Data processing device, data processing system, data processing method, and program

Abstract

Provided is a data processing device which simply and suitably contributes to marketing activities. In a data processing device (110), an input unit (111) receives input data relating to prescribed consumption behavior. A federated learning model (112) is generated such that at least some local learning models generated for each of a plurality of different businesses are federated. The federated learning model (112) is generated by learning relationships between a plurality of customer groups generated from business customer data possessed by a business, and consumption behavior corresponding to the business. The federated learning model (112) is configured to be able to output potential customer data with respect to input data. An output unit (113) outputs the potential customer data.

Description

Technical Field

[0001] The present invention relates to a data processing device, a data processing system, a data processing method, and a program.

Background Art

[0002] Business operators who provide products, services, etc. utilize various market data in order to obtain business opportunities. Also, proposals for using technologies such as machine learning as a method for utilizing market data have been disclosed.

[0003] For example, Patent Document 1 discloses calculating a purchase expectation degree in a trial environment system for allowing a user to try, by monitoring the usage status of application software and considering customer weight information having a correlation with the user's purchase expectation degree.

[0004] Also, Patent Document 2 discloses a technique for creating household account book information by user, income level, and regional area in a marketing server by collecting each user's purchase information from a plurality of customer information databases and based on the collected purchase information.

[0005] Patent Document 3 discloses a technique for integrating a plurality of individually learned models obtained by performing additional learning based on individual operation data obtained by operating each of a plurality of operating devices after incorporating an initial learned model for controlling a predetermined operating device into the plurality of operating devices.

[0006] Patent Document 4 discloses a purchase trend prediction device that predicts a user's purchase trend based on behavior data indicating the user's behavior by an information terminal via a network.

Prior Art Documents

Patent Documents

[0007]

Patent Document 1

Patent Document 2

[0008] If businesses have a wealth of good data that forms the basis for marketing regarding customers' purchasing behavior or service usage (i.e., specific consumer behavior), they can more easily utilize this data for marketing. Therefore, methods for multiple businesses to use each other's customer data can be considered. However, it is not possible to share customers' personal information across multiple businesses.

[0009] The purpose of this disclosure is to provide a technology that contributes to marketing activities in a simple and suitable manner, in light of the above-mentioned problems.

[0010] A data processing device according to one aspect of this disclosure includes an input unit, a federated learning model, and an output unit. The input unit receives input data relating to predetermined consumer behavior. The federated learning model is generated such that at least a portion of local learning models generated for each of several different businesses are federated. The local learning models are generated by learning the relationship between multiple customer groups generated from business customer data owned by the businesses and the consumer behavior corresponding to the businesses. The federated learning model is configured to output prospective customer data for the input data. The output unit outputs prospective customer data.

[0011] A data processing method according to one aspect of this disclosure involves a computer performing the following steps: The computer receives input data relating to a predetermined consumer behavior. The computer supplies the input data to a federated learning model generated by combining multiple different local learning models that have learned the relationship between business customer data and consumer behavior based on business customer data owned by a predetermined business. The computer receives prospective customer data relating to consumer behavior as output from the federated learning model for the input data. The computer outputs the received prospective customer data.

[0012] A program according to one aspect of this disclosure causes a computer to perform the following actions: The computer receives input data relating to a given consumer behavior. The computer supplies the input data to a federated learning model generated by combining multiple different local learning models that have learned the relationship between business customer data and consumer behavior based on business customer data owned by a given business. The computer receives prospective customer data relating to consumer behavior as output from the federated learning model for the input data. The computer outputs the received prospective customer data.

[0013] According to this disclosure, it is possible to provide a data processing device, etc., that can easily and suitably contribute to marketing activities. [Brief explanation of the drawing]

[0014] [Figure 1] This is a block diagram of the data processing device according to Embodiment 1. [Figure 2] This is a flowchart of the data processing method according to Embodiment 1. [Figure 3] This is a block diagram of the associative learning model generation device according to Embodiment 1. [Figure 4] This is a flowchart of the associative learning model generation method according to Embodiment 1. [Figure 5] This is a block diagram of the data processing system according to Embodiment 2. [Figure 6]It is a block diagram of a data processing device according to Embodiment 2. [Figure 7] It is a block diagram of a local data processing device according to Embodiment 2. [Figure 8] It is a diagram showing the processing of the operator customer data generation unit. [Figure 9] It is a diagram showing operator customer data. [Figure 10] It is a diagram showing the processing of the local learning model. [Figure 11] It is a sequence diagram showing the processing executed by the federated learning model generation system. [Figure 12] It is a diagram showing the configuration of the federated learning model. [Figure 13] It is a diagram showing the processing of the federated learning model. [Figure 14] It is a sequence diagram showing the processing executed by the data processing system. [Figure 15] It is a block diagram of a data processing device according to Embodiment 3. [Figure 16] It is a block diagram of a local data processing device according to Embodiment 3. [Figure 17] It is a sequence diagram of the update method according to Embodiment 3. [Figure 18] It is a block diagram exemplifying the hardware configuration of a computer.

Modes for Carrying Out the Invention

[0015] Hereinafter, the present invention will be described through embodiments of the invention, but the invention according to the claims is not limited to the following embodiments. Also, not all of the configurations described in the embodiments are essential as means for solving the problems. For clarity of explanation, the following description and drawings have been appropriately omitted and simplified. In each drawing, the same elements are denoted by the same reference numerals, and duplicate explanations are omitted as necessary.

