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

Abstract

To predict a user's fundamental desire, and determine items in which the user has interest according to the prediction.SOLUTION: An information processing device is configured to: acquire a user feature of a user; predict a desire stage to which the user belongs, among a plurality of preset desire stages, by using a learning model, on the basis of the user feature; detect a life event generated in the user; and determines one or more items in which the user is interested, on the basis of the predicted desire stage, and the detected life event.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to an information processing device, an information processing method, and an information processing program, and in particular to a technique that applies prediction of a user's desires. [Background technology]

[0002] When recommending products to a user, it is important to accurately consider the user's needs. Patent Document 1 discloses a method for estimating a user's potential physiological needs from biological information such as the user's blood glucose level, and presenting the user with information about products and services based on the estimated physiological needs. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2016-071716 A Summary of the Invention [Problem to be solved by the invention]

[0004] According to the technology disclosed in the above document, it is possible to satisfy a user's potential desires by providing the user with information about products and services that the user potentially desires and supporting the user's actions. However, the technology in the literature is limited to estimating a limited range of needs, such as a user's physiological needs, from limited information, such as the user's biometric information, and is unable to estimate the user's deep psychological needs in a broader range. Therefore, it is unable to appropriately determine and recommend services or products that correspond to the user's fundamental needs.

[0005] The present invention has been made in consideration of the above problems, and has an object to predict a user's fundamental desires and determine items in which the user will be interested based on the prediction. [Means for solving the problem]

[0006] In order to solve the above problem, one aspect of an information processing device according to the present invention includes an acquisition means for acquiring user characteristics of a user, a prediction means for predicting, based on the user characteristics, which of a plurality of preset desire stages the user belongs to using a learning model, a detection means for detecting a life event that has occurred to the user, and a determination means for determining one or more items in which the user is interested based on the predicted desire stage and the detected life events.

[0007] In order to solve the above problem, one aspect of an information processing method according to the present invention includes an acquisition step of acquiring user characteristics of a user, a prediction step of predicting, using a learning model based on the user characteristics, which desire stage the user belongs to among a plurality of preset desire stages, a detection step of detecting a life event that has occurred to the user, and a determination step of determining one or more items in which the user is interested based on the predicted desire stage and the detected life events.

[0008] In order to solve the above problems, one aspect of the program according to the present invention is an information processing program for causing a computer to execute information processing, the program being for executing processes including an acquisition process for acquiring user characteristics of a user, a prediction process for predicting, based on the user characteristics, using a learning model, which of a plurality of pre-set desire stages the user belongs to, a detection process for detecting a life event that has occurred to the user, and a determination process for determining one or more items in which the user is interested, based on the predicted desire stage and the detected life events. Effect of the Invention

[0009] According to the present invention, it is possible to predict a user's fundamental desires and determine items in which the user will be interested based on the prediction. The above-mentioned objects, aspects, and advantages of the present invention, as well as objects, aspects, and advantages of the present invention not described above, will be understood by those skilled in the art from the following detailed description of the invention by referring to the accompanying drawings and the claims. [Brief description of the drawings]

[0010] [Figure 1] FIG. 1 shows an example of the configuration of an information processing system according to an embodiment. [Diagram 2] FIG. 2 illustrates an example of a functional configuration of an information processing device according to an embodiment. [Figure 3A] Figure 3A shows Maslow's hierarchy of needs. [Figure 3B] FIG. 3B shows an example of the association between each stage of Maslow's hierarchy of needs and related products and related life events. [Figure 4] Figure 4 shows a conceptual diagram of need stage prediction. [Figure 5A] FIG. 5A shows an example of an interest graph. [Figure 5B] FIG. 5B shows another example of an interest graph. [Figure 6] FIG. 6 shows a flowchart of the entire process executed by the information processing device according to the embodiment. [Figure 7] FIG. 7 illustrates an example of the hardware configuration of an information processing device and a user device according to the embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0011] Hereinafter, with reference to the attached drawings, an embodiment for carrying out the present invention will be described in detail. Among the components disclosed below, those having the same functions are given the same reference numerals, and their description will be omitted. Note that the embodiment disclosed below is an example of a means for realizing the present invention, and should be appropriately modified or changed depending on the configuration of the device to which the present invention is applied and various conditions, and the present invention is not limited to the following embodiment. Furthermore, not all of the combinations of features described in the present embodiment are necessarily essential to the solution of the present invention.

