Information processing apparatus, information processing method, and program

Abstract

To enable extraction of a candidate of an attraction destination which is further matched with characteristics of an activity relating to an economical activity of a user.SOLUTION: The present invention is directed to an information processing apparatus having associating means for associating each of a series of history data relating to at least any one of received money and expenditure of a user with any one of a plurality of activities relating to an economic activity of the user, extracting means for extracting a candidate of an attraction destination responding to activity characteristics of the user based on time-serial change of each of the plurality of activities, and outputting means for outputting information indicating the extracted candidate of the attraction destination to a predetermined output destination.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] The present disclosure relates to an information processing apparatus, an information processing method, and a program.

Background Art

[0002] Various techniques have been studied to induce a target user to purchase a product or use a service. As an example of such a technique, based on the product purchase history of a target user, a mechanism has been proposed that recommends products likely to match the hobbies and preferences of the user and encourages the user to make a purchase. As another example, a mechanism has been proposed that encourages a target user to make a purchase by introducing what other products other users who have purchased the same product as the target user have purchased. Patent Document 1 discloses a technique of extracting and recommending only products in a genre that matches the preferences of a user based on the product purchase history information and the Internet surfing history by the user.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] On the other hand, although candidates for attraction extracted from the product purchase history and service usage status conform to the trend of the user's preferences, they do not necessarily conform to the characteristics of actions related to economic activities such as investment and savings. Against this background, the realization of a mechanism capable of extracting candidates for attraction that conform to the characteristics of actions related to the economic activities of users is expected.

[0005] In view of the above problems, an object of the present invention is to be able to extract candidates for attractive destinations in accordance with the characteristics of actions related to the economic activities of users.

Means for Solving the Problems

[0006] The information processing apparatus according to the present invention associates each series of history data related to at least one of user deposits and withdrawals with any one of a plurality of actions related to the economic activities of the user, and extraction means for extracting candidates for attractive destinations according to the characteristics of the actions of the user, and output means for outputting information indicating the extracted candidates for attractive destinations to a predetermined output destination. including the amount and date of the transaction for each transaction along with each of the series of history data classify based on the purpose of the transaction and classify the historical data obtained any one of a plurality of actions related to the economic activities of the user into consumer behavior, savings behavior, and investment behavior with an associating means for associating, and For the data showing changes along the time series of each of the plurality of behaviors, for a learned model constructed based on supervised learning using teacher data in which information indicating candidates for the attraction destination used by the user who is the subject of the plurality of behaviors is associated as the correct label, for each of the series of historical data associated with changes along the time series of each of the plurality of actions by inputting data indicating It is provided with extraction means for extracting candidates for attractive destinations according to the characteristics of the actions of the user, and output means for outputting information indicating the extracted candidates for attractive destinations to a predetermined output destination.

Effects of the Invention

[0007] According to the present invention, it becomes possible to extract candidates for attractive destinations in accordance with the characteristics of actions related to the economic activities of users.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Embodiments for Carrying Out the Invention

[0009] Hereinafter, preferred embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. In this specification and the drawings, for components having substantially the same functional configuration, the same reference numerals are given to omit redundant explanations.

[0010] <System Configuration> Referring to FIG. 1, an example of the system configuration of the information processing system according to the present embodiment will be described. The information processing system 1 according to the present embodiment uses the history of procedures according to various instructions such as various instructions from a user and various instructions from other financial institutions to determine the characteristics of the behavior of the target user, and outputs candidates for attractive destinations such as products and services that conform to the characteristics of the behavior. Specifically, the information processing system 1 determines (or estimates) the characteristics of the actions related to the economic activities of the target user by using the history of deposits and withdrawals associated with the use of financial institutions, the history of cashless payments using electronic money, etc., the point history given with the purchase of goods or use of services, and the like. Moreover, the information processing system 1 extracts and outputs candidates for attractive destinations such as goods and services that match the characteristics of the actions of the target user (for example, candidates for goods and services to be recommended to the user). In this embodiment, in particular, by focusing on the case where the characteristics of the actions related to the economic activities of the user are determined, and financial products and financial education content that match the characteristics of the actions are extracted and output as candidates for attractive destinations, the configuration and processing of the information processing system 1 will be described.

[0011] The information processing system 1 according to this embodiment includes an information processing device 100 and one or more terminal devices 200. In the example shown in FIG. 1, terminal devices 200a and 200b are provided as the terminal devices 200. The information processing device 100 and the terminal devices 200 are connected via a network N1 so as to be able to mutually transmit and receive various information and data.

[0012] The type of the network N1 that connects the devices constituting the information processing system 1 is not particularly limited. As a specific example, the network N1 may be constituted by a LAN (Local Area Network), the Internet, a dedicated line, a WAN (Wide Area Network), or the like. Further, the network N1 may be constituted by a wired network or a wireless network. Further, the network N1 may include a plurality of networks, and as some of the networks, networks of a different type from other networks may be included. Further, it is only necessary that communication between the devices is logically established, and the physical configuration of the network N1 is not particularly limited. As a specific example, communication between the devices may be relayed by other communication devices or the like. In addition, the series of devices constituting the information processing system 1 do not necessarily have to be connected to a common network. That is, as long as it is possible to establish communication between the devices that transmit and receive information and data, the networks to which some of the devices and other devices are directly connected may be different.

[0013] The information processing device 100 determines the behavior characteristics of the user based on various processes according to the user's instructions and history information of various procedures, and extracts and outputs candidates for attraction destinations according to the behavior characteristics of the user. The output destination of the information according to the extraction result of the candidates for the attraction destination is not particularly limited. As a specific example, the information processing device 100 may output information according to the extraction result of the candidates for the attraction destination to the terminal device 200 via the network N1, and present the information to the user of the terminal device 200 via the terminal device 200. As another example, the information processing device 100 may output information according to the extraction result of the candidates for the attraction destination to another server device or the like that performs various analyzes, or may store it in a predetermined storage area. Details of the mechanism related to the determination of the user's behavior characteristics and the mechanism related to the extraction and output of candidates for the attraction destination according to the behavior characteristics of the user will be described separately later.