[0016] <Embodiment 1> First, Embodiment 1 of this disclosure will be described. Figure 1 is a block diagram of a data processing device according to Embodiment 1. The data processing device 110 shown in Figure 1 receives input data relating to a predetermined consumer behavior and outputs prospective customer data relating to this consumer behavior.

[0017] The specified consumer behaviors include, for example, the purchase of goods or tickets, or the use of services, which involve direct or indirect monetary payment. The prospective customer data is, for example, data that defines customers who are relatively likely to engage in the above consumer behaviors, using a predetermined segment. Specifically, for example, prospective customers may be grouped by age, gender, address, family structure, hobbies, occupation, medical history, or past consumer behavior history. As a result, the data processing device 110 provides, for example, a specified business operator with data that can be used as a reference for determining the target customer segment for marketing activities. Marketing activities are activities carried out by a business operator and include sales, advertising, and the sale or provision of goods and services to customers.

[0018] The data processing device 110 may consist of, for example, a computer, server, or dedicated equipment with communication capabilities. In the following description, "computer" may include server devices, blades, and cloud computing systems. The data processing device 110 mainly consists of an input unit 111, a federated learning model 112, and an output unit 113.

[0019] The input unit 111 receives input data related to predetermined consumer behavior from a predetermined external device or the like. The input data related to predetermined consumer behavior may be, for example, data indicating a predetermined product or service. The input data related to consumer behavior may also indicate detailed specifications of the product or service.

[0020] The federated learning model 112 is generated so that at least a portion of the local learning models generated for each of several different businesses are federated. In this embodiment, "federated" means, for example, connecting the data structures that each of the multiple local learning models possesses, but the definition of federated is not limited to this. By being generated by federating multiple local learning models, the federated learning model 112 includes the features that each of the local learning models possesses. Furthermore, by being generated by federating local learning models, the federated learning model 112 can realize an algorithm that cross-utilizes the data that each of the multiple local learning models possesses. The federated learning model 112 can employ various methods known to those skilled in the art as "federated learning." Federated learning may also be referred to as integration, for example.

[0021] The local learning model described above is a learning model in itself. The local learning model is generated by learning the relationship between multiple customer groups, which are generated from customer data owned by the business, and the corresponding consumer behavior of the business.

[0022] The associative learning model 112 is configured to output prospective customer data for the input data received by the input unit 111. For example, if the associative learning model 112 receives input data indicating a consumer behavior of purchasing a specific product, it outputs prospective customer data for prospective customers who are relatively likely to purchase this product.

[0023] The output unit 113 outputs the prospective customer data output by the associative learning model 112 to the aforementioned predetermined external device. The prospective customer data includes, for example, indicators that show the likelihood that customers in each segment will become customers exhibiting the consumption behavior related to the input data, according to pre-set customer segments.

[0024] Next, the processing performed by the data processing device 110 will be described with reference to Figure 2. Figure 2 is a flowchart of the data processing method according to Embodiment 1. The flowchart shown in Figure 2 is started, for example, when the associative learning model 112 detects that it has received input data.

[0025] First, the input unit 111 receives input data related to predetermined consumption behavior from an external device or the like that is communicably connected to the data processing device 110 (step S11). Upon receiving the input data, the input unit 111 supplies the received input data to the associative learning model 112.

[0026] Next, the associative learning model 112 receives the input data supplied from the input unit 111 (step S12). In other words, the data processing device 110 supplies the input data received by the input unit 111 to the associative learning model 112.

[0027] Next, the associative learning model 112 outputs prospective customer data related to consumer behavior as an output for the input data. In other words, the data processing device 110 receives the prospective customer data from the associative learning model 112 as an output (step S13).

[0028] Next, the output unit 113 outputs the prospective customer data received from the federated learning model 112 to a predetermined output destination (step S14). The output destination is, for example, an external device that received the input data. Once the output unit 113 outputs the output data, the data processing device 110 completes the series of processes.

[0029] The data processing device 110 has been described above. With the above configuration, the data processing device 110 can provide a data processing device that contributes to marketing activities in a simple and suitable manner.

[0030] Next, with reference to Figure 3, we will describe the apparatus for generating the federated learning model 112 of the data processing device 110. Figure 3 is a block diagram of the federated learning model generation apparatus according to Embodiment 1.

[0031] The associative learning model generation device 100 may be configured, for example, by a computer or dedicated equipment. The associative learning model generation device 100 has a local learning model acquisition unit 101 and an associative learning model generation unit 102.

[0032] The local learning model acquisition unit 101 acquires multiple different local learning models that have learned the relationship between multiple customer groups generated from customer data owned by multiple businesses and the corresponding consumer behavior of those businesses. The local learning model acquisition unit 101 acquires local learning models from each business, for example, by communicating with the computers of each business that possess local learning models.

[0033] The federated learning model generation unit 102 combines at least a portion of the acquired local learning models. As a result, the federated learning model generation unit 102 receives predetermined customer consumption behavior as input data and generates a federated learning model 112 that outputs prospective customer data for the input data.

[0034] Referring to Figure 4, the processes executed by the federated learning model generation device 100 will be described. Figure 4 is a flowchart of the federated learning model generation method according to Embodiment 1. The flowchart shown in Figure 4 starts, for example, when the federated learning model generation device 100 detects that it has acquired a local learning model.

[0035] First, the local learning model acquisition unit 101 acquires multiple different local learning models that have learned the relationship between multiple customer groups generated from customer data owned by multiple businesses and the corresponding consumer behavior of the businesses (step S101). The local learning model acquisition unit 101 then supplies the acquired local learning models to the federated learning model generation unit 102.