[0012] [Example of information processing system configuration] Fig. 1 shows an example of the configuration of an information processing system according to the present embodiment. As an example, as shown in Fig. 1, the information processing system includes a user device 11 used by a user u1 and an information processing device 10. In the following description, the terms user device and user may be used synonymously.

[0013] The user device 11 is, for example, a device such as a smartphone or a tablet, and is configured to be able to communicate with the information processing device 10 via a public network such as LTE (Long Term Evolution) or a wireless communication network such as a wireless LAN (Local Area Network). The user device 11 has a display unit (display surface) such as a liquid crystal display, and each user can perform various operations using a GUI (Graphic User Interface) equipped on the liquid crystal display. The operations include various operations on content such as images displayed on the screen, such as tapping, sliding, and scrolling using a finger or a stylus. Note that the user device 11 is not limited to the device in the form shown in Fig. 1, and may be a device such as a tablet terminal or a notebook PC. Also, the user device 11 may be provided with a separate display screen.

[0014] The user u1 can use web services (internet-related services) provided from the information processing device 10 or from another device (not shown) via the information processing device 10. These web services can be used by the user, for example, by registering for use with an API (Application Programming Interface) provided by the information processing device 10 or another device.

[0015] In this embodiment, the web services may include online malls, online supermarkets, or services related to communications, finance, real estate, sports, and travel. By using such web services, the user device 11 can transmit information about the user u1 to the information processing device 10. For example, the user device 11 can transmit to the information processing device 10 characteristic information regarding the user device 11 and the user u1, such as the IP (Internet Protocol) address of the user device 11, the address of the user u1, and the name of the user u1 (first and last name). In addition, the user device 11 performs positioning calculations based on signals received from GPS (Global Positioning System) satellites (not shown), generates the information obtained by the calculations as location information of the user device 11, and transmits it to the information processing device 10.

[0016] The information processing device 10 is configured to obtain various information from the user device 11 and predict the desires of the user u1 based on the information. In this embodiment, a trained learning model (desire stage prediction model 112 described later) is used to predict the desire stage to which the user u1 belongs among a plurality of preset desire stages. Furthermore, the information processing device 10 is configured to determine items (one or more of one or more products and one or more services) in which the user u1 is interested based on the predicted desire stage, and to recommend the determined items.

[0017] In this embodiment, an item may be a tangible or intangible product or service that can be provided for various services. For example, for financial (fintech) services, there are items such as bank accounts, financial products such as stocks, investment trusts, and insurance products, cryptocurrency, and smartphone app payments. For digital content services, there are items such as video content such as movies and animations, and still image content such as photos, illustrations, and text. For e-commerce services, there are items such as intangible or tangible products handled in online shopping. For travel services, there are items such as information and reservations on hotels, package tours, and transportation facilities. For mobile services, there are items such as mobile devices, public network / Internet connections, and communication usage fees. For advertising and media services, there are items such as offline and offline advertising products, direct mail, and advertising via broadcasting and the Internet. For card services, there are items such as credit card payments and point transactions. For sports and cultural services, there are items such as events such as sporting events and concerts, and products sold at events.

[0018] [Functional configuration of information processing device 10] FIG. 2 shows an example of a functional configuration of the information processing device 10 according to the present embodiment. 2 includes a user feature acquisition unit 101, a desire stage prediction unit 102, an interest graph management unit 103, an interest item prediction unit 104, a recommendation unit 105, a learning model storage unit 110, and a graph storage unit 120. The learning model storage unit 110 is configured to store a user feature prediction model 111 and a desire stage prediction model 112, which have been learned. The graph storage unit 120 is configured to store an interest graph 121.

[0019] The user feature acquisition unit 101 acquires factual features (factual information) about the user device 11 and the user u1 as user features from the user device 11. The user features are factual features (information) that are actually or objectively obtained from the user device 11 and the user u1. The user feature acquisition unit 101 can acquire the user features directly from the user device 11, for example. The user feature acquisition unit 101 can also acquire the user features as information registered in a web service by the user u1 of the user device 11. The user features are associated with a user ID that identifies the user.