[0014] The terminal device 200 serves as an input interface for receiving inputs related to the use of functions provided by the information processing device 100, and also serves as an output interface for presenting various types of information to the user.

[0015] Note that the configuration shown in FIG. 1 is merely an example, and the system configuration of the information processing system 1 is not necessarily limited as long as it is possible to realize the functions of each component of the information processing system 1 described separately later. As a specific example, the information processing system 1 may be realized as a so-called stand-alone environment in which the information processing device 100 and the terminal device 200 are integrally configured. Also, as another example, a component corresponding to the information processing device 100 may be realized by a plurality of devices cooperating with each other, or may be realized as a so-called network service.

[0016] <Hardware Configuration> With reference to FIG. 2, an example of the hardware configuration of an information processing device 900 applicable as the information processing device 100 or the terminal device 200 in the information processing system 1 according to the present embodiment will be described. As shown in FIG. 2, the information processing device 900 according to the present embodiment includes a CPU (Central Processing Unit) 910, a ROM (Read Only Memory) 920, and a RAM (Random Access Memory) 930. The information processing device 900 also includes an auxiliary storage device 940 and a network I / F 970. The information processing device 900 may also include at least one of an output device 950 and an input device 960. The CPU 910, the ROM 920, the RAM 930, the auxiliary storage device 940, the output device 950, the input device 960, and the network I / F 970 are interconnected via a bus 980.

[0017] The CPU 910 is a central processing unit that controls various operations of the information processing apparatus 900. For example, the CPU 910 may control the overall operation of the information processing apparatus 900. The ROM 920 stores control programs, boot programs, etc. that are executable by the CPU 910. The RAM 930 is the main memory of the CPU 910 and is used as a work area or a temporary storage area for expanding various programs.

[0018] The auxiliary storage device 940 stores various data and various programs. The auxiliary storage device 940 is realized by a storage device such as an HDD (Hard Disk Drive) or a non-volatile memory typified by an SSD (Solid State Drive) that can store various data temporarily or persistently.

[0019] The output device 950 is a device that outputs various information and is used to present various information to the user. For example, the output device 950 is realized by a display device such as a display. In this case, the output device 950 presents information to the user by displaying various display information. Also, as another example, the output device 950 may be realized by an acoustic output device that outputs sounds such as voices or electronic sounds. In this case, the output device 950 presents information to the user by outputting sounds such as voices or electronic sounds. Also, the device applied as the output device 950 may be appropriately changed according to the medium used to present information to the user.

[0020] The input device 960 is used to receive various instructions from the user. In this embodiment, the input device 960 includes input devices such as a mouse, a keyboard, and a touch panel. As another example, the input device 960 may include a sound collection device such as a microphone, and may collect the sound uttered by the user. In this case, by performing various analysis processes such as acoustic analysis and natural language processing on the collected sound, the content indicated by this sound is recognized as an instruction from the user. Further, the device applied as the input device 960 may be appropriately changed according to the method of recognizing an instruction from the user. Also, a plurality of types of devices may be applied as the input device 960.

[0021] The network I / F 970 is used for communication with an external device via a network. Note that the device applied as the network I / F 970 may be appropriately changed according to the type of communication path and the applied communication method.

[0022] The CPU 910 expands the program stored in the ROM 920 or the auxiliary storage device 940 to the RAM 930 and executes this program, whereby the functional configuration of the information processing apparatus 100 shown in FIG. 9 and the processing of the information processing apparatus 100 described with reference to FIG. 13 are realized.

[0023] <Technical idea> With reference to FIGS. 3 to 8, an outline of the basic technical idea of the functions of the information processing system 1 according to this embodiment will be described below. First, with reference to FIG. 3, an outline of the functions of the information processing system 1 will be described. The information processing system 1 according to this embodiment refers to transaction data related to at least one of deposits and withdrawals managed in a financial database, and associates each transaction data with any one of a plurality of actions related to the economic activities of the user.

[0024] In the financial database, for example, as deposit history, purchase and operation history of financial products, cashless payment history, and common point history, etc., the history data of transactions generated along with the processing of deposits and withdrawals is managed. Hereinafter, the history data of transactions generated along with the processing of deposits is also referred to as "history data related to deposits", and the history data of transactions generated along with the processing of withdrawals is also referred to as "history data related to withdrawals".

[0025] The history data of transactions managed as deposit history includes the deposit balance of the target account, the history data related to deposits to the target account, and the history data related to withdrawals from the account. Also, in the information processing system 1 according to the present embodiment, for the history data related to withdrawals from the account, the history data related to withdrawals for investment purposes and the history data related to withdrawals corresponding to other uses different from investment are distinguished. For example, FIG. 4 is a diagram showing an example of the structure of a database of transaction data managed as deposit history. In the database shown in FIG. 4, a table for managing the deposit balance that fluctuates with deposits and withdrawals to the target account is associated with a table for managing the history of transactions related to the deposits and the withdrawals (for example, the amount, the date and time of deposits and withdrawals, etc.). Also, each transaction related to deposits and withdrawals is classified according to the purpose of the transaction. Thereby, for example, it becomes possible to identify whether the history of transactions related to withdrawals corresponds to the history of withdrawals corresponding to a predetermined use (for example, withdrawals for investment purposes) or the history of withdrawals corresponding to other uses other than the use. Note that the history data related to withdrawals for investment purposes corresponds to an example of "second history data related to withdrawals corresponding to a predetermined use", and the history data related to withdrawals corresponding to other uses different from investment corresponds to an example of "third history data related to withdrawals corresponding to other uses other than the predetermined use". Also, the history data related to deposits corresponds to an example of "first history data".

[0026] The historical data of transactions managed as the purchase and operation history of financial products includes historical data related to withdrawals for the purpose of purchasing and operating financial products. The historical data related to withdrawals for the purpose of purchasing and operating this financial product is classified as historical data related to withdrawals for investment purposes (second historical data). For example, FIG. 5 is a diagram showing an example of the structure of a database of transaction data managed as the purchase and operation history of financial products. In the database shown in FIG. 5, information indicating the financial products to be operated, the type of operation (for example, Order , holding, selling, purchasing, reservation, etc.), the amount applied to the operation, and information such as the trading date and time are managed as the purchase and operation history of financial products.