[0036] Next, the federated learning model generation unit 102 combines at least a portion of the local learning models acquired by the local learning model acquisition unit 101 to generate a federated learning model (step S102). The federated learning model 112 generated by the federated learning model generation unit 102 is configured to accept predetermined customer consumption behavior as input data and to output prospective customer data for the input data. Once the federated learning model generation unit 102 has generated the federated learning model 112, the federated learning model generation device 100 completes the series of processes.

[0037] The associative learning model generation device 100 has been described above. According to the above configuration, an associative learning model and its generation device that contribute to marketing activities in a simple and suitable manner can be provided.

[0038] Embodiment 1 has been described above. The data processing device 110 and the federated learning model generation device 100 may be separate devices or may be included in a single device or system.

[0039] Each of the data processing unit 110 and the federated learning model generation unit 100 has a processor and a storage device, although these are not shown in the diagram. The storage devices of the data processing unit 110 and the federated learning model generation unit 100 include, for example, non-volatile memory such as flash memory or SSD (Solid State Drive). In this case, the storage device stores a computer program (hereinafter also simply referred to as "the program") for executing the above-described method. The processor loads the computer program from the storage device into a buffer memory such as DRAM (Dynamic Random Access Memory) and executes the program.

[0040] Each component of the data processing device 110 and the federated learning model generation device 100 may be implemented with dedicated hardware. Furthermore, some or all of each component may be implemented by general-purpose or dedicated circuits, processors, etc., or combinations thereof. These may be implemented by a single chip or by multiple chips connected via a bus. Some or all of each component of each device may be implemented by a combination of the aforementioned circuits, etc., and programs. Furthermore, a CPU (Central Processing Unit), GPU (Graphics Processing Unit), FPGA (field-programmable gate array), etc., can be used as the processor. Note that the descriptions of the configurations described herein may also apply to other devices or systems described below in this disclosure.

[0041] Furthermore, if some or all of the components of the data processing device 110 and the federated learning model generation device 100 are implemented by multiple information processing devices or circuits, these multiple information processing devices or circuits may be centrally located or distributed. For example, the information processing devices or circuits may be implemented in a form in which each is connected via a communication network, such as a client-server system or a cloud computing system. Also, the functions of the information processing device 10 may be provided in SaaS (Software as a Service) format. Moreover, the above method may be stored on a computer-readable medium in order for a computer to execute the above method.

[0042] As described above, this embodiment provides a federative learning model that can easily and effectively contribute to marketing activities, a device for generating the same, and a data processing device using this federative learning model.

[0043] <Embodiment 2> Next, the information processing system will be described with reference to Figure 5. Figure 5 is a block diagram of the data processing system 1 according to Embodiment 2. The data processing system 1 mainly consists of a data processing device 120 and a plurality of local data processing devices 200.

[0044] The data processing device 120 shown in Figure 5 is communicated with each of the two local data processing devices 200 via network N1. One of the local data processing devices 200 is owned by operator A. The other local data processing device 200 is owned by operator B.

[0045] Business A is, for example, a car dealership. Business A uses a local data processing device 200 that it owns for its own business. For example, customers P1 and P2 visit Business A regarding the purchase of a car. As a result, Business A has obtained the personal information of customers P1 and P2.

[0046] Business operator B is a financial services provider that handles, for example, certain financial services. Business operator B uses its own local data processing device 200 for its own business. For example, customers P2 and P3 visit business operator B in connection with a service contract. As a result, business operator B has obtained the personal information of customers P2 and P3.

[0047] In the situation described above, businesses A and B manage customer personal information (also referred to as customer data) in the local data processing device 200. Customer data includes personal information such as the customer's name and address, as well as information about the customer's consumer behavior, such as the products they have purchased. Businesses A and B each generate statistical data that does not contain personal information from the customer data they possess, and generate a local learning model from this statistical data. Businesses A and B each supply the local learning model they have generated to the data processing device 120 via the network N1.

[0048] When the data processing device 120 receives local learning models from both business operator A and business operator B, it combines the received local learning models to generate a federated learning model. The data processing device 120 then uses the generated federated learning model to perform predetermined data processing. Specifically, the data processing device 120 accepts data on customer consumption behavior as input data. The data processing device 120 then outputs prospective customer data for the received input data.

[0049] The data processing device 120 will be further described with reference to Figure 6. Figure 6 is a block diagram of the data processing device 120 according to Embodiment 2. The data processing device 120 mainly consists of a federated learning model generation device 100, an input unit 111, a federated learning model 112, an output unit 113, and a communication unit 114.

[0050] The associative learning model generation device 100 has the same functions and configuration as described in Embodiment 1. That is, the associative learning model generation device 100 mainly includes a local learning model acquisition unit 101 and an associative learning model generation unit 102.

[0051] The local learning model acquisition unit 101 in this embodiment acquires local learning models from local data processing devices 200 owned by business operators A and B, respectively. The local learning model acquisition unit 101 supplies the acquired local learning models to the federated learning model generation unit 102.

[0052] The federated learning model generation unit 102 in this embodiment generates a federated learning model 112 by combining the local learning models of business operator A and business operator B. More specifically, the federated learning model generation unit 102 generates a federated learning model by combining at least a portion of the features extracted from each of the multiple local learning models.

[0053] The input unit 111, the associative learning model 112, and the output unit 113 have the same functions and configurations as those described in Embodiment 1. The input unit 111 receives input data from, for example, event participant A or business operator B.