[0020] User characteristics include the IP address of the user device, the user's address, the user's name, the credit card number held by the user, the user's demographic information (demographic user attributes such as gender, age, area of ​​residence, occupation, family composition, etc.), etc. The user characteristics may also include the user's registration number and registered name when using the web service. The user characteristics may also include call history, a delivery address for products when using the web service, and information regarding points that can be accumulated by using the service. The user characteristics may also include a user's behavioral history in the web service (a user's behavioral history when using the web service). The behavioral history may include a purchase history, a usage history (including whether or not a click was made, the number of clicks, and viewing time), a search history, and a registration history for the service with respect to items in the web service. In other words, the behavioral history may include information about items. The user characteristics may also include data indicating the user's preferences corresponding to the behavioral history. In this way, the user characteristics may include any information including information related to the user device or the user himself / herself, and information related to the use of web services via communication. The user characteristics acquisition unit 101 acquires such user characteristics for the user u1.

[0021] The user feature acquisition unit 101 is also configured to apply the acquired user features of the user u1 to the trained user feature prediction model 111 to acquire user features (estimated user features (attributes)) estimated for the user features. That is, the user features of the user u1 acquired by the user feature acquisition unit 101 may include the user features (actual features) of the user u1 and the estimated user features.

[0022] The user feature prediction model 111 is configured to receive user features (factual features) of user u1 as input and output the probability (correspondence probability) that each of a plurality of user features corresponds (matches) user u1. The plurality of user features include user features acquired from a plurality of users (not shown). The user feature acquisition unit 101 finally determines an estimated user feature of user u1 from the correlation probability.

[0023] For example, the user feature acquisition unit 101 inputs data indicating demographic information of the user u1 and purchase history and preferences in past web services as the user features of the user u1 to the user feature prediction model 111. From the user feature prediction model 111, the probability for each of a plurality of items estimated to be purchased by the user u1, a plurality of preferences that the user u1 may have, the behavioral tendency of the user u1, and the economic feature of the user u1 is output as the corresponding probability. Then, the user feature acquisition unit 101 can acquire the items, preferences, behavioral tendency, and / or economic feature having a probability equal to or greater than a predetermined value as the estimated user features of the user u1. The behavioral tendency includes, for example, a characteristic related to a life stage such as having a child, and a characteristic related to a lifestyle such as working from home. In addition, the economic feature indicates the income characteristic of the user u1 and / or the household of the user u1, and is classified as high income, medium income, or low income. The income threshold for classifying high income, medium income, and low income can be set arbitrarily.

[0024] The need stage prediction unit 102 predicts the need stage of the user u1 from the user characteristics of the user u1 acquired by the user characteristic acquisition unit 101. The need stage prediction unit 102 predicts the need stage to which the user u1 belongs from among a plurality of hierarchical need stages set in advance, using a need stage prediction model 112. One example of the plurality of hierarchical need stages is a hierarchical stage that represents human psychological needs. In this embodiment, Maslow's Hierarchy of Need Stages is used. Maslow's hierarchy of needs is a theorization of human needs into a five-stage hierarchy.

[0025] Here, we will explain Maslow's hierarchy of needs. Figure 3A shows Maslow's hierarchy of needs. According to Maslow's law, human needs are classified into five stages: (1) physiological needs 31, (2) safety needs 32, (3) social needs (belonging and love needs) 33, (4) esteem needs 34, and (5) self-actualization needs 35. As shown in Figure 3A, the five stages form a pyramid, with physiological needs 31 at the lowest (base) need stage and self-actualization needs 35 at the highest need stage.

[0026] Physiological needs 31 are the most basic needs and refer to the instinctive needs to maintain a minimum level of life. Examples of physiological needs 31 include eating, sleeping, and excretion. Generally, humans do not limit themselves to physiological needs 31, but other animals rarely exceed physiological needs 31. Safety needs 32 refer to the desire to live in an environment that is both physically and mentally safe and economically stable. Examples of safety needs 32 include the desire to obtain safety, economic security, good health, a good standard of living, and a safety net against illness and unexpected accidents. Social needs 33 refer to the desire to belong to some kind of social group (community) and feel a sense of security, and are sometimes called the "needs for belonging and love." Examples of social needs 33 include the desire to avoid loneliness, to be loved, to interact with others, and to avoid being left out. The need for recognition 34 is the desire to be recognized and to be an influential figure in the social group (community) to which one belongs. An example of the need for recognition 34 includes the desire to have one's posts on SNS (social network service) get "Likes." Self-actualization need 35 refers to the desire to accomplish things that only one can do and to realize the desires that come from within oneself. An example of self-actualization need 35 includes the desire to develop one's talents, invent something, and contribute to humanity, and to take on unknown challenges.

