Information processing device, information processing method, and information processing program

Abstract

We provide a system that makes data sphere member companies want to release their own data. [Solution] The information processing device includes an acquisition unit that acquires anonymous information that has been vectorized from customer data from each of a plurality of companies, a learning unit that performs machine learning on AI using the anonymous information from each company, an estimation unit that uses the AI ​​that has undergone machine learning to estimate customers from among the plurality of companies that are suitable for the objectives of a specific company, and a provision unit that provides information based on the estimation results by the estimation unit.

Description

[Technical Field]

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program. [Background technology]

[0002] A technology has been disclosed that allows companies to safely utilize data they hold in a data sharing system without disclosing the contents to other companies. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent No. 6803598 Summary of the Invention [Problem to be solved by the invention]

[0004] However, the above-mentioned conventional technology merely encrypts and processes sensitive data provided by organizations participating in the data sharing system using a specified encryption method (an encryption method that allows calculations, including searches and analyses, to be performed while the data remains encrypted), while the keys used for encryption and decryption are managed by the organizations that provided the sensitive data.

[0005] Furthermore, forcing all clients to integrate their IDs using a common ID (ID Connect) would pose the problem of not increasing the number of companies joining the data sphere. Even if companies join the data sphere, it would be unrealistic to expect each company to send all of their customer data (personal information) to the mother server (many companies would likely not agree to this). Therefore, it is necessary to design a system that prevents companies joining the data sphere from releasing too much of their own data and from relying solely on other companies' data for marketing purposes. In other words, it is necessary to have a system that prevents free riding and encourages companies to release their own data.

[0006] The present application has been made in light of the above, and aims to provide a system that will encourage data sphere member companies to release their own data. [Means for solving the problem]

[0007] The information processing device according to the present application includes an acquisition unit that acquires anonymous information obtained by vectorizing customer data from each of a plurality of companies; Perform machine learning using the above as input to generate a customer evaluation model a learning unit for performing the machine learning; The customer evaluation model generated by , a specific company among the plurality of companies may enter anonymous information from the The system includes an estimation unit that estimates customers, and a provision unit that provides information based on the estimation result by the estimation unit. The learning unit performs machine learning using input obtained by adding anonymous information from each of the companies to anonymous information obtained by vectorizing customer data independently collected by the information processing device separately from each of the companies, and generates the customer evaluation model. It is characterized by: [Effects of the Invention]

[0008] According to one aspect of the embodiment, a mechanism can be provided that makes data sphere member companies want to release their own data. [Brief explanation of the drawings]

[0009] [Figure 1] FIG. 1 is an explanatory diagram showing an overview of an information processing method according to an embodiment. [Figure 2] Figure 2 is an explanatory diagram showing an overview of AI learning on the server side. [Figure 3] FIG. 3 is an explanatory diagram showing an example of learning from campaign objectives. [Figure 4] FIG. 4 is an explanatory diagram showing an overview of vectorization of customer data. [Figure 5] FIG. 5 is an explanatory diagram showing an overview of inference on the server side. [Figure 6] FIG. 6 is a simplified schematic diagram of inference on the server side. [Figure 7] FIG. 7 is an explanatory diagram showing an overview of vector neighborhood customer expansion. [Figure 8] Figure 8 is a conceptual diagram showing the image of the 1st Party Data Zone concept. [Figure 9]FIG. 9 is a diagram illustrating an example of the configuration of the mother server according to the embodiment. [Figure 10] FIG. 10 is a diagram illustrating an example of the user information database. [Figure 11] FIG. 11 is a diagram illustrating an example of the history information database. [Figure 12] FIG. 12 is a diagram illustrating an example of the anonymous information database. [Figure 13] FIG. 13 is a flowchart showing a processing procedure according to the embodiment. [Figure 14] FIG. 14 is a diagram illustrating an example of a hardware configuration. DETAILED DESCRIPTION OF THE INVENTION

[0010] Hereinafter, an information processing device, an information processing method, and an information processing program according to the present application (hereinafter referred to as "embodiments") will be described in detail with reference to the drawings. Note that the information processing device, the information processing method, and the information processing program according to the present application are not limited to these embodiments. Furthermore, the same components in the following embodiments will be denoted by the same reference numerals, and duplicated descriptions will be omitted.

[0011] [1. Overview of information processing method] First, an overview of an information processing method performed by an information processing device according to an embodiment will be described with reference to Fig. 1. Fig. 1 is an explanatory diagram showing an overview of an information processing method according to an embodiment. Note that Fig. 1 explains an example in which a system is provided that encourages data sphere member companies to release their own data.

[0012] 1, an information processing system 1 according to the embodiment includes a client terminal 10 and a mother server 100. The client terminal 10 and the mother server 100 are connected to each other via a network N so as to be able to communicate with each other via a wired or wireless connection. The network N is, for example, a LAN (Local Area Network) or a WAN (Wide Area Network) such as the Internet.

[0013] In addition, such a client terminal 10 can connect to the network N via a wireless communication network such as LTE (Long Term Evolution), 4G (4th Generation), or 5G (5th Generation: 5th generation mobile communication system), or via short-range wireless communication such as Bluetooth (registered trademark) or wireless LAN (Local Area Network), and communicate with the mother server 100.

[0014] The client terminal 10 is an information processing device used by each client company (a data zone member company). For example, the client terminal 10 is a personal computer (PC), a server device, or a smart device such as a smartphone or a tablet terminal. It is preferable that each client company is a company in a different industry. Of course, the data zone member companies may include competing companies (companies in the same industry).

[0015] The client terminal 10 of each company collects customer data on each user U (customer) of the company. For example, the client terminal 10 of each company collects, as customer data, registration information, store visit records, product purchase history / service usage history, questionnaire responses, etc. of each user U who has purchased the company's products / used the company's services.

[0016] In the example shown in FIG. 1, for the purpose of identification, the client terminal 10 used by company A is referred to as "client terminal 10A," the client terminal 10 used by company B is referred to as "client terminal 10B," and the client terminal 10 used by company C is referred to as "client terminal 10C."

[0017] The mother server 100 is, for example, a PC, a server device, a mainframe, or a workstation. The mother server 100 may be realized by cloud computing. The mother server 100 may also collect customer data (mother-side data) on each user U as a customer who uses the mother-side service, separately from the client terminals 10 of each company.

[0018] For example, the mother server 100 may cooperate with the terminal device of each user U and provide API (Application Programming Interface) services for various applications (hereinafter referred to as apps) and various data to the terminal device of each user U.

[0019] Furthermore, the mother server 100 may be an information processing device that provides some kind of online web service to the terminal device of each user U. For example, the mother server 100 may provide such web services as internet connection, search service, SNS (Social Networking Service), e-commerce, electronic payment, online games, online banking, online trading, hotel and ticket reservations, video and music distribution, news, maps, route search, route guidance, line information, operation information, and weather forecast. In practice, the mother server 100 may cooperate with various servers that provide the above-mentioned web services, and may act as an intermediary for the web services or may be responsible for processing the web services.