[0054] The associative learning model 112 is an associative learning model generated by the associative learning model generation device 100. That is, the associative learning model 112 according to this embodiment includes at least a portion of the features of the local learning model generated by business operator A and at least a portion of the features of the local learning model generated by business operator B.

[0055] The output unit 113 outputs the prospective customer data output by the associative learning model 112 to a predetermined output destination. The predetermined output destination is the source of the input data received by the input unit 111. For example, if input data is received from business A, the output unit 113 outputs the output data to business A. If input data is received from business B, the output unit 113 outputs the output data to business B.

[0056] The communication unit 114 includes an interface for connecting to the network N1. That is, the communication unit 114 receives various data from the local data processing unit 200 via the network N1. The communication unit 114 supplies the received data to each component of the data processing unit 120. The communication unit 114 also receives various data from each component of the data processing unit 120 and supplies the received data to the local data processing unit 200 via the network N1.

[0057] The local data processing device 200 will be described with reference to Figure 7. Figure 7 is a block diagram of the local data processing device according to Embodiment 2. The local data processing device 200 mainly consists of a customer information acquisition unit 201, a business customer data generation unit 202, a local learning model generation unit 203, a communication unit 204, an operation reception unit 205, an information presentation unit 206, a local learning model 210, and a storage unit 220.

[0058] The customer information acquisition unit 201 acquires customer information from the business operator. This customer information may be information received via the operation reception unit 205 or information received via the communication unit 204. The customer information acquisition unit 201 supplies the acquired customer information to the business operator customer data generation unit 202.

[0059] The business customer data generation unit 202 receives customer information from the customer information acquisition unit 201 and generates business customer data 221 in a predetermined format from the received customer information. The business customer data 221 is statistical data that does not include specific personal information. The business customer data generation unit 202 stores the generated business customer data 221 in the storage unit 220.

[0060] The local learning model generation unit 203 generates a local learning model 210 from the business customer data 221 generated by the business customer data generation unit 202. At this time, the local learning model generation unit 203 generates the learning model using a portion of the business customer data 221 as input data and the other portion of the business customer data 221 as output data.

[0061] In generating learning models, various methods known in the field of machine learning can be employed. For example, a local learning model may have a structure called a decision tree, or it may be composed of a predetermined neural network.

[0062] The communication unit 204 includes an interface for the local data processing unit 200 to connect to the network N1. That is, the communication unit 204 receives various data from the data processing unit 120 via the network N1. The communication unit 204 supplies the received data to each component of the local data processing unit 200. The communication unit 204 also receives various data from each component of the local data processing unit 200 and supplies the received data to the data processing unit 120 via the network N1.

[0063] The operation reception unit 205 receives predetermined operations performed by an administrator managing the local data processing device 200. These predetermined operations include information and instructions input via input devices such as switches, buttons, keyboards, mice, touch panels, or remote controllers. More specifically, for example, the operation reception unit 205 receives customer information via an input device operated by the administrator and supplies the received customer information to the customer information acquisition unit 201. Alternatively, the operation reception unit 205 receives predetermined input data related to the local learning model from the administrator. Upon receiving the input data, the operation reception unit 205 supplies this input data to the local learning model 210. Alternatively, the operation reception unit 205 receives predetermined input data related to the federated learning model from the administrator. Upon receiving the input data, the operation reception unit 205 supplies this input data to the data processing device 120 via the communication unit 204.

[0064] The information display unit 206 is a means for presenting information related to various processes performed by the local data processing device 200 to the administrator of the local data processing device 200. Specifically, for example, the information display unit 206 is a display device including a liquid crystal panel or organic electroluminescence. The information display unit 206 may also present predetermined information to other devices (computers, smartphones, etc.) via the communication unit 204.

[0065] The local learning model 210 is a learning model generated by the local learning model generation unit 203. The local learning model 210 is a learning model trained using the business customer data 221. The local learning model 210 may, for example, take a predetermined customer group as input and output the consumption behavior corresponding to this customer group. The local learning model 210 may, for example, take a business's consumption behavior as input and output the customer group corresponding to this consumption behavior.

[0066] The storage unit 220 is a storage device that includes non-volatile memory such as an HDD (Hard Disk Drive), SSD (Solid State Drive), or flash memory. The storage unit 220 stores the business customer data 221 generated by the business customer data generation unit 202. The business customer data 221 includes a common area shared by multiple businesses and a unique area shared by each of the multiple businesses.

[0067] Next, the business customer data generation unit 202 will be described with reference to Figure 8. Figure 8 is a diagram showing the processing of the business customer data generation unit 202. Figure 8 shows the customer information D10 input to the business customer data generation unit 202 and the customer statistics data D11 output by the business customer data generation unit 202.

[0068] Customer information D10 includes the customer's name, address, age, occupation, and other personal information, and is unique to each customer. In addition to the above items, customer information D10 may also include family structure, marital status, hobbies, preferences, behavioral history, purchase history, etc.

[0069] Customer statistics data D11 is information processed as statistical data by dividing customer information D10 into predetermined categories and groups. Customer statistics data D11 does not include personal names or information that can identify individuals. Customer statistics data D11 is grouped by, for example, a predetermined range of addresses, a predetermined range of ages, a predetermined occupation category, etc. Therefore, for example, customer group #0001 includes multiple customers that fall under the category of customer group #0001.