[0027] In this embodiment, the desire stage prediction unit 102 predicts the desire stage to which the user u1 belongs among the five desire stages shown in Fig. 3A by using the desire stage prediction model 112. The procedure for predicting the desire stage will be described later.

[0028] The interest graph management unit 103 manages the interest graph 121 stored in the graph storage unit 120. In this embodiment, the interest graph 121 is a graph that expresses the user's interests for each of the five desire stages shown in Fig. 3A. The interest graph 121 will be described later.

[0029] The interest item prediction unit 104 predicts and determines the interest of the user u1 based on the desire stage of the user u1 predicted by the desire stage prediction unit 102. The interest may be a current interest or an interest in the near future. In this embodiment, the "interest of the user u1" means "items in which the user u1 is interested (one or more products and one or more services)". In this embodiment, the interest item prediction unit 104 predicts the interest of the user u1 based on the interest graph 121, triggered by the occurrence of a life event for the user u1. The procedure for predicting the interest of the user u1 will be described later.

[0030] The recommendation unit 105 performs control for recommending to the user u1 items in which the user u1 is interested, which are predicted (determined) by the interest item prediction unit 104. For example, the recommendation unit 105 provides the user u1 with advertisements related to items in which the user u1 is interested. The provision may be any output process, and may be output to an external device via a communication I / F (communication I / F 77 in FIG. 7) or display on a display unit (display unit 76 in FIG. 7).

[0031] [Procedure for predicting desire stage] Next, a procedure for predicting the desire stage of user u1 by the desire stage prediction unit 102 will be described. In this embodiment, the desire stage prediction unit 102 applies the user features of user u1 acquired by the user feature acquisition unit 101 to a trained desire stage prediction model 112 to predict the desire stage of user u1. The desire stage prediction model 112 is a machine learning model, and is, for example, a learning model based on CatBoost. Alternatively, the desire stage prediction model 112 may be another boosting-based learning model such as XGBoost or LightGBM.

[0032] First, the learning process of the need stage prediction model 112 will be described. The learning process may be performed by one or more processing units in the information processing device 10, or may be performed by an external device. The need stage prediction model 112 is trained using learning data (teacher data) based on the association between each need stage and related items (related products or related services) that represent items related to each need stage. FIG. 3B shows an example of the association between each need stage of Maslow's need hierarchy shown in FIG. 3A and related items. The association also includes life events (related life events) that are closely related to each need stage.

[0033] Examples of related items corresponding to physiological needs 31 are medicine and mobile connectivity. It is assumed that medicine (medicine) is essential for living a healthy life, and that having mobile connectivity (i.e., having a communication device capable of communication) is essential in modern life. Note that in FIG. 3B, it is assumed that there are no related life events corresponding to physiological needs 31.

[0034] An example of a life event that corresponds to the need for safety is education (general higher education) or starting work after completing one's studies. It is based on the assumption that education and employment are life events that occur in order to live in an environment that is both physically and mentally safe and economically stable. Examples of items related to the need for security32 are insurance and investment. Insurance is based on the assumption that you need to purchase insurance in order to live a financially stable life even if you get sick or injured. Investment is based on the assumption that you need to make investments in order to live a financially stable life, regardless of your income.

[0035] Examples of related life events that correspond to social needs 33 are marriage and the birth of a child. It is based on the assumption that marriage and the birth of a child are life events that occur in order to satisfy the desire of users who have satisfied their physiological needs 31 and safety needs 32 to create a community and feel a sense of security. An example of a relevant item corresponding to the social need 33 is a low-priced car. For example, it is based on the assumption that it is convenient to own a car when the birth of a child occurs as a relevant life event corresponding to the social need 33.

[0036] Examples of related life events corresponding to the need for recognition 34 are promotion and overseas posting. Promotion and overseas posting are based on the assumption that they are life events that occur in order to satisfy the desire of a user whose social needs 33 have been satisfied to be recognized in the community to which he or she belongs. Luxury cars, real estate, and credit cards are examples of related items corresponding to the need for recognition 34. It is assumed that luxury cars, real estate, and credit cards are items that can express the high social status of the user himself / herself.