[0027] The historical data of transactions managed as the cashless settlement history includes historical data related to withdrawals associated with the use of cashless settlement. The historical data related to withdrawals associated with the use of this cashless settlement is classified as historical data related to withdrawals for other purposes different from investment (third historical data). In addition, the historical data of transactions managed as the common point history includes historical data associated with the use of so-called common points applied in a point program (point service) that can be used by multiple affiliated companies. The historical data associated with the use of this common point is classified as historical data related to withdrawals for other purposes different from investment (third historical data). For example, FIG. 6 is a diagram showing an example of the structure of a database of transaction data managed as the cashless settlement history or the common point history. In the database shown in FIG. 6, information indicating the balance, the type of points, the date and time of income and expenditure, etc. are managed as the usage history.

[0028] As shown in FIG. 3, the information processing system 1 associates each piece of historical data related to deposits and withdrawals managed in the above-described financial database with any one of consumption behavior, savings behavior, and investment behavior, which are actions related to the user's economic activities. Specifically, the information processing system 1 associates the history data related to deposits with "savings behavior". That is, the history data related to deposits that are managed as deposit histories will be associated with savings behavior. In addition, the information processing system 1 associates the history data related to withdrawals for investment purposes among the history data related to withdrawals with "investment behavior". That is, the history data related to withdrawals for investment that are managed as deposit histories, and the history data related to withdrawals for the purpose of purchasing and operating financial products that are managed as the purchase and operation histories of financial products will be associated with investment behavior. In addition, the information processing system 1 associates the history data related to withdrawals corresponding to other uses different from investment among the history data related to withdrawals with "consumption behavior". That is, the history data related to withdrawals corresponding to other uses different from investment that are managed as deposit histories will be associated with consumption behavior. Also, the history data related to withdrawals managed as cashless settlement histories and the history data related to withdrawals managed as common point histories will also be associated with either investment behavior or consumption behavior according to whether it is for investment purposes or not. As a specific example, when common points are transferred to investment-related products or used for automatic allocation to the purchase of financial products, the target history data will be associated with investment behavior.

[0029] The information processing system 1 determines the behavioral characteristics of the target user based on the changes over time of each behavior (savings behavior, investment behavior, and consumption behavior) associated with the history data related to deposits and withdrawals. For example, FIG. 7 is a diagram showing an example of a determination result of behavior characteristics based on changes in a time series of a plurality of behaviors related to a user's economic activities. In the example shown in FIG. 7, four cases are shown as (a) to (d). Further, in each case shown as (a) to (d), the left table shows, as numerical values, the degrees of consumption behavior, investment behavior, and savings behavior for each month in the target period (the period from January to June). The middle table shows an example of the analysis result of the user's behavior characteristics based on the changes along the time series of consumption behavior, investment behavior, and savings behavior respectively, and the result of plotting this analysis result for each month is shown as a graph on the right side. Specifically, Data A is data obtained by subtracting the total accumulated value of consumption behavior from the total accumulated value of savings behavior and then further subtracting the total accumulated value of investment behavior. Data B is data showing the total accumulated value of investment behavior for each month. Data C is data showing the total value of consumption behavior for each month.

[0030] For example, in the samples illustrated as (a) and (b), both Data A and Data B are fluctuating around 0. In such a case, since savings and consumption are in balance and no investment is being made, it can be inferred that the target user has no margin in life. Also, in the sample illustrated as (c), although Data B is fluctuating around 0, Data A is rising linearly. In such a case, since savings exceed consumption and no investment is being made, it can be inferred that the target user is saving money through non-investment means such as a regular savings account. Also, in the sample illustrated as (d), both Data A and Data B are rising linearly. In such a case, it can be seen that savings exceed consumption and investment is also being made. Further, since savings exceed investment, it can be inferred that the target user has a margin in life even after making investments.

[0031] In addition, regarding the data of investment behavior, it may be possible to distinguish for what types of financial products and how much investment has been made. As a specific example, as historical data related to the purchase and operation of financial products, when the classification of the financial products to be operated is managed, it becomes possible to distinguish for what types of financial products and how much investment has been made for the target investment behavior. For example, FIG. 8 is a diagram showing an example of a case where data on investment behavior is managed by distinguishing it for each type of financial product to be operated. In the example shown in FIG. 8, the data on investment behavior is managed separately for "yen-denominated bonds", "foreign-currency-denominated bonds", "domestic stocks", "overseas stocks", "investment trusts", "gold", "FX", "REIT", and "others", which are the types of financial products to be operated. Also, in the example shown in FIG. 8, the data on investment behavior for which the type of financial product to be operated could not be specified is classified as "unknown". That is, in the example shown in FIG. 8, the data on investment behavior corresponding to the historical data on withdrawals related to the purpose of investment is managed by distinguishing it for each more detailed purpose classified from the said purpose (for example, the purpose for each type of financial product as the investment destination). Note that in the example shown in FIG. 8, the explanation is mainly focused on financial products, but for financial education content, etc., it is also possible to manage them by distinguishing according to the type and content of the said content.

[0032] And as shown in FIG. 3, the information processing system 1 extracts candidates for the target user's attraction destination (for example, financial products, financial education content, etc.) according to the determination result of the behavior characteristics of the target user, and then outputs information indicating the candidates for the attraction destination to a predetermined output destination.

[0033] Note that the conditions for extracting candidates for the attraction destination according to the behavior characteristics (in other words, the correspondence between the behavior characteristics and the candidates for the attraction destination) are constructed according to the history of the use of each candidate for the attraction destination by each user and the behavior characteristics of the said user. As a specific example, a plurality of users who have had experience in financial activities (such as purchasing and managing financial products and using financial education content) over a certain period in the past are used as samples, and they are asked to evaluate candidates for attractive targets with which they have experience. Then, for example, the candidates for attractive targets to be evaluated are managed in association with data indicating the behavioral characteristics of the samples that have performed the evaluation. As a result, for example, for candidates for attractive targets that have received good evaluations, other users who exhibit the same behavioral characteristics as the samples that are the subjects of the evaluations are likely to receive similarly good evaluations, and thus can become more suitable candidates for attractive targets for such other users. That is, in order to be able to extract candidates for attractive targets that are more suitable for the target user according to the behavioral characteristics of the target user, it is sufficient to manage the candidates for attractive targets to be evaluated in association with data indicating the behavioral characteristics of the samples that have given good evaluations to the candidates for attractive targets.