[0070] The business customer data generation unit 202 takes in customer information D10 and generates customer statistical data D11 from the taken customer information D10. The customer statistical data D11 is categorized by a common definition that spans multiple businesses. Therefore, "customer group #0001" is data that is categorized by a common definition across multiple different businesses. In addition, the customer statistical data D11 does not contain personal information. Therefore, the data processing system 1 can use the customer statistical data D11 across multiple businesses without providing customer information managed by each business to others.

[0071] Next, we will explain the business customer data with reference to Figure 9. Figure 9 is a diagram showing the business customer data 221. The business customer data 221 has the customer statistics data D11, which was explained in Figure 8, as a common area. The common area refers to the area of ​​data set in the data processing system 1 by a definition common to multiple different businesses. The common area includes customer groups.

[0072] The business customer data 221 includes both common and unique areas. The unique area contains data specific to the business of the business and is linked to the customer statistics data D11 in the common area. For example, in customer group #0001 shown in Figure 9, the purchase indicator for product type 1 is 20%, and the purchase indicator for product type 2 is 42%, etc. The purchase indicator is a statistic of the consumption behavior of customers belonging to the customer group.

[0073] The purchase indicator may be a probability as shown in Figure 9, or it may be a rank or score, as long as the associative learning model generator 100 can generate an associative learning model. In other words, the business customer data 221 includes customer groups included in the business customer data and the consumption behavior associated with those customer groups.

[0074] The local learning model generation unit 203 in this embodiment generates a local learning model 210 using business customer data 221 as shown in Figure 9. That is, the local learning model generation unit 203 generates a local learning model that has been trained based on business customer data that includes common areas shared by multiple businesses and unique areas shared by each of the multiple businesses.

[0075] Alternatively, the local learning model generation unit 203 generates a local learning model trained on business customer data, including customer groups as a common area. The local learning model generation unit 203 also generates a local learning model that has data classifying customers into multiple groups based on predetermined attributes of the customer as a common area.

[0076] Next, we will explain the processing of the local learning model 210 with reference to Figure 10. Figure 10 is a diagram illustrating the processing of the local learning model. The local learning model 210 shown in Figure 10 is a local learning model owned by business operator A. The local learning model 210 accepts data such as the type of automobile being type 1 and the color being C2 as input data regarding customer consumption behavior. When the local learning model 210 receives the input data, it outputs a purchase index for each customer group as corresponding output data. In this way, when the local learning model 210 receives the business operator's consumption behavior as input, it outputs customer groups corresponding to that consumption behavior.

[0077] The processing performed by the local learning model 210 is not limited to the above. For example, the local learning model 210 may take a customer group as input and output the consumption behavior corresponding to that customer group.

[0078] Next, the process of generating a federated learning model will be described with reference to Figure 11. Figure 11 is a sequence diagram showing the process performed by the federated learning model generation system. In Figure 11, the local data processing device 200 owned by business operator A, the local data processing device 200 owned by business operator B, and the federated learning model generation device 100 are shown as the federated learning model generation system 2, which is the configuration for generating a federated learning model.

[0079] In other words, the federated learning model generation system 2 includes multiple local data processing devices 200 and a federated learning model generation device 100. The local data processing devices 200 generate a local learning model 210 for each business from the business customer data managed by each business. The federated learning model generation device 100 acquires the local learning models 210 from each of the multiple businesses and generates a federated learning model 112.

[0080] In the sequence diagram shown in Figure 11, first, the local data processing device 200 of business operator A acquires customer information (step S201). Next, the local data processing device 200 of business operator A generates business customer data from the acquired customer information (step S202). Furthermore, the local data processing device 200 of business operator A generates a local learning model from the business customer data (step S203).

[0081] The local data processing device 200 of business operator B, similar to that of business operator A, first acquires customer information (step S211) and then generates business operator customer data (step S212). Then, the local data processing device 200 of business operator B generates a local learning model from the generated business operator customer data (step S213).

[0082] Next, the federated learning model generator 100 acquires the local learning model of business operator A and the local learning model of business operator B, respectively (step S221). Then, the federated learning model generator 100 combines the acquired local learning models to generate a federated learning model (step S222).

[0083] The federated learning model generation system 2 has been described above with reference to Figure 11. As described above, the federated learning model generation device 100 works in conjunction with the local data processing device 200 to constitute the federated learning model generation system 2.

[0084] Next, we will explain the configuration of the associative learning model. Figure 12 is a diagram illustrating the configuration of the associative learning model. The associative learning model 112 shown in Figure 12 is an example that is constructed using a decision tree. When the associative learning model 112 shown in Figure 12 receives input data, it processes the input data using the multiple nodes that make up the model. The associative learning model 112 then outputs the output data shown at the bottom of Figure 12. Note that the diagram in Figure 12 is a schematic representation of the data structure of a decision tree, and the number of nodes, layers, etc., are not limited to the configuration in Figure 12.

[0085] The federated learning model 112 has elements E1 and E2. Element E1 includes at least a portion of the features of the local learning model 210 owned by business operator A. Element E2 includes at least a portion of the features of the local learning model 210 owned by business operator B. Thus, the federated learning model 112 is generated by extracting at least a portion of the features owned by each local learning model and combining the extracted features. In other words, the federated learning model generation unit 102 of the federated learning model generation device 100 generates the federated learning model 112 by combining at least a portion of the features extracted from each of the multiple local learning models.

[0086] Although Figure 12 shows an example of a decision tree, the associative learning model 112 of the data processing system 1 is not limited to a decision tree. The associative learning model 112 may also have a predetermined perceptron or neural network as its constituent elements.