[0037] Examples of life events that correspond to the need for self-actualization 35 are adult education and a trip around the world. Adult education is based on the assumption that it is a life event that occurs in order to satisfy the desire to realize aspirations that arise from within a user whose need for recognition 34 has been satisfied. A trip around the world is based on the assumption that it is a life event that occurs in order to satisfy the desire to achieve something that only a user can do in order to satisfy the desire to achieve. Examples of related items corresponding to the need for self-actualization 35 are hobbies (items related to hobbies), travel, and education. It is based on the assumption that hobbies, travel, and education are items that allow a user whose need for recognition 34 has been satisfied to further identify himself / herself with an ideal self-image.

[0038] The association between each need stage, related life event, and related item shown in Fig. 3B may be generated based on the assumptions as described above, or may be generated based on a network based on conditional probabilities obtained from past purchase histories, such as a Bayesian network.

[0039] In the learning stage, user features of multiple users (not shown) are used. The user features of the multiple users include actual features and estimated features of the multiple users, similar to the user features of user u1 acquired by the user feature acquisition unit 101. The user features of the multiple users include behavioral history in the web service, and items included in the behavioral history (for example, items included in the purchase history or usage history) are associated with desire stages according to the association shown in FIG. 3B. Then, learning data combining the user features of the multiple users and the desire stages (correct answer data) is prepared, and the desire stage prediction model 112 is trained using the learning data. The trained desire stage prediction model 112 is stored in the learning model storage unit 110. The trained desire stage prediction model 112 is configured to input the user features of any user and output the probability (0 to 1) of each of the multiple desire stages to which the user may belong.

[0040] In this embodiment, the need stage prediction model 112 is configured to output a probability of falling into each of the five need stages shown in Fig. 3A. Specifically, the need stage prediction model 112 is configured to input the user characteristics of an arbitrary user and output a probability of falling into each of the need stages of physiological needs 31, safety needs 32, social needs 33, esteem needs 34, and self-actualization needs 35.

[0041] Next, the prediction process will be described. The need stage prediction unit 102 inputs the user characteristics of the user u1 acquired by the user characteristic acquisition unit 101 into the need stage prediction model 112, and predicts the need stage with the highest output probability as the need stage to which the user u1 belongs. In this embodiment, the need stage prediction unit 102 predicts the need stage with the highest probability of being included in each of the need stages of physiological needs 31, safety needs 32, social needs 33, esteem needs 34, and self-actualization needs 35 output from the need stage prediction model 112 as the need stage to which the user u1 belongs.

[0042] FIG. 4 shows an example of a conceptual diagram of desire stage prediction. In FIG. 4, the user features 40 acquired by the user feature acquisition unit 101 include demographic information 401, purchase history 402, and economic features 403. The demographic information 401 indicates demographic user attributes such as the gender, age, residential area, occupation, and family structure of the user u1. The purchase history 402 indicates the purchase history of the user u1 on the web service. The economic features 403 are included in the estimated user features estimated using the user feature prediction model 111 based on the user features (factual features) of the user u1 acquired by the user feature acquisition unit 101, as described above.

[0043] The need stage prediction unit 102 inputs the user characteristics 40 of the user u1 into a need stage prediction model 112, and predicts the need stage with the highest probability of being output from the model 112 as the need stage 41 to which the user u1 belongs. In this embodiment, the need stage prediction unit 102 predicts one of the need stages of physiological needs 31, safety needs 32, social needs 33, esteem needs 34, and self-actualization needs 35 as the need stage 41 to which the user u1 belongs. 4, the user characteristics 40 are configured to include the demographic information 401, the purchase history 402, and the economic characteristics 403, but are not limited to this. In other words, the user characteristics 40 may include user characteristics (actual characteristics or estimated characteristics) other than the demographic information 401, the purchase history 402, and the economic characteristics 403.

[0044] User characteristics, particularly the behavioral history of a user in a web service, may reveal trends when observed for a certain period of time (e.g., quarterly or semiannual). Therefore, the desire stage prediction unit 102 may be configured to predict the desire stage to which the user u1 belongs, using the user characteristics of the user u1 collected over a certain period of time. In addition, since the user's desires may change over time, the desire stage prediction unit 102 may be configured to periodically predict the desire stage to which the user u1 belongs.

[0045] [Example of interest graph] Next, the interest graph 121 managed by the interest graph management unit 103 will be described with reference to Fig. 5A and Fig. 5B. Initially, the interest graph 121 may be input by an operator via an input unit (input unit 75 in Fig. 7) or via a communication unit (communication I / F 77 in Fig. 7). Alternatively, the interest graph 121 may be set in advance in the system, or may be set by an arbitrary program stored in a storage unit (ROM 72 or RAM 73 in Fig. 7).