[0034] In addition, a learned model constructed based on so-called machine learning may be used to extract candidates for attractive targets according to behavioral characteristics. In this case, for example, teacher data in which information indicating the candidates for attractive targets (such as financial products and financial education content) evaluated by the samples is attached as a label (correct label) may be used for the data indicating the behavioral characteristics of the samples to construct a learned model. As a result, by inputting data indicating the behavioral characteristics of the target user (data indicating changes along the time series of consumption behavior, investment behavior, and savings behavior) into the learned model, it becomes possible to obtain, as output, information indicating candidates for attractive targets that conform to the behavioral characteristics.

[0035] As described above, the information processing system according to the present embodiment analyzes at least one of the deposit and withdrawal history data to determine the behavioral characteristics of the target user, and extracts attractive targets that conform to the behavioral characteristics. As a result, for example, it becomes possible to present more suitable candidates for products and services (that is, candidates for attractive targets) to the target user. Therefore, hereinafter, the configuration and processing of the information processing system according to the present embodiment will be described in more detail.

[0036] <Functional configuration> Referring to FIG. 9, an example of the functional configuration of the information processing system 1 according to the present embodiment will be described, focusing particularly on the configuration of the information processing apparatus 100. Hereinafter, it is assumed that a learned model constructed based on machine learning is used for determining the behavior characteristics of the target user and extracting candidates for the attraction destination according to the behavior characteristics.

[0037] The information processing apparatus 100 includes an analysis unit 101, a teacher data generation unit 102, a learning processing unit 103, a storage unit 104, an extraction processing unit 105, and an output control unit 106.

[0038] The analysis unit 101 analyzes history data related to at least one of deposits and withdrawals, and associates the history data with any one of a plurality of actions related to the economic activities of the user. In the present embodiment, as described above, the analysis unit 101 associates each history data related to deposits and withdrawals with any one of consumption behavior, savings behavior, and investment behavior.

[0039] For example, FIG. 10 is a diagram showing an example of history data. Specifically, FIG. 10(a) shows an example of history data associated with consumption behavior, specifically, an example of history data related to a withdrawal (i.e., a withdrawal) from a predetermined account. Further, FIG. 10(b) shows an example of history data associated with savings behavior, specifically, an example of history data related to a deposit or transfer to a predetermined account. In the examples shown in FIGS. 10(a) and 10(b), information indicating the date on which the deposit or withdrawal was made, the content of the procedure (either deposit or withdrawal), and the amount, etc. is recorded. By using this information, it is possible to identify which of deposits and withdrawals the target history data corresponds to, and thus it is possible to identify which of consumption behavior and savings behavior the history data is associated with. Also, by using the above information, it is possible to identify the timing at which the deposit or withdrawal was made, that is, the timing at which the savings behavior or consumption behavior was performed, and the amount of the deposit or withdrawal, etc.

[0040] In addition, FIGS. 10(c) to 10(f) show an example of historical data associated with investment behavior for each type of financial product. Specifically, FIG. 10(c) shows an example of historical data related to the purchase and operation of yen-denominated bonds. FIG. 10(d) shows an example of historical data related to the purchase and operation of foreign-currency-denominated bonds. FIG. 10(e) shows an example of historical data related to the purchase and operation of domestic stocks. FIG. 10(f) shows an example of historical data related to the purchase and operation of overseas stocks. Note that FIGS. 10(c) to 10(f) are merely examples, and the data structure of the historical data may be appropriately changed according to the type of financial product, financial education content, etc. targeted. In the examples shown in FIGS. 10(c) to 10(f), information indicating the date on which the investment (purchase or operation) was made, the amount of money operated, and the object of the investment, etc. is recorded. By using this information, it becomes possible to associate the target historical data with investment behavior, or to identify the financial products, financial education content, etc. that are the targets of the investment behavior. Also, by using the above information, it becomes possible to identify the timing when the investment behavior was performed, the amount of money operated, etc.

[0041] The teacher data generation unit 102 generates teacher data used for constructing a learned model related to the extraction of attractive destinations according to the behavior characteristics of the target user, based on the analysis results of the historical data of each sample by the analysis unit 101. Specifically, the teacher data generation unit 102 associates, for each sample (user), data indicating the change along the time series of each of a plurality of behaviors related to the economic activities of the sample (in other words, data indicating behavior characteristics) with information indicating candidates for attractive destinations evaluated by the sample as a label, thereby generating teacher data. In the above manner, the teacher data generation unit 102 generates teacher data for each of a series of samples, and outputs the teacher data to the learning processing unit 103.

[0042] The learning processing unit 103 acquires the teacher data generated for each of a series of samples from the teacher data generation unit 102, and constructs a model based on machine learning using the teacher data. In the above manner, by inputting data indicating changes along the time series of a plurality of actions related to the economic activities of the target user (i.e., consumption actions, savings actions, and investment actions), the action characteristics of the user are determined, and a learned model that outputs candidates for attractive destinations according to the determination result of the action characteristics is constructed. The data according to the construction result of the learned model (i.e., the data according to the learning result of the model) is stored in the storage unit 104. The storage unit 104 schematically shows a storage area for holding various data.

[0043] Here, with reference to FIG. 11, an example of the learning result of the relationship between the action characteristics of the user and the candidates for attractive destinations to be extracted will be described. Note that the features shown as data A, data B, and data C in the example shown in FIG. 11 correspond to the data A, data B, and data C described above with reference to FIG. 7. Specifically, in the example shown in FIG. 11, at least one of the features of a series of features including the features of data A, data B, and data C and the features of combinations thereof, the action characteristics corresponding thereto, and the candidates for attractive destinations used by the user (sample) showing the action characteristics are associated. Further, for each of the plurality of features, when determining the action characteristics of the user by combining the plurality of features, it may be set whether to use an AND condition or an OR condition for the relationship with other features.