[0020] The mother server 100 can acquire user information about the user U. For example, the mother server 100 acquires information about the attributes of the user U, such as the gender, age, and area of ​​residence of the user U. The mother server 100 then stores and manages the information about the attributes of the user U together with identification information (such as a user ID) that indicates the user U.

[0021] The mother server 100 also acquires various types of history information (log data) indicating the behavior of the user U from the user U's terminal device or from various servers based on the user ID, etc. For example, the mother server 100 acquires the location history, which is a history of the user U's location and date and time, from the terminal device. The mother server 100 also acquires the search history, which is a history of search queries entered by the user U, from a search server (search engine). The mother server 100 also acquires the browsing history, which is a history of content viewed by the user U, from a content server. The mother server 100 also acquires the purchase history (payment history), which is a history of the user U's product purchases and payment processes, from an e-commerce server or a payment processing server. The mother server 100 may also acquire the listing history and sales history, which are a history of the user U's listings on the marketplace, from an e-commerce server or a payment server. The mother server 100 also acquires the posting history, which is a history of the user U's posts, from a posting server or SNS server that provides a word-of-mouth posting service.

[0022] The number of devices included in the information processing system 1 shown in Fig. 1 is not limited to that shown in the figure. For example, in Fig. 1, for the sake of simplicity, only three client terminals 10 are shown, but this is merely an example and is not limiting, and the number may be four or more. Furthermore, the mother server 100 may also be distributed across multiple server devices.

[0023] For example, as shown in FIG. 1, the mother server 100 distributes AI (Artificial Intelligence) for anonymizing (vectorizing) customer data to the client terminals 10 (10A, 10B, 10C, etc.) of each company joining the data zone in advance (such as when each company joins the data zone) via the network N (step S1).

[0024] At this time, the mother server 100 may customize the AI ​​for each company. In other words, the mother server 100 may distribute a different AI to each company. Since the content of the information acquired as customer data may differ depending on the industry and business type of the company, it is possible to change the specifications of the AI ​​according to the content of the acquired customer data, and it is not necessary to use the same AI for all companies. It is sufficient for the mother server 100 (server side) to understand and manage the specifications of the AI ​​distributed to each company.

[0025] Next, the mother server 100 collects anonymous information obtained by anonymizing (vectorizing) customer data of each company using AI from the client terminals 10 of each company that is a member of the data zone (step S2).

[0026] In this embodiment, the client terminal 10 of each company acquires customer data related to its customers and generates anonymous information by converting (vectorizing) the acquired customer data into vectors using AI. At this time, the client terminal 10 may perform so-called embedding and generate embedding vectors (embedding vectors) based on the customer data as anonymous information. Then, the client terminal 10 of each company transmits the anonymous information, which is the anonymized (vectorized) customer data, to the mother server 100 in response to a request from the mother server 100 or on its own initiative.

[0027] For example, the client terminal 10 inputs the customer data into a vector conversion model constructed by the mother server 100 (server side) using AI, such as a machine learning technique using a neural network, and converts the customer data into a multidimensional real-valued vector. Alternatively, the client terminal 10 may convert the customer data into a vector in accordance with a predetermined conversion rule (rule-based).

[0028] Anonymous information is irreversible, meaning that the original customer data cannot be restored from the vector values ​​of the anonymous information. Here, "unable to restore" not only refers to the concept that the original data cannot be definitively calculated from the vector values, but also includes the concept that it is difficult to calculate or estimate the original data, i.e., the customer data, such as when it cannot be estimated. In other words, anonymous information cannot be decrypted like encrypted data. For example, when anonymizing (vectorizing) customer data, it can be placed into specific groups or segments based on common attributes, or numbers can be "rounded" by rounding (rounding up or down). Furthermore, customer data and vector values ​​are not necessarily related in a one-to-one relationship, where the same vector value is obtained from the same data. There can also be a many-to-one relationship, where the same vector value is obtained from different data, or a many-to-many relationship, where different vector values ​​are obtained depending on the combination of data, even if some of the data contains the same data. Furthermore, when different AIs are distributed to different companies, the meaning of the vector value itself may differ depending on the company. Therefore, a single vector value is not necessarily based on the same data. Moreover, what the mother server 100 (server side) needs is the vector value, not the customer data that is the basis (source) of the vector value.

[0029] Furthermore, since the vector values ​​of anonymous information and their meanings are defined by the mother server 100 (server side) that distributes the AI, companies and third parties cannot understand what the vector values ​​mean by looking at only the vector values ​​of anonymous information, so even if anonymous information is leaked, customer data will not be leaked, making it safe. Note that a vector will be different each time it is created, even for the same user, and will not necessarily be the same each time.

[0030] Next, the mother server 100 learns the objective setting data as the correct answer using all the anonymous information collected from the client terminals 10 of each company (step S3). For example, the mother server 100 performs machine learning using the anonymous information of all customers of all companies participating in the data zone to generate a customer evaluation model. At this time, the mother server 100 may learn using customer data (mother-side data) independently acquired on the mother side in addition to all the anonymous information. Note that the mother server 100 may generate anonymous information by anonymizing (vectorizing) the mother-side data using AI.

[0031] Next, the mother server 100 performs inference for each company from the learning results (step S4). For example, the mother server 100 generates a list of customers who are likely to purchase a specific product of a specific company based on the results of learning using all anonymous information collected from the client terminals 10 of each company. At this time, the mother server 100 may input the anonymous information from the specific company into a customer evaluation model and obtain as output a list of customers who are likely to purchase a specific product of the specific company.

[0032] Next, the mother server 100 provides the inference results corresponding to each company and information based on the inference results to the client terminal 10 of each company via the network N (step S5). For example, the mother server 100 provides the client terminal 10 of a specific company with a list of customers who are likely to purchase a specific product of the specific company.

[0033] [1-1. AI learning on the server side] Next, AI learning on the mother server 100 (server side) according to this embodiment will be described with reference to Fig. 2. Fig. 2 is an explanatory diagram showing an overview of AI learning on the server side.

[0034] 2, the data domain is configured with a CDP (Customer Data Platform) on the client side and a server on the mother side, that is, a client terminal 10 of each company and a mother server 100.

[0035] In the client-side CDP, each company's client terminal 10 uses AI to convert (vectorize) its own customer data (first-party data) into vectors to generate anonymous information. At this time, each company's client terminal 10 generates a vector for each company's ID as anonymous information. In other words, it generates a combination of each company's customer ID and vector data.

[0036] The customer data held by each company differs from company to company. For example, Company A collects information such as "store visits," "purchases," and "requests and conversations during business negotiations" as customer data. Company B collects information such as "product purchases," "survey responses," and "interaction with online advertisements" as customer data. Company C collects information such as "searches," "exposure to news," and "products purchased" as customer data. However, these are just a few examples.

[0037] Then, the client terminal 10 of each company transmits the generated anonymous information to the mother server 100 (server side) as transmission data in response to a request from the mother server 100 or spontaneously. For security reasons, the client terminal 10 of each company may further encrypt the generated anonymous information before transmitting it to the mother server 100 (server side). Encrypting and transmitting data is itself possible within the scope of common technical knowledge. Also, a closed virtual direct connection line may be established between the mother side and the client side by tunneling. For example, a VPN (Virtual Private Network) may be constructed between the mother side and the client side.