[0046] The interest graph 121 is created for each of a plurality of preset need stages. In this embodiment, the interest graph 121 is created for each need stage of physiological needs 31, safety needs 32, social needs 33, esteem needs 34, and self-actualization needs 35.

[0047] 5A shows an interest graph 121A as an example of the interest graph 121. The interest graph 121A is an example of the interest graph 121 when the desire stage of the user u1 predicted by the desire stage prediction unit 102 is social desire 33. As shown in FIG. 5A, the interest graph 121A is a directed graph configured to include a directed connection from a life event (life event node) 50 to an item of interest (item node) 51.

[0048] The interest graph 121A may be a graph constructed at least in part using a Bayesian network that expresses the causal relationship between a life event (cause) and an item of interest (result) by likelihood (probability). Thus, each connection between a life event node and an item node may have a likelihood that a user in whom a life event 50 has occurred will be interested in the item 51. The likelihood is represented by a value between 0 and 1 (1 being the maximum likelihood (probability)). The likelihood may be subject to restrictive conditions such as a time condition or an extended condition. Alternatively, or in addition, interest graph 121A may be a graph constructed, at least in part, based on marketing assumptions or heuristics, such that each connection between a life event node and an item node does not necessarily have a likelihood.

[0049] In the interest graph 121A, it is shown that a user who has experienced a life event of "getting married" (life event 501) will go on a trip within three months (or show an interest in traveling) with a likelihood of 0.9 (item node 511). Here, a time condition of "within three months" is imposed on the likelihood of the item node 511 indicating "travel". It is also shown that a user who has experienced the life event 501 will purchase (contract for) an item related to "mobile connection" (or show an interest in "mobile connection") (item node 512). It is also shown that a user who has experienced the life event 501 will subscribe to "insurance" (or show an interest in "insurance") with a likelihood of 1 (item node 513). Here, an extension condition of "including family" (i.e., adding family to an insurance already subscribed) is imposed on the likelihood of the item node 513 indicating "insurance". Furthermore, interest graph 121A indicates that a user for whom the life event "buy a car" (life event node 502) has occurred will go on a trip with a likelihood of 0.9 ("travel" 514).

[0050] The interest graph 121A shown in FIG. 5A is configured with connections between life events 50 and items of interest 51, but may also include user features. FIG. 5B shows an interest graph 121B as another example of the interest graph 121. The interest graph 121B is obtained by adding a relationship with a user feature to the interest graph 121A shown in FIG. 5A. Specifically, as shown in FIG. 5B, the interest graph 121B is a directed graph configured to include a directed connection to an item of interest (item node) 51, starting from a life event (life event node) 50 and a user feature 52 (user feature node). The connections added to the interest graph 121A will be described below.

[0051] The interest graph 121B may be a graph constructed at least in part using a Bayesian network that expresses the causal relationship between the user feature (cause) and the item of interest (result) by likelihood (probability). Thus, each connection between a user feature node and an item node may have a likelihood that a user having the user feature 40 is interested in the item 51. The likelihood is represented by a value between 0 and 1 (1 being the maximum likelihood (probability)). Alternatively, or in addition, the interest graph 121B may be a graph that is constructed, at least in part, based on marketing assumptions or heuristics, such that each connection between a user feature node and an item node does not necessarily have a likelihood.

[0052] For example, interest graph 121B indicates that a user (user feature node 521) whose user feature 52 includes "female" is interested in "travel" (item node 511, item node 514) with a likelihood of 1.0. Similarly, a user (user feature node 522) whose user feature 52 includes "hobby of travel" is also interested in "travel" (item node 511, item node 514). In addition, a user (user feature node 523) whose user feature 52 includes "age < 30 years old" is also interested in "insurance" (item node 513).

[0053] An interest graph 121 configured as an interest graph 121A shown in Fig. 5A or an interest graph 121B shown in Fig. 5B is stored in the graph storage unit 120. The interest graph management unit 103 can update the interest graph 121 every time new cause-effect data is acquired.

[0054] [Procedure for predicting user interests] Next, a prediction procedure by the interest item prediction unit 104 will be described. When the interest item prediction unit 104 detects that a life event has occurred in the user u1, it predicts and determines the interest of the user u1 (i.e., items in which the user u1 is interested) based on the interest graph 121 stored in the graph storage unit 120. The interest item prediction unit 104 can detect the occurrence of a life event in the user u1 based on the behavior history (search history, purchase history, etc.) included in the user characteristics of the user u1 in the web service.