[0044] As a specific example, focusing on the candidate for an attractive destination, "Cumulative NISA (○○ Securities)", it can be seen that samples satisfying the conditions of "Feature of A = changing over 100,000 yen", "Feature of B = changing at 0 yen", or "Feature of C = smaller than income" and also satisfying the condition of "Feature of A×C = changing with A>C" are being used. Also, as another example, focusing on "Stock account opening (OO Securities)", which is a candidate for the target of attraction, it can be seen that samples are used that satisfy the conditions of "Characteristic of C = smaller than income" or "Investment ratio of investment behavior = 25% or more of investment trusts", and also satisfy the conditions of "Characteristic of A = changing at 500,000 yen or more" and "Characteristic of A×C = changing with A > C". By utilizing the characteristics as described above, for example, by matching the data indicating the behavioral characteristics of the target user with the data indicating the behavioral characteristics of the samples associated with each candidate for the target of attraction, it becomes possible to extract the candidates for the target of attraction that utilized samples showing behavioral characteristics similar or analogous to those of the target user. That is, by inputting data indicating the behavioral characteristics of the target user (in other words, data showing changes along the time series of each of a plurality of behaviors related to economic activities) into the constructed learned model, it becomes possible to cause the learned model to output candidates for the target of attraction that utilized samples showing substantially the same behavioral characteristics.

[0045] The extraction processing unit 105 extracts candidates for the target of attraction according to the behavioral characteristics of the target user (for example, the user targeted for attraction to financial products, financial education content, etc.) by utilizing the characteristics described above. Specifically, the extraction processing unit 105 acquires data showing changes along the time series of each of a plurality of behaviors related to the economic activities of the user (that is, consumption behavior, savings behavior, and investment behavior) according to the analysis result of the history data of the target user from the analysis unit 101. The extraction processing unit 105 inputs the data acquired from the analysis unit 101 to the learned model in which the data is held in the storage unit 104, thereby causing the learned model to output candidates for the target of attraction according to the behavioral characteristics of the target user. In the above manner, the extraction processing unit 105 extracts candidates for the target of attraction according to the behavioral characteristics of the target user, and outputs the data indicating the extracted candidates for the target of attraction to the output control unit 106.

[0046] For example, FIG. 12 shows an example of candidates for attractions extracted according to the behavior characteristics of a user. In the example shown in FIG. 12, not only financial products but also financial education content are included as extraction targets for candidates for attractions. Generally, when purchasing a financial product, it is expected that by sufficiently educating about risks and the like associated with the product, more returns can be obtained or losses can be reduced. Also, regarding education, for investment activities where education is received through some content, other education for avoiding risks associated with the investment activities is provided as a set, making it possible to conduct more effective education. According to the information processing system according to this embodiment, it is possible to provide a user with a set of candidates for attractions that systematically organize the above-described content according to the behavior characteristics of the user.

[0047] The output control unit 106 acquires data indicating candidates for attractions according to the behavior characteristics of the target user from the extraction processing unit 105 and outputs the data to a predetermined output destination. As a specific example, the output control unit 106 may output the above data to an output device such as a display, and notify an operator (for example, an operator who operates the system) of information indicating candidates for attractions according to the behavior characteristics of the target user via the output device. Also, as another example, the output control unit 106 may output the above data to an information processing device (for example, a server, etc.) that performs various analyses. Thereby, the information processing device can use the extraction result of candidates for attractions according to the behavior characteristics of the target user for various analyses.

[0048] Note that the above configuration is merely an example, and the functional configuration of the information processing system 1 according to the present embodiment is not necessarily limited to the example shown in FIG. 9. For example, the functional configuration of the information processing apparatus 100 described with reference to FIG. 9 may be realized by a plurality of apparatuses cooperating with each other. As a specific example, the functions of some of the components of the information processing apparatus 100 may be realized by other apparatuses. Further, as another example, the processing load of at least some of the components of the information processing apparatus 100 may be distributed among a plurality of apparatuses. Also, the functions of at least some of the components of the information processing apparatus 100 may be realized as so-called network services typified by cloud services. As described above, with reference to FIG. 9, an example of the functional configuration of the information processing system 1 according to the present embodiment has been described, particularly focusing on the configuration of the information processing apparatus 100.

[0049] <Processing> With reference to FIG. 13, an example of the processing of the information processing system according to the present embodiment will be described. Particularly focusing on the processing of the information processing apparatus 100, the processing related to the construction of the learned model and the processing related to the extraction of the attraction destination using the learned model will be separately described.

[0050] First, with reference to FIG. 13(A), an example of the processing related to the construction of the learned model will be described. In S101, the analysis unit 101 analyzes the history data related to at least either the deposit or withdrawal of a sample of a plurality of users who have had experience in financial activities in a past certain period. Then, based on the result of the analysis, the analysis unit 101 associates each piece of history data with one of a plurality of actions related to the economic activities of the user, that is, either a consumption action, a savings action, or an investment action.

[0051] In S102, the teacher data generation unit 102 generates teacher data used for constructing a learned model related to extraction of candidate attraction destinations according to the behavior characteristics of the target user, based on the analysis results of the history data of each sample by the analysis unit 101 in S101. Specifically, for each sample, the teacher data generation unit 102 associates information indicating a candidate attraction destination evaluated by the sample as a label with data indicating changes along the time series of each of a plurality of behaviors related to the economic activities of the sample, thereby generating teacher data.

[0052] In S103, the learning processing unit 103 acquires the teacher data generated for each of a series of samples by the teacher data generation unit 102 in S102, and constructs a model based on machine learning using the teacher data. Thereby, a learned model that outputs candidate attraction destinations in accordance with the behavior characteristics is constructed by inputting data indicating the behavior characteristics of the target user (in other words, data indicating changes along the time series of each of a plurality of behaviors related to the economic activities of the user). Data corresponding to the construction result of the learned model is stored in the storage unit 104. Time series Next, with reference to FIG. 13(B), an example of the process related to extraction of an attraction destination using the learned model will be described.

[0053] In S201, the analysis unit 101 analyzes history data related to at least one of the deposits and withdrawals of the user who is the target of extraction of candidate attraction destinations. Then, based on the result of the analysis, the analysis unit 101 associates each history data with a plurality of behaviors related to the economic activities of the user, that is, any one of consumption behavior, savings behavior, and investment behavior. Thereby, for example, by targeting the history data for a specified period, it becomes possible to generate data indicating changes along the time series of each of the consumption behavior, savings behavior, and investment behavior of the target user during the period.

