs to a predetermined category are relaxed. This makes it possible to actively provide statistical information.

In addition, in the provision device 100 according to the embodiment or modification, the provision unit 135 provides information based on the scores output from the classification model (for example, average score of users belonging to each cluster) as statistical information. may This makes it possible to provide an objective index for easily identifying information correlated with a predetermined category (for example, "resident in Chuo Ward"+"50's"+"male") in the statistical information.

In addition, in the provision device 100 according to the embodiment or modification, the provision unit 135 provides a combination of statistical information of users classified into each category. This makes it possible to provide statistical information about users belonging to a desired category without reducing the accuracy of classifying users belonging to a predetermined category.

As described above, the embodiments of the present application have been described in detail based on several drawings, but these are examples, and various modifications and It is possible to carry out the invention in other forms with modifications.

Also, the above-mentioned "section, module, unit" can be read as "means" or "circuit". For example, the control unit can be read as control means or a control circuit.

10 provider terminal 100 providing device 110 communication unit 120 storage unit 121 service user information storage unit 122 model storage unit 130 control unit 131 reception unit 132 generation unit 133 classification unit 134 grouping unit 135 provision unit

Claims (7)

a reception unit that receives user attribute information belonging to a predetermined category from the first business operator;
A model for classifying users included in the user information using user information preliminarily accumulated by a second business operator, the classification model classifying users having the attribute information received by the reception unit into the same category. a generator that generates
a classification unit that classifies users belonging to the predetermined category from users included in the user information using the classification model generated by the generation unit;
a grouping unit that groups the users classified by the classifying unit;
a providing unit that provides the first business operator with statistical information about users grouped into the same set by the grouping unit. The providing unit
2. The statistical information relating to the users belonging to the same group is provided on condition that the number of users grouped into the same group by the grouping unit is equal to or greater than a predetermined threshold. providing equipment. The providing unit
If the number of users grouped into the same set by the grouping unit is less than a predetermined threshold, the users are classified as to whether they belong to the predetermined category such that the number of users is equal to or greater than the predetermined threshold. 3. The providing device according to claim 1, wherein a threshold condition for providing is relaxed. The providing unit
The providing apparatus according to any one of claims 1 to 3, wherein information based on a score output from the classification model is provided as the statistical information. The providing unit
5. The providing device according to any one of claims 1 to 4, wherein the statistical information of users classified into each category is combined and provided. A computer-implemented method of providing, comprising:
a receiving step of receiving user attribute information belonging to a predetermined category from a first business operator;
A model for classifying users included in the user information by using user information accumulated in advance by a second business operator, the classification model for classifying users having the attribute information received by the receiving step into the same category. a generating step for generating
a classification step of classifying users belonging to the predetermined category from users included in the user information using the classification model generated in the generation step;
a grouping step of grouping the users classified by the classification step;
and a provision step of providing statistical information on the users grouped into the same set by the grouping step to the first business operator. to the computer,
a reception procedure for receiving user attribute information belonging to a predetermined category from a first business operator;
A model for classifying users included in the user information by using user information accumulated in advance by a second business operator, the classification model for classifying users having the attribute information received by the reception procedure into the same category. a generation step that generates
a classification procedure for classifying users belonging to the predetermined category from users included in the user information using the classification model generated by the generation procedure;
a grouping procedure for grouping the users classified by the classification procedure;
and a providing procedure for providing the first business operator with statistical information relating to the users grouped into the same set by the grouping procedure.

Images