[0038] At this time, each company's client terminal 10 transmits only vector data linked to the common ID as anonymous information. The common ID is a customer ID common to the data domain. The vector data linked to the common ID is vector data of customers who have consented to data provision to the data domain. Throughout, the link between the mother side and the client side is an exchange of vector data only between users who have consented to common ID linkage. Note that what is transmitted is only vector data (anonymous information), and no customer data itself is transmitted. Even vectorized data may be interpreted as personal information, but it is far more secure than raw personal information data. The common ID identifies the same person, but personal information such as place of residence, age, and gender is either not included or is anonymized (vectorized) and therefore cannot be identified.

[0039] In addition, it is not necessary to link the vector data of all customers to a common ID. For example, if a company simply participates in the data sphere without obtaining permission from each individual customer to use their data within the data sphere, it is not necessary to link the common ID to the vector data of the customers.

[0040] In this embodiment, ID integration (ID Connect) using a common ID is "optional." In other words, the system is designed to function as a whole even if there are customers who "do not" use ID integration (ID Connect) using a common ID.

[0041] In addition, the system is designed to achieve the purpose by preventing companies that join the data zone from sending raw customer data (personal information) to the mother server. In other words, raw customer data (personal information) held by each company will not be sent to the mother server 100. Raw customer data (personal information) held by each company will be anonymized (vectorized) and sent to the mother server 100 together with the ID.

[0042] In addition, by introducing distributed AI that uses only one company's own data and anonymizes (vectorizes) one's own customer data, a system will be created in which the accuracy of one's own targeting will not improve unless one company provides a large amount of data in both quality and quantity. This prevents free riding, where one company uses only other companies' data without providing their own data.

[0043] Furthermore, the greater the quantity and quality of a company's customer data, the greater the benefits the company will receive. In other words, the more vector data is linked to a common ID and permission to use data within the data area is obtained on a customer-by-customer basis, the greater the accuracy will be.

[0044] Each customer consents to their own customer data being anonymized (vectorized) and used within the data sphere. Companies may offer incentives or benefits to customers in return for their consent. For example, if a customer consents, they will receive discount coupons and other benefits from companies within the data sphere. Customers who consent will also be issued rewards based on their activities within the data sphere (purchasing products / using services from companies participating in the data sphere, interacting with advertisements, answering surveys, etc.).

[0045] The mother server 100 (server side) sets a list of past product purchaser IDs from each company's CDP as correct answer data and performs AI learning. For example, if company A wants to create a product purchase probability list (recommendation list) that is a list of customers who have a high probability of purchasing its company's product X, the mother server 100 (server side) sets a list of IDs of past product X purchasers from company A's CDP as correct answer data as a goal and performs AI learning.

[0046] Here, an example of learning from campaign objectives will be described with reference to Figure 3. Figure 3 is an explanatory diagram showing an example of learning from campaign objectives. It should be noted that AI learning is performed for each campaign. As shown in Figure 3, the mother server 100 (server side) sets the objective to a list of IDs of past purchasers of product X from the data of company A and performs AI learning.

[0047] At this time, the mother server 100 (server side) may set the correct answer data by taking into account customer data (mother side data) that it holds in addition to the anonymous information of each company. For example, if company A is an automobile manufacturer and product X is a specific vehicle, the mother server 100 (server side) may set the correct answer data by taking into account customer data of "people who are interested in golf" as mother side data.

[0048] Then, the mother server 100 (server side) performs AI learning using all records of the customer data of the company to which the correct answer data was given, and all records of other affiliated companies that are linked by a common ID among all records of the customer data of the company, as learning data.

[0049] [1-2. Vectorization of customer data] Next, vectorization of customer data according to this embodiment will be described with reference to Fig. 4. Fig. 4 is an explanatory diagram showing an overview of vectorization of customer data. In this embodiment, by vectorizing the personal information of each customer of each company, it is no longer personal information, but customer characteristics can be processed.

[0050] As shown in FIG. 4, for example, if Company A is an automobile manufacturer and Product X is a specific vehicle, Company A's client terminal 10 collects information such as purchased products, store visits and conversations, inquiries and catalog requests, owned web logs, advertising contacts, and sales promotion contacts as customer data.

[0051] Furthermore, the client terminal 10 of company A uses either of the following methods (1) or (2) as a method for vectorizing customer data.

[0052] (1) The client terminal 10 of Company A assumes that products (models of cars) that customers simultaneously negotiate or purchase are similar in purpose and vectorizes the products. Next, using the product vectors, two customers are randomly extracted, and the proximity of the product vectors is learned as positive and negative examples.

[0053] (2) Company A's client terminal 10 learns using a BERT (Bidirectional Encoder Representations from Transformers) model in a fill-in-the-blank manner and acquires the data from the embedding layer. BERT is a Transformer-based machine learning method for pre-training natural language processing (NLP). In practice, other natural language processing methods may be used instead of BERT. For example, distributed representations (word vectors) such as Word2Vec may be used. Distributed representations embed characters and words in a vector space and represent them as a single point in that space; they are also called word embeddings.

[0054] Figure 4 shows an example of anonymization (vectorization) of Company A's customer data. Here, for simplicity, the vector dimension is expressed in two dimensions, and the vector group is classified into several segments based on the proximity of the vector. For example, they can be classified into segments such as "luxury sports car preference," "luxury SUV preference," and "price range preference."

[0055] In the example of anonymization (vectorization) of Company A's customer data shown in Figure 4, the vector dimension is expressed in two dimensions, but the actual vector dimension is not limited to two dimensions, and each customer is expressed in multiple dimensions, such as 100 dimensions, taking into account all factors.

[0056] [1-3. Server-side inference] Next, inference on the server side according to this embodiment will be described with reference to Figures 5 and 6. Figure 5 is an explanatory diagram showing an overview of inference on the server side. Figure 6 is a schematic diagram that simplifies inference on the server side. Note that inference on the server side may be performed after AI learning on the server side, or may be performed simultaneously.

[0057] As shown in Figure 5, the mother server 100 (server side) inputs a vector (anonymous information) of customer data for each company into the mother side AI and obtains a product purchase probability list (recommendation list) for each customer of each company.

[0058] At this time, the mother server 100 (server side) obtains a purchase probability list of a specific product by a specific company ID (e.g., a product X purchase probability company A ID list) that targets only customers of the specific company, and a purchase probability list of a specific product by a common ID (e.g., a product X purchase probability common ID list) that targets customers of all companies.

[0059] Then, the mother server 100 (server side) transmits a purchase probability list of a specific product by a specific company ID targeted only at customers of the specific company only to the client terminal 10 of the specific company. At this time, the client terminal 10 of the specific company performs the process of vector neighborhood customer expansion, which will be described later, based on the purchase probability list of a specific product by a specific company ID targeted only at customers of the specific company.

[0060] In addition, the mother server 100 (server side) can deliver advertisements to all company customers and distribute coupons on social media based on a list of purchase probabilities of specific products using a common ID targeted at all company customers.