[0054] ​In S202, the extraction processing unit 105 inputs data corresponding to the analysis result of the history data of the target user by the analysis unit 101 in S201 into the learned model in which the data is held in the storage unit 104. Thereby, the learned model determines the behavior characteristics of the target user, and information indicating the extraction result of candidates for the attraction destination according to the determination result of the behavior characteristics is output from the learned model. As described above, the extraction processing unit 105 extracts candidates for the attraction destination according to the behavior characteristics of the target user.

[0055] In S203, the output control unit 106 outputs data indicating candidates for the attraction destination according to the behavior characteristics of the target user, extracted by the extraction processing unit 105 in S202, to a predetermined output destination. As a specific example, the output control unit 106 may output the above data to an output device such as a display, and notify the operator of information indicating candidates for the attraction destination according to the behavior characteristics of the target user via the output device.

[0056] As described above, with reference to FIG. 13, an example of the processing of the information processing system according to the present embodiment has been described separately for the processing related to the construction of the learned model and the processing related to the extraction of the attraction destination using the learned model, paying particular attention to the processing of the information processing apparatus 100.

[0057] <Modification Example> A modification example of the information processing system according to the present embodiment will be described below.

[0058] (Modification Example 1) First, a first modification example of the information processing system according to the present embodiment will be described. Among the candidates for the target of solicitation, like some financial products, there are those with age restrictions regarding use. As a specific example, while there is no age restriction for opening a stock account, there are cases where age restrictions are individually set for each type of financial product, such as starting from 18 years old for opening a virtual currency account and starting from 20 years old for opening a foreign exchange account. In view of such a situation, in this modification example, an example of a mechanism for controlling the extraction target by taking into account the characteristics and attributes (e.g., age) of the user when extracting candidates for the target of solicitation according to the determination result of the behavioral characteristics of the target user will be described.

[0059] For example, FIG. 14 is a diagram showing an example of the learning result of the relationship between the behavioral characteristics of a user and the candidates for the target of solicitation in this modification example. Note that the features shown as data A, data B, and data C in the example shown in FIG. 14 correspond to the data A, data B, and data C described above with reference to FIG. 7, similar to the example shown in FIG. 11. In the example shown in FIG. 14, it is different from the example shown in FIG. 11 in that an age condition is provided. In this modification example, when the extraction processing unit 105 extracts candidates for the target of solicitation of the target user, if a candidate with an age condition set is included in a series of candidates, the extraction target is determined by collating the age condition with the age of the user. As a result, it is possible to realize a filtering process (that is, a filtering process for restricting the extraction target) for excluding candidates for the target of solicitation whose age of the target user does not meet the condition from the extraction target.

[0060] In addition, if it is possible to control (for example, limit) candidates for attractive destinations to be extracted according to the age of the target user, the configuration and method therefor are not particularly limited. As a specific example, for the extraction of candidates for attractive destinations according to the behavioral characteristics of the target user, a learned model performs the extraction, and filtering processing based on age conditions may be separately applied to a series of candidates for attractive destinations output from the learned model. As another example, if it is possible to construct a learned model such that constraints for extracting candidates for attractive destinations are applied taking into account age conditions, the extraction of candidates for attractive destinations finally output using the learned model (that is, candidates for attractive destinations taking into account age conditions) may be performed.

[0061] In addition, in the above, an example of the case of controlling candidates for attractive destinations to be extracted mainly according to age conditions has been described. However, not only age but also other characteristics and attributes of the user may be taken into account to control candidates for attractive destinations to be extracted. As a specific example, control (for example, limitation) of candidates for attractive destinations to be extracted may be performed taking into account the gender, nationality, etc. of the target user.

[0062] As described above, with reference to FIG. 14, Modification Example 1 of the information processing system according to the present embodiment has been described.

[0063] (Modification Example 2) Next, Modification Example 2 of the information processing system according to the present embodiment will be described. In the above-described embodiment, an example of the case of extracting candidates for attractive destinations suitable for the user according to the behavioral characteristics of the target user has been described. On the other hand, it is also conceivable that among the candidates for attractive destinations, there are candidates for attractive destinations for which it is not desirable to attract the target user, or candidates for attractive destinations for which the attraction of the user should be restricted (and thus candidates for which it is desirable to prohibit the attraction). Therefore, in this modification example, an example of a mechanism for extracting candidates for attractive destinations for which it is not desirable to attract the target user or candidates for attractive destinations for which the attraction of the user should be restricted according to the behavioral characteristics of the target user will be described.

[0064] First, an example of a mechanism for extracting candidates for an attraction target that are not desirable for attracting a user from among the candidates for the attraction target according to the behavior characteristics of the target user will be described. In this case, first, as in the above-described embodiment, a plurality of users who have had experience in financial activities in a certain past period are used as samples, and the candidates for the attraction target with utilization experience are evaluated. Then, for example, the candidate for the evaluation target and the data indicating the behavior characteristics of the sample that gave a bad evaluation to the candidate for the attraction target may be managed in association with each other. Further, when using a learned model for extracting candidates for the attraction target, teacher data in which information indicating the candidates for the attraction target that the sample gave a bad evaluation is given as a label to the data indicating the behavior characteristics of the sample may be used to construct the learned model.

[0065] For example, FIG. 15 is a diagram showing an example of the learning result of the relationship between the behavior characteristics of the user and the candidates for the attraction target to be extracted in this modified example. Note that the features shown as data A, data B, and data C in the example shown in FIG. 15 correspond to the data A, data B, and data C described above with reference to FIG. 7, similar to the example shown in FIG. 11. As a specific example, focusing on the candidate for the attraction target, "Domestic Stock MIX Investment Trust (×□ Trust Bank)", it can be seen that samples satisfying the conditions of "Characteristic of B = changing at 0 yen" or "Characteristic of A×C = changing with A<C" and satisfying the condition of "Investment ratio of investment behavior = 10% of investment trust" are being used. That is, for the above candidate for the attraction target, since users showing the same behavior characteristics as above are likely to receive a bad evaluation, it can become a candidate for the attraction target that is not preferable for attracting the user. Above That is, for the above candidate for the attraction target, since users showing the same behavior characteristics as above are likely to receive a bad evaluation, it can become a candidate for the attraction target that is not preferable for attracting the user.