[0087] Next, with reference to Figure 13, the processes performed by the associative learning model 112 will be explained. Figure 13 is a diagram illustrating the processes of the associative learning model. The associative learning model 112 shown in Figure 13 is generated by associating the local learning model possessed by business operator A with the local learning model possessed by business operator B. Therefore, the associative learning model 112 can accept data on consumer behavior related to business operator A's business as input data, and can also accept data on consumer behavior related to business operator B's business.

[0088] In the upper part of Figure 13, the associative learning model 112 receives input data regarding the consumption behavior of automobiles related to the business of business operator A. In this case, the associative learning model 112 outputs a purchase index for each customer group as output data corresponding to the type and specifications of the automobiles received.

[0089] In the lower part of Figure 13, the associative learning model 112 receives input data regarding consumer behavior for financial services related to the business of business operator B. In this case, the associative learning model 112 outputs a purchase index for each customer group as output data corresponding to the specifications of the financial services received.

[0090] In this way, the federated learning model 112 accepts input data regarding the consumer behavior of multiple businesses and outputs prospective customer data corresponding to the accepted input data. When the federated learning model 112 accepts input data relating to a predetermined consumer behavior, it is configured to output prospective customer data for that consumer behavior. At this time, the business customer data held by business A, the business customer data held by business B, and the prospective customer data output by the federated learning model 112 each include customer groups common to them. The federated learning model 112 outputs prospective customer data that includes at least a portion of the multiple customer groups. At this time, the federated learning model 112 may also output prospective customer data that includes estimated purchase indicators showing the trends in the consumer behavior of the customer groups.

[0091] Next, with reference to Figure 14, the processes performed by the data processing system 1 will be described. Figure 14 is a sequence diagram showing the processes performed by the data processing system 1. The sequence diagram in Figure 14 shows the flow in which a local data processing device 200 owned by business operator B, a financial services provider, receives input data for the federated learning model 112, and in response, the federated learning model 112 outputs output data, which is the response, to the local data processing device 200.

[0092] First, the local data processing device 200 receives a predetermined query from a user who is a user of the data processing system 1 (step S301). The query is, for example, "What are the potential customers for automobile type 1?"

[0093] Next, the local data processing device 200 generates input data to be sent to the data processing device 120 from the contents of the query, and sends the generated input data to the data processing device 120 (step S302).

[0094] Next, the data processing device 120 acquires input data from the local data processing device 200 (step S11) and supplies the acquired input data to the federated learning model 112 (step S12).

[0095] When the associative learning model 112 receives input data, it outputs prospective customer data, which is output data corresponding to the received input data. The data processing device 120 receives the output data from the associative learning model 112 (step S13) and sends the received output data to the local data processing device 200 (step S14).

[0096] Next, the local data processing device 200 acquires output data from the data processing device 120 (step S303), and uses the acquired output data to present a response to the user (step S304). The response includes the customer group.

[0097] The above describes the processing performed by the data processing system 1. The data processing system 1 uses an associative learning model 112, which receives input regarding consumer behavior, to output prospective customers. The associative learning model does not use personal information and handles information on customer groups grouped according to predetermined criteria. Therefore, through associative learning, the data processing system 1 can link data from unique domains held by different businesses via data in a common domain. As a result, the data processing system 1 can output prospective customer data for each business using data in a common domain that spans multiple businesses. In other words, the data processing system 1 can estimate prospective customer data related to each business's business with high accuracy.

[0098] As shown in Figure 14, the data processing system 1 can send queries to the data processing device 120 using consumer behavior related to businesses different from its own as input data, and obtain prospective customer data as a response to these queries. This allows businesses using the data processing system 1 to obtain prospective customer data on consumer behavior that would be difficult to obtain from their own businesses. Therefore, the data processing system 1 can contribute to expanding the scope of businesses' marketing activities.

[0099] The administrator managing the data processing device 120 may be business operator A or business operator B, or it may be a person different from business operator A and business operator B. The federated learning model 112 may be owned by the administrator of the data processing device 120.

[0100] As described above, Embodiment 2 provides a federative learning model that contributes to marketing activities in a simple and suitable manner, a device for generating the same, and a data processing device using this federative learning model.

[0101] <Embodiment 3> Next, Embodiment 3 will be described. Embodiment 3 differs from the embodiments described above in that it has the function of updating the local learning model and the federated learning model.

[0102] Figure 15 is a block diagram of the data processing device 130 according to Embodiment 3. The data processing device 130 shown in Figure 15 differs from the data processing device 120 described above in that it has an update control unit 115.

[0103] The update control unit 115 controls the acquisition of the local learning model when the local learning model is updated. Specifically, the update control unit 115 appropriately acquires update information from multiple local data processing devices. When it detects that the local learning model has been updated, it instructs the local learning model acquisition unit 101 to acquire the update data from the local data processing device that has the updated local learning model.

[0104] The update control unit 115 also controls the updating of the federated learning model using the updated local learning model. In other words, the update control unit 115 instructs the federated learning model generation unit 102 to update the federated learning model using the acquired update data.

[0105] Figure 16 is a block diagram of the local data processing device 300 according to Embodiment 3. The local data processing device 300 differs from the local data processing device 200 described above in that it has an update information management unit 207. The local data processing device 300 also stores update information 222 in the storage unit 220.

[0106] The local learning model 210 is updated, for example, when the customer information acquisition unit 201 acquires new customer data. Alternatively, the local learning model 210 may be updated in conjunction with updates to the business customer data 221.

[0107] The update information management unit 207 manages the version information or update date and time of the local learning model 210 by updating the update information 222 stored in the storage unit 220 when the local learning model 210 is updated. The update information management unit 207 may also have a function to notify the data processing device 130 that the local learning model 210 has been updated.