[0061] As shown in Figure 6, if Company A desires a list of customers who are likely to purchase its product X, the mother server 100 (server side) inputs information on customers who have previously purchased Company A's product X and who have consented to common ID federation as correct answer data into the mother-side AI, and obtains a purchase probability list (recommendation list) for Company A's product X. Then, based on the purchase probability list for product X, the mother server 100 (server side) performs advertisement distribution, coupon distribution on SNS, etc. The mother server 100 (server side) may also return the list to Company A's CRM (Customer Relationship Management) and perform vector neighborhood expansion within Company A's CDP, etc., enabling marketing activities to be carried out using Company A's CRM after vector neighborhood expansion.

[0062] [1-4. Vector Neighborhood Customer Expansion] Next, vector neighborhood customer expansion according to this embodiment will be described with reference to Fig. 7. Fig. 7 is an explanatory diagram showing an overview of vector neighborhood customer expansion.

[0063] When the mother server 100 (server side) infers correct data, it is based only on the vector data on the mother side (vector data of ID Connected customers), so when it is returned to the client side CDP, neighborhood expansion is performed there using vectors from the company's own data.

[0064] For example, the client terminal 10 of Company A assumes that, based on the customer data of Company A, users U1 and U2, who are customers, are candidates for purchasing product X. Then, the client terminal 10 of Company A executes a vectorization learning process within Company A's CDP (see "Company A Customer Vector" in FIG. 7).

[0065] At this time, the mother server 100 (server side) uses the data of users U1 to U4 who have ID integration (ID Connect) with a common ID to learn with the mother side AI. Here, we are using users U1 to U4 as an example, but in reality, the amount of data is much larger, including customers of other companies (ID Connected users).

[0066] The mother server 100 (server side) infers people who have a high probability of purchasing product X from company A from the vectors of all customers of all companies on the mother side. At this time, user U1 is extracted from company A's data (user U2 is not ID connected).

[0067] Then, the mother server 100 (server side) feeds back U1 to the CDP of company A as a candidate for purchasing product X. Then, the client terminal 10 of company A performs vector neighborhood customer expansion in response to the feedback from the mother side. Specifically, as shown in "Company A Customer Vector" in Figure 7, in the vector of customer data within company A, user U2 exists in the vicinity of user U1, so the client terminal 10 of company A also considers user U2 to be a candidate for purchasing product X.

[0068] [1-5. 1st Party Data Zone Concept] As mentioned above, by joining the 1stParty Data Zone initiative, companies will be able to conduct highly accurate personal marketing. Figure 8 is a conceptual diagram showing the image of the 1stParty Data Zone initiative.

[0069] For example, targeting can be performed using AI that has learned the data of all customers who have agreed to the common ID Connect of all companies that are members of the data zone (high-precision targeting).

[0070] Furthermore, each company's customer data itself is not passed on to other companies or the parent company. Each company's customer data is converted into a list of numerical values ​​(vectorized) and then learned by the parent company's AI (it is OK not to release the company's own customer data itself).

[0071] Additionally, the higher the quantity and quality of a company's vector data (anonymous information based on its own customer data), the more accurate its targeting will be. Therefore, the more of a company's vector data it provides to the first-party data sphere, the higher its accuracy will be. Conversely, the fewer of a company's vector data it provides, the lower its accuracy will be (a system that does not allow for free riding).

[0072] If a company does not provide any of its own vector data, the data sphere will only contain unknown vector data from other companies, and the relationship between that data and its own customer data will be unclear. By providing its own vector data, it is possible to link its own vector data with that of other companies for targeting. Furthermore, the more content and number of items in its own vector data, the more connections it will have with other companies' vector data, and the more information it will be able to obtain from the data sphere.

[0073] In order to improve the accuracy of their own targeting, each company actively provides its own vector data, creating a virtuous cycle in which large amounts of vector data naturally accumulate on the mother side.

[0074] According to this embodiment, data sphere collaboration and targeting accuracy can be improved without each company disclosing raw personal information. Furthermore, because a system that prevents free riding is in place, overall accuracy improves synergistically. From each company's perspective, even if they join the data sphere, their own data and situation cannot be seen by other companies (for example, Company A cannot know how many customers visit its stores). Furthermore, if the parent company offers multiple services (such as a portal site), it can also be used for mass marketing, such as advertising distribution and social networking sites.

[0075] [2. Mother Server Configuration Example] Next, the configuration of the mother server 100 according to the embodiment will be described with reference to Fig. 9. Fig. 9 is a diagram showing an example of the configuration of the mother server 100 according to the embodiment. As shown in Fig. 9, the mother server 100 has a communication unit 110, a storage unit 120, and a control unit 130.

[0076] (Communication unit 110) The communication unit 110 is realized by, for example, a network interface card (NIC), etc. The communication unit 110 is connected to a network N by wire or wirelessly.

[0077] (Storage unit 120) The storage unit 120 is realized by, for example, a semiconductor memory element such as a RAM (Random Access Memory) or a flash memory, or a storage device such as a hard disk or an optical disk. As shown in Fig. 9, the storage unit 120 has a user information database 121, a history information database 122, and an anonymous information database 123.

[0078] (User Information Database 121) The user information database 121 stores user information about the user U (user) that is independently collected by the mother server 100. For example, the user information database 121 stores various information such as the attributes of the user U. FIG. 10 is a diagram showing an example of the user information database 121. In the example shown in FIG. 10, the user information database 121 has items such as "User ID (Identifier)," "Age," "Gender," "Home," "Workplace," and "Interests."

[0079] The "user ID" indicates identification information for identifying the user U. The "user ID" may be the contact information of the user U (telephone number, email address, etc.), or may be identification information for identifying the client terminal 10 of the user U.

[0080] Furthermore, "age" indicates the age of user U identified by the user ID. Note that "age" may be information indicating the specific age of user U (e.g., 35 years old), or may be information indicating the generation of user U (e.g., 30s). Alternatively, "age" may be information indicating the date of birth of user U, or may be information indicating the generation of user U (e.g., born in the 1980s). Furthermore, "gender" indicates the gender of user U identified by the user ID.

[0081] Furthermore, "home" indicates the location information of the home of user U identified by the user ID. In the example shown in FIG. 10, "home" is illustrated as an abstract code such as "LC11," but it may also be latitude and longitude information, etc. Furthermore, for example, "home" may also be the name of an area or an address.

[0082] Furthermore, "workplace" indicates location information of the workplace (school in the case of a student) of user U identified by the user ID. In the example shown in FIG. 10, "workplace" is illustrated as an abstract code such as "LC12," but it may also be latitude and longitude information, etc. Furthermore, for example, "workplace" may also be the name of a region or an address.

[0083] Furthermore, "interests" indicate the interests of user U identified by the user ID. In other words, "interests" indicate subjects in which user U identified by the user ID is highly interested. For example, "interests" may be search queries (keywords) entered by user U into a search engine. In the example shown in FIG. 10, each user U is shown with one "interest," but there may be multiple "interests."