[0066] As described above, by associating and managing candidates for the target of attraction and data indicating the behavioral characteristics of samples that have given a negative evaluation to the candidates for the target of attraction, it becomes possible to extract candidates for the target of attraction that are not preferable for attracting the target user, using data indicating the behavioral characteristics of the target user as input.

[0067] Next, an example of a mechanism for extracting candidates for the target of attraction that should restrict the attraction of the target user (and thus candidates for which it is desirable to prohibit attraction) from among the candidates for the target of attraction according to the behavioral characteristics of the target user will be described. In this case, a plurality of users (a plurality of users listed on the so-called blacklist) whose accident information has been recorded in a credit information institution (for example, CIC, JICC, KSC etc.) in a certain past period are used as samples, and the candidates for the target of attraction with which they have experience of use are selected. At this time, among a series of candidates for the target of attraction, financial products of a predetermined type (for example, loan-based financial products, etc.) or financial education content may be excluded from the selection target. Then, for example, the selected candidates for the target of attraction and the data indicating the behavioral characteristics of the samples for which the candidates for the target of attraction were the selection target may be associated and managed. Also, when using a learned model for extracting candidates for the target of attraction, teacher data in which information indicating the candidates for the target of attraction selected by the sample is given as a label may be used for the data indicating the behavioral characteristics of the sample to construct the learned model.

[0068] For example, FIG. 16 is a diagram showing another example of the learning result of the relationship between the behavioral characteristics of a user and the candidates for the target of extraction in this modification. The features shown as data A, data B, and data C in the example shown in FIG. 16 correspond to data A, data B, and data C described above with reference to FIG. 7, similar to the example shown in FIG. 11. As a specific example, focusing on "Temporary FX Account (○○ Securities)", which is a candidate for the target of solicitation, it can be seen that samples are selected that meet the conditions of "Characteristic of A = fluctuating below 100,000 yen", "Characteristic of C = smaller than income", "Characteristic of A×B = in any consecutive six months, there are four or more months where B < A", and "Investment behavior input ratio = 10% or less for domestic stocks". That is, regarding the above candidate for the target of solicitation, since it is conceivable that an accident may occur when a user showing the same behavioral characteristics as above uses it, it can be a candidate for the target of solicitation for which the solicitation of the user should be restricted.

[0069] As described above, by associating and managing the candidate for the target of solicitation to be evaluated and the data indicating the behavioral characteristics of the sample in which the candidate for the target of solicitation is selected from the samples in which accident information is recorded, it is possible to extract the candidate for the target of solicitation for which the solicitation of the target user should be restricted, using the data indicating the behavioral characteristics of the target user as input.

[0070] As described above, by learning the relationship between each candidate for the target of solicitation and the behavioral characteristics of the user according to the conditions related to the relationship between the user and the candidate for the target of solicitation, it is possible to control the candidate for the target of solicitation to be extracted according to the purpose of use of the extraction result. As described above, with reference to FIGS. 15 and 16, Modification Example 2 of the information processing system according to the present embodiment has been described.

[0071] <Conclusion> As described above, in the information processing system according to the present embodiment, the information processing device associates each series of history data related to at least one of the user's deposits and withdrawals with any one of a plurality of behaviors related to the user's economic activities. Further, the information processing device extracts candidates for the target of solicitation according to the behavioral characteristics of the user based on the changes along the time series of each of the above plurality of behaviors. Moreover, the information processing device outputs information indicating the extracted candidates for the target of solicitation to a predetermined output destination. With the configuration as described above, it becomes possible to present, to the target user, an attraction destination (for example, a candidate that it is desirable to attract the user, a candidate that it is not desirable to attract the user, and a candidate for which the attraction of the user should be restricted, etc.) that conforms to the characteristics of the user's behavior.

[0072] Note that the following configurations also belong to the technical scope of the present disclosure. (1) An information processing apparatus including: an association means for associating each of a series of history data related to at least any one of the user's deposits and withdrawals with any one of a plurality of actions related to the user's economic activities; an extraction means for extracting candidates for attraction destinations according to the user's behavior characteristics based on changes along the time series of each of the plurality of actions; and an output means for outputting information indicating the extracted candidates for attraction destinations to a predetermined output destination. (2) The information processing apparatus according to (1), wherein the association means associates the first history data related to a deposit, the second history data related to a withdrawal corresponding to a predetermined use, and the third history data related to a withdrawal corresponding to another use other than the predetermined use with different actions. (3) The information processing apparatus according to (2), wherein the extraction means extracts candidates for attraction destinations according to the user's behavior characteristics, taking into account a more detailed use classified from the predetermined use. (4) The information processing apparatus according to (3), wherein the predetermined use includes at least any one of use for investment in financial products and use of financial education content, and the extraction means extracts at least any one of candidates for attraction destinations of financial products and financial education content as candidates for attraction destinations according to the user's behavior characteristics. (5) The information processing apparatus according to any one of (1) to (4), wherein the extraction means extracts candidates for attraction destinations according to the user's behavior characteristics, taking into account conditions related to the relationship between the user and the candidates for attraction destinations. (6) The extraction means extracts at least one of candidates that are desirable to attract the user, candidates that are not desirable to attract the user, and candidates for which the attraction of the user should be restricted, as candidates for an attraction destination according to the relationship between the user and the candidates for the attraction destination according to the behavior characteristics of the user, for the information processing apparatus according to (5). (7) The extraction means extracts candidates for an attraction destination according to the behavior characteristics of the user, taking into account the characteristics of the target user, for the information processing apparatus according to any one of (1) to (6). (8) The extraction means restricts the candidates for the attraction destination to be extracted according to the age of the user, for the information processing apparatus according to (7). (9) The extraction means inputs data indicating changes along the time series of each of the plurality of actions into a learned model constructed based on supervised learning using teacher data in which information indicating the candidate for the attraction destination used by the user who is the subject of the plurality of actions is associated as a correct label with respect to the data indicating changes along the time series of each of the plurality of actions, thereby extracting candidates for the attraction destination according to the behavior characteristics of the user, for the information processing apparatus according to any one of (1) to (8). (10) The extraction means inputs data indicating changes along the time series of each of the plurality of actions of the user, which indicate a predetermined characteristic regarding the relationship with the candidate for the attraction destination, into a learned model constructed based on supervised machine learning using teacher data in which information indicating the candidate for the attraction destination used by the user is associated as a correct label, for the learned model corresponding to the predetermined characteristic, thereby extracting candidates for the attraction destination according to the behavior characteristics of the user, taking into account the predetermined characteristic regarding the relationship with the target user, for the information processing apparatus according to (9). An information processing method executed by an information processing apparatus, comprising: an association step of associating each of a series of history data related to at least one of a user's deposits and withdrawals with any one of a plurality of actions related to the user's economic activities; an extraction step of extracting candidates for an attraction destination according to the user's behavior characteristics based on changes along the time series of each of the plurality of actions; and an output step of outputting information indicating the extracted candidates for the attraction destination to a predetermined output destination. A program for causing a computer to execute: an association step of associating each of a series of history data related to at least one of a user's deposits and withdrawals with any one of a plurality of actions related to the user's economic activities; an extraction step of extracting candidates for an attraction destination according to the user's behavior characteristics based on changes along the time series of each of the plurality of actions; and an output step of outputting information indicating the extracted candidates for the attraction destination to a predetermined output destination.