[0108] The update information 222 stored in the memory unit 220 may contain version information of the local learning model 210, or it may contain the latest update date and time of the local learning model 210.

[0109] Next, with reference to Figure 17, the update process performed by the data processing system 1 will be described. Figure 17 is a sequence diagram of the update method according to Embodiment 3.

[0110] First, the update information management unit 207 of the local data processing device 300 determines whether the customer information or business customer data 221 has been updated (step S401). If it is determined that the customer information or business customer data 221 has not been updated (step S401: NO), the local data processing device 300 repeats step S401. If it is determined that the customer information or business customer data 221 has been updated (step S401: YES), the local data processing device 300 updates the local learning model 210 using the updated data (step S402).

[0111] Next, the data processing device 130 retrieves the updated local learning model 210 from the local data processing device 300 (step S403).

[0112] Next, the data processing unit 130 updates the federated learning model 112 using the updated local learning model 210 (step S404).

[0113] The process by which the data processing device 130 updates the federated learning model using the updated local learning model has been described above. The data processing system 1 according to this embodiment can appropriately contribute to marketing activities by appropriately updating the federated learning model 112. Therefore, according to this embodiment, it is possible to provide a federated learning model, a device for generating the same, and a data processing device using this federated learning model that can easily and appropriately contribute to marketing activities.

[0114] <Example hardware configuration> The following describes how each functional configuration of the determination device in this disclosure is realized through a combination of hardware and software.

[0115] Figure 18 is a block diagram illustrating the hardware configuration of a computer. The management device in this disclosure can realize the above-described functions using a computer 500 including the hardware configuration shown in the figure. The computer 500 may be a portable computer such as a smartphone or tablet terminal, or a stationary computer such as a PC. The computer 500 may be a dedicated computer designed to realize each device, or it may be a general-purpose computer. The computer 500 can realize the desired functions by installing a predetermined program.

[0116] Computer 500 has a bus 502, a processor 504, memory 506, a storage device 508, an input / output interface 510 (an interface is also called an I / F (Interface)), and a network interface 512. Bus 502 is a data transmission path for the processor 504, memory 506, storage device 508, input / output interface 510, and network interface 512 to send and receive data to and from each other. However, the method of connecting the processor 504 and the other components to each other is not limited to bus connection.

[0117] Processor 504 is a variety of processors such as a CPU, GPU, or FPGA. Memory 506 is main memory implemented using RAM (Random Access Memory), etc.

[0118] The storage device 508 is an auxiliary storage device implemented using a hard disk, SSD, memory card, or ROM (Read Only Memory). The storage device 508 stores a program for implementing a desired function. The processor 504 reads this program into memory 506 and executes it to implement each functional component of each device.

[0119] The input / output interface 510 is an interface for connecting the computer 500 with input / output devices. For example, input devices such as keyboards and output devices such as display devices are connected to the input / output interface 510.

[0120] Network interface 512 is an interface for connecting computer 500 to a network.

[0121] The above describes examples of hardware configurations in this disclosure, but the embodiments described above are not limited thereto. This disclosure can also be implemented by having a processor execute a computer program to perform any processing.

[0122] In the examples described above, the program includes a set of instructions (or software code) that, when loaded into a computer, cause the computer to perform one or more of the functions described in the embodiments. The program may be stored on a non-temporary computer-readable medium or a physical storage medium. Examples, but not limited to, include random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drive (SSD) or other memory technologies, CD-ROM, digital versatile disc (DVD), Blu-ray® disc or other optical disc storage, magnetic cassette, magnetic tape, magnetic disk storage or other magnetic storage devices. The program may be transmitted over a temporary computer-readable medium or a communication medium. Examples, but not limited to, include temporary computer-readable medium or a communication medium that includes electrical, optical, acoustic or other forms of propagating signals.

[0123] Although the present invention has been described above with reference to embodiments, the present invention is not limited thereto. Various modifications to the structure and details of the present invention can be made that are understandable to those skilled in the art within the scope of the invention.