[0084] For example, in the example shown in FIG. 10, the age of user U identified by user ID "U1" is "20s" and the gender is "male." Furthermore, for example, the home address of user U identified by user ID "U1" is "LC11." Furthermore, for example, the workplace of user U identified by user ID "U1" is "LC12." Furthermore, for example, the user U identified by user ID "U1" is interested in "sports."

[0085] 10, abstract values ​​such as "U1", "LC11", and "LC12" are used for illustration, but "U1", "LC11", and "LC12" are assumed to store information such as specific character strings and numerical values. Below, abstract values ​​may also be illustrated in diagrams relating to other information.

[0086] The user information database 121 may store various types of information depending on the purpose, without being limited to the above. For example, the user information database 121 may store various types of information related to the client terminal 10 of the user U. The user information database 121 may also store information related to the user U's attributes, such as demographic attributes, psychographic attributes, geographic attributes, and behavioral attributes. For example, the user information database 121 may store information such as name, family structure, hometown (hometown), occupation, job title, income, qualifications, type of residence (detached house, apartment, etc.), whether or not the user has a car, commuting time, commuting route, commuter pass area (station, line, etc.), frequently used stations (other than the station nearest to home or workplace), extracurricular activities (location, time zone, etc.), hobbies, interests, lifestyle, etc.

[0087] (History Information Database 122) The history information database 122 stores various information related to history information (log data) that indicates the behavior of the user U and that is independently collected by the mother server 100. Fig. 11 is a diagram showing an example of the history information database 122. In the example shown in Fig. 11, the history information database 122 has items such as "user ID," "location history," "search history," "browsing history," "purchase history," and "posting history."

[0088] "User ID" indicates identification information for identifying user U. "Location history" indicates the location history, which is the history of user U's location and movements. "Search history" indicates the search history, which is the history of search queries entered by user U. "Browsing history" indicates the browsing history, which is the history of content viewed by user U. "Purchase history" indicates the purchase history, which is the history of purchases made by user U. "Posting history" indicates the posting history, which is the history of posts made by user U. "Posting history" may also include questions about user U's possessions.

[0089] For example, in the example shown in Figure 11, user U, identified by user ID "U1," moved as shown in "Location History #1," searched as shown in "Search History #1," viewed content as shown in "Viewing History #1," purchased specified products at specified stores as shown in "Purchase History #1," and posted as shown in "Posting History."

[0090] Here, in the example shown in Figure 11, abstract values ​​such as "U1", "Location History #1", "Search History #1", "Browse History #1", "Purchase History #1", and "Post History #1" are used for the illustration, but "U1", "Location History #1", "Search History #1", "Browse History #1", "Purchase History #1", and "Post History #1" are assumed to store specific information such as character strings and numbers.

[0091] The history information database 122 is not limited to the above and may store various types of information depending on the purpose. For example, the history information database 122 may store the user U's usage history of a specific service. The history information database 122 may also store the user U's store visit history or facility visit history. The history information database 122 may also store the user U's payment history (electronic payment) using the client terminal 10.

[0092] (Anonymous Information Database 123) The anonymous information database 123 stores various information related to anonymous information (vector data) of the user U. Fig. 12 is a diagram showing an example of the anonymous information database 123. In the example shown in Fig. 12, the anonymous information database 123 has items such as "common ID," "company ID," "company," "vector," and "inference."

[0093] "Common ID" indicates a customer ID that is common across the data domain. "Company ID" indicates a customer ID unique to each company. "Company" indicates the company that provided the anonymous information (vector data).

[0094] Furthermore, a "vector" indicates a vector value obtained by anonymizing (vectorizing) customer data for each company. A "vector" may be multidimensional. In other words, a "vector" may be a collection of multiple vector values.

[0095] Furthermore, "inference" refers to the result of inference based on a vector. For example, an "inference" may be a product purchase probability list (recommendation list) for each customer of each company. Note that "inference" is not limited to the result of inference for each company (for each company ID), but also includes the result of inference common to all companies (common ID).

[0096] For example, in the example shown in Figure 12, user U, identified by common ID "U1," is managed with customer data from "Company A" using company-specific ID "U1A," and "inference A1" can be obtained based on "vector A1," which is customer data that has been anonymized (vectorized).

[0097] 12, abstract values ​​such as "U1", "U1A", "Company A", "Vector A1", and "Inference A1" are used for illustration, but "U1", "U1A", "Company A", "Vector A1", and "Inference A1" are assumed to store information such as specific character strings and numerical values. Below, abstract values ​​may also be illustrated in diagrams relating to other information.

[0098] The anonymous information database 123 may store various types of information depending on the purpose, not limited to the above. For example, the anonymous information database 123 may store information about the client-side and mother-side AIs. The anonymous information database 123 may also store information about AI learning and inference techniques.

[0099] (control unit 130) 9, the explanation will be continued. The control unit 130 is a controller, and is realized by, for example, a CPU (Central Processing Unit), an MPU (Micro Processing Unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or the like, executing various programs (corresponding to an example of an information processing program) stored in a storage device inside the mother server 100 using a storage area such as a RAM as a working area. In the example shown in FIG. 9, the control unit 130 has an acquisition unit 131, a learning unit 132, an estimation unit 133, and a provision unit 134.

[0100] (Acquisition part 131) The acquisition unit 131 acquires a search query input by the user U. For example, when the user U inputs a search query into a search engine or the like to perform a keyword search, the acquisition unit 131 acquires the search query via the communication unit 110.

[0101] Furthermore, the acquisition unit 131 acquires user information about the user U via the communication unit 110. For example, the acquisition unit 131 acquires identification information (such as a user ID) indicating the user U, location information of the user U, attribute information of the user U, etc. from the client terminal 10 of the user U. Furthermore, the acquisition unit 131 may acquire the identification information indicating the user U, attribute information of the user U, etc. when registering the user U. Then, the acquisition unit 131 registers the user information in the user information database 121 of the storage unit 120.

[0102] Furthermore, the acquisition unit 131 acquires various types of history information (log data) indicating the behavior of the user U via the communication unit 110. For example, the acquisition unit 131 acquires various types of history information indicating the behavior of the user U from the client terminal 10 of the user U or from various servers based on the user ID or the like. Then, the acquisition unit 131 registers the various types of history information in the history information database 122 of the storage unit 120.

[0103] Furthermore, the acquisition unit 131 acquires anonymous information obtained by anonymizing (vectorizing) customer data from the client terminal 10 of each company participating in the data zone via the communication unit 110. For example, the acquisition unit 131 acquires anonymous information obtained by anonymizing (vectorizing) customer data by a client-side AI from the client terminal 10 of each company via the communication unit 110.

[0104] At this time, the acquiring unit 131 acquires anonymous information obtained by anonymizing (vectorizing) customer data from each of the client terminals 10 of a plurality of companies in different industries. Note that the plurality of companies may include competing companies (other companies in the same industry).

[0105] The anonymous information is multidimensional vector data obtained by vectorizing customer data, and is irreversible information with no means of restoring the customer data. In other words, the anonymous information is not encrypted customer data and cannot be decrypted.