Explanation of Signs

[0073] 1 Information processing system 100 Information processing apparatus 101 Analysis unit 102 Teacher data generation unit 103 Learning processing unit 104 Storage unit 105 Extraction processing unit 106 Output control unit 200 Terminal device

Claims

1. For each transaction related to at least one of the user's deposits and withdrawals, a series of historical data including the amount and date of the transaction is classified based on the purpose of the transaction, and the classified historical data is associated with any one of consumption behavior, savings behavior, and investment behavior, which are a plurality of actions related to the user's economic activities, and an association means; For data indicating changes along the time series of each of the plurality of actions, based on supervised learning using teacher data in which information indicating candidates for the attraction destination used by the user who is the subject of the plurality of actions is associated as the correct label with respect to a trained model constructed based on supervised learning, by inputting data indicating changes along the time series of each of the plurality of actions associated with each of the series of historical data, an extraction means for extracting candidates for the attraction destination according to the behavior characteristics of the user; An output means for outputting information indicating the extracted candidate for the attraction destination to a predetermined output destination; An information processing apparatus comprising:

2. The association means associates the first historical data related to the deposit, the second historical data related to the withdrawal corresponding to a predetermined purpose, and the third historical data related to the withdrawal corresponding to other purposes other than the said purpose with different actions. The information processing apparatus according to claim 1.

3. The extraction means extracts candidates for the attraction destination according to the behavior characteristics of the user, taking into account more detailed purposes classified from the said predetermined purpose. The information processing apparatus according to claim 2.

4. The said predetermined purpose includes at least one of investment in financial products and use of financial education content. The extraction means extracts at least one of candidates for financial products and financial education content as candidates for the attraction destination according to the behavior characteristics of the user. The information processing apparatus according to claim 3.

5. The extraction means extracts candidates for the attraction destination according to the behavior characteristics of the user, taking into account conditions related to the relationship between the user and the candidates for the attraction destination. The information processing apparatus according to claim 1.

6. The extraction means, as candidates for the attraction destination according to the behavior characteristics of the user, extracts at least one of candidates for which it is desirable to attract the user, candidates for which it is not desirable to attract the user, and candidates for which the attraction of the user should be restricted, according to conditions related to the relationship between the user and the candidates for the attraction destination. The information processing apparatus according to claim 5.

7. The extraction means extracts candidates for the attraction destination according to the behavior characteristics of the user, taking into account the characteristics of the target user, according to the information processing apparatus described in claim 1.

8. The extraction means restricts candidates for the attraction destination to be extracted according to the age of the user, according to the information processing apparatus described in claim 7.

9. The extraction means For data indicating changes over time in each of the plurality of behaviors of the user that exhibit a predetermined characteristic regarding the relationship with the candidate for the attraction destination, using teacher data in which information indicating the candidate for the attraction destination used by the user is associated as a correct label, for a learned model constructed based on supervised machine learning corresponding to the predetermined characteristic, By inputting data indicating changes over time in each of the plurality of behaviors associated with each of the series of history data, candidates for the attraction destination according to the behavior characteristics of the user, taking into account the predetermined characteristic regarding the relationship with the target user, are extracted. The information processing apparatus described in claim 1.

10. An information processing method executed by an information processing apparatus, An association step of classifying each of a series of history data including the amount and date of each transaction related to at least one of the user's deposits and withdrawals based on the purpose of the transaction, and associating the classified history data with any one of consumption behavior, savings behavior, and investment behavior, which are a plurality of behaviors related to the user's economic activities; An extraction step of extracting candidates for the attraction destination according to the behavior characteristics of the user by inputting data indicating changes over time in each of the plurality of behaviors associated with each of the series of history data into a learned model constructed based on supervised learning using teacher data in which information indicating the candidate for the attraction destination used by the user, who is the subject of the plurality of behaviors, is associated as a correct label; An output step of outputting information indicating the extracted candidates for the attraction destination to a predetermined output destination; An information processing method including:

11. On a computer, For each transaction related to at least one of the user's deposits and withdrawals, classify a series of historical data including the amount and date of the transaction based on the purpose of the transaction, and associate the classified historical data with any one of consumption behavior, savings behavior, and investment behavior, which are a plurality of actions related to the user's economic activities. Association step, For data showing changes along the time series of each of the plurality of actions, input data showing changes along the time series of each of the plurality of actions associated with each of the series of historical data into a trained model constructed based on supervised learning using teacher data in which information indicating candidates for attractive destinations used by the user who is the subject of the plurality of actions is associated as the correct label. An extraction step of extracting candidates for attractive destinations according to the behavior characteristics of the user, An output step of outputting information indicating the extracted candidates for attractive destinations to a predetermined output destination, A program that causes the above to be executed.

Images