[0124] Some or all of the above embodiments may also be described as follows, but are not limited to the following: (Note 1) An input unit that receives input data related to a predetermined consumer behavior, A federated learning model is generated by learning the relationship between multiple customer groups generated from customer data owned by a business and the corresponding consumer behavior of the business, thereby creating a federated learning model in which at least a portion of the local learning models generated for each of the multiple different businesses are federated, and which is configured to output prospective customer data for the input data, An output unit that outputs the aforementioned prospective customer data, A data processing device equipped with the following features. (Note 2) The aforementioned associative learning model is Outputting prospective customer data that includes at least a portion of multiple customer groups, The data processing device described in Appendix 1. (Note 3) The aforementioned associative learning model is Outputs prospective customer data including estimated purchase indicators that show the trends in the consumer behavior of the customer group. The data processing device described in Appendix 2. (Note 4) The aforementioned associative learning model is When the aforementioned customer group is received as input, the local learning model that outputs the consumption behavior corresponding to the aforementioned customer group is generated by associating it with the customer group, The data processing device described in Appendix 3. (Note 5) The aforementioned associative learning model is When the aforementioned business operator's consumer behavior is received as input, the local learning model that outputs the customer group corresponding to the consumer behavior is generated by combining the following: The data processing device described in Appendix 3. (Note 6) The aforementioned associative learning model is At least some of the features extracted from each of the multiple local learning models are combined to generate: The data processing device described in Appendix 1. (Note 7) The aforementioned associative learning model is The local learning models, which are trained based on the customer data of the businesses, each of which includes a common domain shared by multiple businesses and a unique domain shared by each of the businesses, are generated by combining these local learning models. A data processing device as described in any one of the appendices 1 to 6. (Note 8) The aforementioned associative learning model is The local learning model, which has been trained based on the business customer data including the customer group, is generated by combining the above common domains. The data processing device described in Appendix 7. (Note 9) The aforementioned associative learning model is The aforementioned customer group is generated by combining the local learning model, which has as a common area data that classifies the customers into multiple groups based on predetermined attributes of the customers, The data processing device described in Appendix 8. (Note 10) A local data processing device that generates the local learning model for each of the aforementioned businesses based on customer data managed by each of the aforementioned businesses, An update control unit that controls the acquisition of the local learning model when the local learning model is updated, A data processing device described in Appendix 1 updates the federated learning model using the acquired local learning model, Equipped with, Data processing system. (Note 11) Computers We accept input data regarding specified consumer behavior. The input data is supplied to a federated learning model generated by combining multiple different local learning models that have learned the relationship between the business customer data and consumer behavior based on the business customer data owned by a specified business operator. The associative learning model receives prospective customer data for the consumer behavior as output for the input data, Output the received prospective customer data. Data processing method. (Note 12) We accept input data regarding specified consumer behavior. The input data is supplied to a federated learning model generated by combining multiple different local learning models that have learned the relationship between the business customer data and consumer behavior based on the business customer data owned by a specified business operator. The associative learning model receives prospective customer data for the consumer behavior as output for the input data, Output the received prospective customer data. To have a computer execute a data processing method. program.

[0125] This application claims priority based on Japanese Patent Application No. 2022-089721, filed on 1 June 2022, and incorporates all of its disclosures herein. [Industrial applicability]

[0126] This disclosure can be used, for example, as a data processing device for businesses to conduct marketing activities. [Explanation of symbols]

[0127] 1. Data Processing System 2. Associative Learning Model Generation System 100 Associative Learning Model Generator 101 Local Learning Model Acquisition Unit 102 Associative Learning Model Generation Unit 110 Data Processing Devices 111 Input Section 112 Associative Learning Models 113 Output section 114 Communications Department 115 Update Control Unit 120 Data Processing Devices 130 Data Processing Devices 200 Local Data Processing Units 201 Customer Information Acquisition Department 202 Business Customer Data Generation Department 203 Local Learning Model Generation Unit 204 Communications Department 205 Operation Reception Section 206 Information Presentation Department 207 Update information management department 210 Local Learning Models 220 Storage section 221 Business Customer Data 222 Update information 300 Local Data Processing Units 500 Computers 504 Processors 506 memory 508 Storage Devices 510 Input / Output Interfaces 512 Network Interfaces A10 Readable area G10 guidance image G11 First Guidance Image G12 Second-guided image N1 Network P1 Customer P2 Customer D10 Customer Information D11 Customer Statistics Data

Claims

1. An input means for receiving input data related to a predetermined consumer behavior, A federated learning model is generated by learning the relationship between multiple customer groups generated from customer data owned by a business and the corresponding consumer behavior of the business, thereby creating a federated learning model in which at least a portion of the local learning models generated for each of the multiple different businesses are federated, and which is configured to output prospective customer data for the input data, Output means for outputting the prospective customer data, A data processing device equipped with the following features.

2. The aforementioned associative learning model is Outputting prospective customer data that includes at least a portion of multiple customer groups, The data processing device according to claim 1.

3. The aforementioned associative learning model is Outputs prospective customer data including estimated purchase indicators that show the trends in the consumer behavior of the customer group. The data processing device according to claim 2.

4. The aforementioned associative learning model is When the aforementioned customer group is received as input, the local learning model that outputs the consumption behavior corresponding to the aforementioned customer group is generated by associating it with the customer group, The data processing device according to claim 3.

5. The aforementioned associative learning model is When the aforementioned business operator's consumer behavior is received as input, the local learning model that outputs the customer group corresponding to the consumer behavior is generated by combining the following: The data processing device according to claim 3.

6. The aforementioned associative learning model is At least some of the features extracted from each of the multiple local learning models are combined to generate: The data processing device according to claim 1.

7. The aforementioned associative learning model is The local learning models, which are trained based on the customer data of the businesses, each of which includes a common domain shared by multiple businesses and a unique domain shared by each of the businesses, are generated by combining these local learning models. A data processing device according to any one of claims 1 to 6.

8. A local data processing device that generates the local learning model for each of the aforementioned businesses based on customer data managed by each of the aforementioned businesses, An update control unit that controls the acquisition of the local learning model when the local learning model is updated, A data processing device according to claim 1, which updates the federated learning model using the acquired local learning model, Equipped with, Data processing system.

9. Computers We accept input data regarding specified consumer behavior. The input data is supplied to a federated learning model generated by combining multiple different local learning models that have learned the relationship between the business customer data and consumer behavior based on the business customer data owned by a specified business operator. The associative learning model receives prospective customer data for the consumer behavior as output for the input data, Output the received prospective customer data. Data processing method.

10. We accept input data regarding specified consumer behavior. The input data is supplied to a federated learning model generated by combining multiple different local learning models that have learned the relationship between the business customer data and consumer behavior based on the business customer data owned by a specified business operator. The associative learning model receives prospective customer data for the consumer behavior as output for the input data, Output the received prospective customer data. To have a computer execute a data processing method. program.

Images