[0106] (Learning Section 132) The learning unit 132 performs machine learning of the AI ​​using anonymous information from the client terminal 10 of each company. At this time, the learning unit 132 may perform machine learning of the AI ​​by adding anonymous information obtained by anonymizing (vectorizing) customer data (mother-side data) independently collected by the mother server 100 separately from each company to the anonymous information from the client terminal 10 of each company.

[0107] Furthermore, during machine learning of the AI, the learning unit 132 sets the objective of the specific company using data corresponding to the objective from among the anonymous information from the client terminal 10 of each company as correct answer data. At this time, the learning unit 132 may set the objective of the specific company using data obtained by adding customer data independently collected by the mother server 100 separately from each company to the data corresponding to the objective from among the anonymous information from the client terminal 10 of each company as correct answer data. Furthermore, if the mother-side AI can also anonymize (anonymize (vectorize)) customer data and convert it into anonymous information in the same way as the client-side AI, the customer data independently collected by the mother server 100 may be input directly to the mother-side AI.

[0108] (Estimation part 133) The estimation unit 133 estimates customers who are suitable for the purpose of a specific company among a plurality of companies, using AI that has undergone machine learning by the learning unit 132.

[0109] Note that the more anonymous information there is from a specific company among the multiple companies, the more accurately the estimation unit 133 estimates customers who fit the purpose of the specific company. Conversely, the fewer anonymous information there is from a specific company, the lower the accuracy of estimating customers who fit the purpose of the specific company.

[0110] The estimation unit 133 also estimates customers who have a high probability of purchasing a specific product from a specific company among multiple companies, and generates a product purchase probability list that lists the customers.

[0111] (Provider 134) The providing unit 134 provides, via the communication unit 110, the client-side AI for anonymizing (vectorizing) customer data to each of the client terminals 10 of the multiple companies.

[0112] Furthermore, the providing unit 134 provides information based on the estimation result by the estimation unit 133 via the communication unit 110. For example, the providing unit 134 provides a product purchase probability list that lists customers who have a high probability of purchasing a specific product of a specific company to the client terminal 10 of the specific company via the communication unit 110. This allows the specific company (client side) to distribute advertisements and coupons to customers on the list from the mother side, thereby acquiring new customers and providing anonymous information about the new customers to the mother side, thereby further improving the accuracy of AI inference.

[0113] Furthermore, the providing unit 134 distributes advertisements and coupons for specific products of specific companies to customers listed on the product purchase probability list via the communication unit 110. As a result, the specific company (client side) acquires new customers and provides anonymous information about the new customers to the mother side, thereby further improving the accuracy of AI inference.

[0114] [3. Processing Procedure] Next, a processing procedure by the mother server 100 according to the embodiment will be described with reference to Fig. 13. Fig. 13 is a flowchart showing the processing procedure according to the embodiment. The processing procedure shown below is repeatedly executed by the control unit 130 of the mother server 100.

[0115] As shown in FIG. 13, the providing unit 134 provides the client-side AI for anonymizing (vectorizing) customer data to each of the client terminals 10 of a plurality of companies via the communication unit 110 (step S101).

[0116] Next, the acquisition unit 131 of the mother server 100 acquires anonymous information, which is customer data anonymized (vectorized) by the client-side AI, from the client terminals 10 of each company participating in the data sphere via the communication unit 110 (step S102). The companies participating in the data sphere include competitors and companies in different industries.

[0117] Next, the learning unit 132 of the mother server 100 performs machine learning of the AI ​​using the anonymous information from the client terminal 10 of each company (step S103). At this time, the learning unit 132 performs machine learning of the AI ​​using the anonymous information from the client terminal 10 of each company and anonymous information obtained by anonymizing (vectorizing) customer data (mother-side data) that the mother server 100 has independently collected separately from each company.

[0118] Next, during AI machine learning, the learning unit 132 of the mother server 100 sets the objective of the specific company by taking the data corresponding to the objective from the anonymous information from the client terminal 10 of each company and the customer data independently collected by the mother server 100 separately from each company as the correct answer data (step S104). Note that it is optional whether or not to take into account the customer data from the mother server. Of course, there is no problem in using only the data corresponding to the objective from the anonymous information from the client terminal 10 of each company as the correct answer data.

[0119] Next, the estimation unit 133 of the mother server 100 estimates customers who fit the purpose of the specific company among the multiple companies using AI that has undergone machine learning by the learning unit 132 (step S105). Note that the estimation unit 133 estimates customers who fit the purpose of the specific company with higher accuracy as the number of anonymous information items from the specific company among the multiple companies increases.

[0120] Next, the estimation unit 133 of the mother server 100 generates a list of customers who fit the purpose of the specific company (step S106). For example, the estimation unit 133 generates a product purchase probability list that lists customers who have a high probability of purchasing a specific product of a specific company.

[0121] Next, the providing unit 134 of the mother server 100 provides a list of customers who fit the purpose of the specific company to the client terminal 10 of the specific company via the communication unit 110 (step S107). For example, the providing unit 134 provides a product purchase probability list that lists customers who have a high probability of purchasing a specific product of the specific company to the client terminal 10 of the specific company via the communication unit 110.

[0122] Next, the providing unit 134 of the mother server 100 provides information in line with the objectives of the specific company to customers who fit the objectives of the specific company via the communication unit 110 (step S108). For example, the providing unit 134 distributes advertisements and coupons related to the specific product to customers listed on the product purchase probability list via the communication unit 110.

[0123] [4. Modifications] The above-described client terminal 10 and mother server 100 may be implemented in various different forms other than the above-described embodiment. Therefore, modifications of the embodiment will be described below.

[0124] (Standalone) In the above embodiment, some or all of the processing executed by the mother server 100 may actually be executed by the client terminal 10. For example, the processing may be completed in a stand-alone manner (by the client terminal 10 alone). In this case, the client terminal 10 is assumed to have the functions of the mother server 100 in the above embodiment. Also, in the above embodiment, the client terminal 10 is linked to the mother server 100, so from the perspective of the user U, it appears that the processing of the mother server 100 is also executed by the client terminal 10. In other words, from another perspective, the client terminal 10 can also be said to be equipped with the mother server 100. Also, the mother server 100 may be one of multiple client terminals 10.

[0125] (Basis Vectors) In addition, in the above embodiment, in order to further increase the anonymity of the anonymous information, the mother server 100 may determine a basis vector in advance and collect difference data between the basis vector and the anonymous information from the client terminal 10. In this case, only the vector difference data is transmitted from the client terminal 10 to the mother server 100.

[0126] (Anonymous information format changes for each company) Furthermore, in the above embodiment, the format of the anonymous information may be different for each company. For example, if the anonymous information is multidimensional vector data, it is not necessary for the vector data of Company A and the vector data of Company B to be arranged in the same dimension. Even if the information contains the same content, the vector data of Company A and the vector data of Company B may be arranged in different dimensions. It is sufficient for only the mother-side AI to recognize which data is arranged in which dimension of each company's multidimensional vector data. In this case, the mother-side AI may ensure consistency (alignment) of the vector data of each company prior to learning and inference. For example, the vector data of each company is converted into vector data in a common format, and then learning and inference are performed.

[0127] (AI learning on the client side) Furthermore, in the above embodiment, the mother server 100 distributes AI to each company's client terminal 10, but in reality, each company's client terminal 10 may independently generate AI for its own use and provide information regarding the specifications of the generated AI (e.g., vector conversion rules) or the AI ​​itself (e.g., vector conversion model) to the mother server 100.

[0128] For example, the client terminal 10 uses customer data and vector values ​​as a data set to construct a learning model using a neural network-based machine learning technique. The client terminal 10 then inputs the customer data into the constructed learning model and converts the customer data into a multidimensional real-valued vector. That is, the client terminal 10 inputs the customer data into a vector conversion model generated through machine learning using, for example, multiple fully connected layers, a variational autoencoder (VAE), a recurrent neural network (RNN), a long short-term memory (LSTM), a transformer, or a BERT, and obtains anonymous information obtained by anonymizing (vectorizing) the customer data as output.

[0129] The RNN and LSTM may be neural networks based on an attention mechanism. Attention can handle data where the order of precedence and followance is important, such as text. The client terminal 10 may also use a similar natural language processing model. By using such a model to generate vector values ​​from customer data, it is possible to generate vector values ​​that place more emphasis on the order of information.

[0130] Then, when each company's client terminal 10 sends anonymous information obtained by anonymizing (vectorizing) customer data to the mother server 100 in response to a request from the mother server 100 or voluntarily, or at any time, each company's client terminal 10 sends differential data of the learned vector conversion model (or the model itself) to the mother server 100.

[0131] Furthermore, each company's client terminal 10 may generate a vector conversion model for converting customer data into vectors using a technique such as federated learning. Federated learning is a machine learning technique in which a learning model is jointly constructed without exposing the data to outside the company. For example, in federated learning, each company's client terminal 10 performs machine learning on a vector conversion model using customer data as learning data, and transmits differential data between the original model and the learned model to the mother server 100. The mother server 100 receives the differential data from each client terminal 10, integrates the differential data, and generates a common model. The mother server 100 then provides parameters of the generated common model to each client terminal 10. In this way, using federated learning makes it possible to perform learning without providing customer data, which is learning data, from the client side to the server side, thereby taking into consideration the personal information and privacy of customers.

[0132] [5. Effects] As described above, the information processing device (mother server 100) according to the present application comprises an acquisition unit that acquires anonymous information obtained by vectorizing customer data from each of a plurality of companies, a learning unit that performs machine learning on an AI using the anonymous information from each company, an estimation unit that uses the AI ​​that has undergone machine learning to estimate customers who are suitable for the objectives of a specific company among the plurality of companies, and a provision unit that provides information based on the estimation results by the estimation unit.

[0133] The anonymous information is multidimensional vector data obtained by vectorizing customer data, and is irreversible information for which there is no means to restore the customer data.

[0134] Furthermore, the estimation unit estimates with higher accuracy customers who are suitable for the purpose of a specific company as the number of anonymous information items from the specific company increases among the multiple companies.

[0135] The providing unit provides each of the multiple companies with a client-side AI for vectorizing customer data, and the acquiring unit acquires anonymous information obtained by vectorizing customer data using the client-side AI from each of the multiple companies.

[0136] In addition, the learning unit performs machine learning on the AI ​​by adding anonymous information from each company to anonymous information that has been vectorized from customer data collected independently by the information processing device, separate from each company.

[0137] In addition, the learning unit sets objectives for specific companies by adding customer data collected independently by the information processing device, separate from each company, to data corresponding to the objectives from the anonymous information from each company, and performs machine learning on the AI.

[0138] The estimation unit estimates customers who are likely to purchase a specific product from a specific company among multiple companies, and generates a product purchase probability list that lists the customers.The provision unit then provides the product purchase probability list to the specific company.

[0139] The providing unit also distributes advertisements and coupons related to specific products to customers listed on the product purchase probability list.

[0140] The acquisition unit also acquires anonymous information obtained by vectorizing customer data from each of a plurality of companies in different industries.

[0141] By using any one or a combination of the above processes, the information processing device according to the present application can provide a mechanism that encourages data sphere member companies to release their data. Of course, security can be further enhanced by encrypting vectorized anonymous information. This allows data sphere collaboration and targeting accuracy to be improved without each company disclosing raw personal information. Furthermore, since a system that prevents free riding is in place, overall accuracy improves synergistically. Furthermore, from each company's perspective, even if they join the data sphere, their own data and status cannot be seen by other companies (for example, Company A cannot know how many customers visit its stores). Furthermore, if the parent company offers multiple services (such as a portal site), it can also be used for mass marketing, such as advertising distribution and social networking.

[0142] [6. Hardware Configuration] The client terminal 10 and the mother server 100 according to the above-described embodiments are realized by a computer 1000 having a configuration as shown in Fig. 14, for example. The mother server 100 will be described below as an example. Fig. 14 is a diagram showing an example of a hardware configuration. The computer 1000 is connected to an output device 1010 and an input device 1020, and has a configuration in which an arithmetic unit 1030, a primary storage device 1040, a secondary storage device 1050, an output I / F (Interface) 1060, an input I / F 1070, and a network I / F 1080 are connected via a bus 1090.

[0143] The arithmetic device 1030 operates based on programs stored in the primary storage device 1040 and the secondary storage device 1050, programs read from the input device 1020, and the like, and executes various processes. The arithmetic device 1030 is realized by, for example, a CPU (Central Processing Unit), a GPU, a TPU, an MPU (Micro Processing Unit), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or the like.

[0144] The primary storage device 1040 is a memory device such as a RAM (Random Access Memory) that temporarily stores data used by the arithmetic device 1030 for various calculations. The secondary storage device 1050 is a storage device in which data used by the arithmetic device 1030 for various calculations and various databases are registered, and is realized by a ROM (Read Only Memory), an HDD (Hard Disk Drive), an SSD (Solid State Drive), a flash memory, or the like. The secondary storage device 1050 may be an internal storage device or an external storage device. The secondary storage device 1050 may also be a removable storage medium such as a USB (Universal Serial Bus) memory or an SD (Secure Digital) memory card. The secondary storage device 1050 may also be cloud storage (online storage), a NAS (Network Attached Storage), a file server, or the like.

[0145] The output I / F 1060 is an interface for transmitting information to be output to an output device 1010 that outputs various types of information, such as a display, a projector, a printer, etc., and is realized by a connector conforming to a standard such as USB (Universal Serial Bus), DVI (Digital Visual Interface), or HDMI (High Definition Multimedia Interface), etc. The input I / F 1070 is an interface for receiving information from various input devices 1020, such as a mouse, a keyboard, a keypad, a button, a scanner, etc., and is realized by a USB, etc.

[0146] Furthermore, the output I / F 1060 and the input I / F 1070 may be wirelessly connected to the output device 1010 and the input device 1020, respectively. That is, the output device 1010 and the input device 1020 may be wireless devices.

[0147] The output device 1010 and the input device 1020 may be integrated into one device, such as a touch panel. In this case, the output I / F 1060 and the input I / F 1070 may also be integrated into one device as an input / output I / F.

[0148] The input device 1020 may be a device that reads information from, for example, an optical recording medium such as a CD (Compact Disc), a DVD (Digital Versatile Disc), or a PD (Phase Change Rewritable Disk), a magneto-optical recording medium such as an MO (Magneto-Optical disk), a tape medium, a magnetic recording medium, or a semiconductor memory.

[0149] The network I / F 1080 receives data from other devices via the network N and sends it to the arithmetic device 1030, and also transmits data generated by the arithmetic device 1030 to other devices via the network N.

[0150] The arithmetic unit 1030 controls the output device 1010 and the input device 1020 via the output I / F 1060 and the input I / F 1070. For example, the arithmetic unit 1030 loads a program from the input device 1020 or the secondary storage device 1050 onto the primary storage device 1040 and executes the loaded program.

[0151] For example, when the computer 1000 functions as the mother server 100, the arithmetic unit 1030 of the computer 1000 executes a program loaded onto the primary storage device 1040 to realize the functions of the control unit 130. The arithmetic unit 1030 of the computer 1000 may also load a program acquired from another device via the network I / F 1080 onto the primary storage device 1040 and execute the loaded program. The arithmetic unit 1030 of the computer 1000 may also cooperate with the other device via the network I / F 1080 to call and use the functions and data of a program from another program of the other device.

[0152] [7. Other] Although the embodiments of the present application have been described above, the present invention is not limited to the contents of these embodiments. Furthermore, the above-described components include those that can be easily imagined by a person skilled in the art, those that are substantially the same, and those that are within the scope of so-called equivalents. Furthermore, the above-described components can be combined as appropriate. Furthermore, various omissions, substitutions, or modifications of the components can be made without departing from the spirit of the above-described embodiments.

[0153] Furthermore, among the processes described in the above embodiments, all or part of the processes described as being performed automatically can be performed manually, or all or part of the processes described as being performed manually can be performed automatically using a known method. In addition, the information including the processing procedures, specific names, various data, and parameters shown in the above documents and drawings can be changed as desired unless otherwise specified. For example, the various information shown in each drawing is not limited to the information shown in the drawings.

[0154] Furthermore, the components of each device shown in the figure are conceptual functional components and do not necessarily have to be physically configured as shown in the figure. In other words, the specific form of distribution and integration of each device is not limited to that shown in the figure, and all or part of them can be functionally or physically distributed and integrated in any unit depending on various loads, usage conditions, etc.

[0155] For example, the above-mentioned mother server 100 may be realized by multiple server computers, and depending on the function, the configuration can be flexibly changed, such as by calling an external platform using an API (Application Programming Interface) or network computing.

[0156] Furthermore, the above-described embodiments and modifications can be combined as appropriate within the scope of not causing any contradiction in the processing content.

[0157] Furthermore, the above-mentioned "section, module, unit" can be read as "means" or "circuit," etc. For example, an acquisition unit can be read as an acquisition means or an acquisition circuit. [Explanation of symbols]

[0158] 1. Information Processing Systems 10 Client Terminal 100 Mother Server 110 Communications Department 120 Storage section 121 User Information Database 122 Historical Information Database 123 Anonymous Information Database 130 control section 131 Acquisition Department 132 Learning Department 133 Estimation Department 134 Provision Department

Claims

1. an acquisition unit that acquires anonymous information obtained by vectorizing customer data from each of a plurality of companies; a learning unit that performs machine learning using the anonymous information from each company as input and generates a customer evaluation model; an estimation unit that inputs anonymous information from a specific company among the plurality of companies into the customer evaluation model generated by the machine learning and estimates customers who are likely to purchase a specific product from the specific company; a providing unit that provides information based on the estimation result by the estimation unit; Equipped with The learning unit performs machine learning using, as input, the anonymous information from each of the companies plus anonymous information obtained by vectorizing customer data independently collected by the information processing device separately from each of the companies, and generates the customer evaluation model.

1. An information processing device comprising:

2. The anonymous information is multidimensional vector data obtained by vectorizing the customer data, and is irreversible information for which there is no means to restore the customer data.

2. The information processing apparatus according to claim 1, wherein:

3. The estimation unit estimates with higher accuracy customers who are suitable for the purpose of the specific company as the number of anonymous information items from the specific company among the plurality of companies increases.

3. The information processing apparatus according to claim 1, wherein the information processing apparatus is a computer.

4. the providing unit provides each of the plurality of companies with a vector conversion model for inputting the customer data held by each of the plurality of companies and converting it into vectorized anonymous information; The acquisition unit acquires anonymous information obtained by vectorizing customer data from each of the plurality of companies using the vector conversion model.

4. The information processing device according to claim 1, wherein the information processing device is a computer.

5. The learning unit, when performing machine learning of the customer evaluation model, sets the objective of the specific company using data obtained by adding customer data independently collected by the information processing device separately from each company to data corresponding to the objective among the anonymous information from each company as correct answer data.

5. The information processing device according to claim 1, wherein the information processing device is a computer.

6. the estimation unit estimates customers who are highly likely to purchase a specific product of the specific company among the plurality of companies, and generates a product purchase probability list that lists the customers; The providing unit provides the product purchase probability list to the specific company.

6. The information processing device according to claim 1, wherein the information processing device is a computer.

7. The provision unit distributes advertisements and coupons related to the specific product to customers listed on the product purchase probability list.

7. The information processing apparatus according to claim 6,

8. The acquisition unit acquires anonymous information obtained by vectorizing customer data from each of a plurality of companies in different industries.

8. The information processing device according to claim 1, wherein the information processing device is a computer.

9. An information processing method executed by an information processing device, an acquisition step of acquiring anonymous information obtained by vectorizing customer data from each of a plurality of companies; a learning process in which machine learning is performed using the anonymous information from each company as input to generate a customer evaluation model; an estimation step of inputting anonymous information from a specific company among the plurality of companies into the customer evaluation model generated by the machine learning, and estimating customers who are likely to purchase a specific product from the specific company; a providing step of providing information based on the estimation result obtained by the estimation step; Including, In the learning step, machine learning is performed using input obtained by adding anonymous information from each company to anonymous information obtained by vectorizing customer data independently collected by the information processing device separately from each company, to generate the customer evaluation model. An information processing method comprising:

10. An acquisition procedure for acquiring anonymous vectorized customer data from each of a plurality of companies; a learning procedure in which machine learning is performed using the anonymous information from each company as input to generate a customer evaluation model; an estimation step of inputting anonymous information from a specific company among the plurality of companies into the customer evaluation model generated by the machine learning, and estimating customers who are likely to purchase a specific product from the specific company; a provision step of providing information based on an estimation result obtained by the estimation step; An information processing program for causing a computer to execute the above, In the learning procedure, machine learning is performed using input obtained by adding anonymous information from each company to anonymous information obtained by vectorizing customer data independently collected by the information processing device separately from each company, and the customer evaluation model is generated. An information processing program characterized by:

Images