[0055] For example, if the web service behavior history of user u1 acquired by the user characteristic acquisition unit 101 includes a purchase history indicating that user u1 has recently frequently purchased car-related products, the interest item prediction unit 104 can detect that a life event has occurred for user u1, that is, purchasing a car. Furthermore, for example, if the web service behavior history of user u1 acquired by the user characteristic acquisition unit 101 includes a recent registration history for a mommy service (childcare support service), the interest item prediction unit 104 can detect that a life event has occurred for user u1, namely, giving birth to a baby.

[0056] The interest item prediction unit 104 predicts the interest of the user u1 based on the interest graph 121 according to the detected life event (based on the detected life event). For example, when the life event is "getting married", the interest item prediction unit 104 can predict that the user u1 is interested in "travel", "mobile connection", and / or "insurance" based on the interest graph 121A shown in FIG. 5A. When the connection of each node in the interest graph 121 has a likelihood, the item in which the user u1 is interested can be predicted according to the likelihood, that is, in descending order of interest degree. In addition, the interest item prediction unit 104 may select one or more items based on the likelihood of the one or more items connected to the life event based on the likelihood shown in the interest graph 121A.

[0057] For example, based on the interest graph 121A, when a life event of "getting married" (life event 501) occurs to the user u1, the interest item prediction unit 104 can predict that the user u1 will be interested in all of "travel", "mobile connection", and "insurance". Alternatively, since the likelihood of "insurance" is high (=1) among items connected to the life event of "getting married" in the interest graph 121A, the interest item prediction unit 104 may select only "insurance". In other words, the interest item prediction unit 104 can predict that the user u1 will be interested in "insurance".

[0058] Furthermore, the interest item prediction unit 104 can predict the user's interest more accurately by using the user characteristics of the user u1 themselves in addition to the life events detectable based on the user characteristics of the user u1. For example, referring to FIG. 5B, when the user characteristics of the user u1 who has experienced the life event "getting married" include "female", there are more connections to the item node 511 indicating "travel" than to the item nodes 512 and 513 indicating "mobile connection" and "insurance". Therefore, the interest item prediction unit 104 can predict that the user u1 is more interested in "travel" among "travel", "mobile connection", and "insurance" based on the interest graph 121B shown in FIG. 5B. In this way, according to the number of connections to the item node, the interest item prediction unit 104 can predict the items that the user u1 is interested in. Furthermore, according to the likelihood to the item node or the sum of the likelihoods (when multiple likelihood connections are set from the life event node and / or the user feature node to the item node), the interest item prediction unit 104 may predict the items that the user u1 is interested in.

[0059] The interest item prediction unit 104 determines the one or more items in which the user u1 is interested, predicted by such a procedure, as the one or more items in which the user u1 is interested, and outputs (transmits) information on the determined one or more items to the recommendation unit 105. The interest item prediction unit 104 may be configured to output information on the determined one or more items to an external device.

[0060] [Overall process flow] Fig. 6 shows a flowchart of the entire process executed by the information processing device 10 according to this embodiment. For the explanation of Fig. 6, the information processing system shown in Fig. 1 will be referred to. The trained user feature prediction model 111 and desire stage prediction model 112 are stored in the trained model storage unit 110. The interest graph 121 is stored in the graph storage unit 120 and managed by the interest graph management unit 103.

[0061] In S61, the user feature acquisition unit 101 acquires user features (factual features) of a user u1 who is a target user. The user feature acquisition unit 101 may also acquire user features (estimated user features (attributes)) estimated for the user u1 using the user feature prediction model 111 based on the factual features.

[0062] In S62, the need stage prediction unit 102 uses the user characteristics acquired in S61 to predict the need stage to which the user u1 belongs among a predetermined number of need stages, using the need stage prediction model 112. As described above, in this embodiment, the five-stage need stage model defined by Maslow shown in Fig. 3A is used. Therefore, the need stage prediction unit 102 predicts the need stage to which the user u1 belongs among the five need stages shown in Fig. 3A, namely, physiological needs 31, safety needs 32, social needs 33, esteem needs 34, and self-actualization needs 35.

[0063] In S63, the interest item prediction unit 104 acquires the interest graph 121 corresponding to the desire stage to which the user u1 belongs, which was predicted in S62. For example, in S62, when the desire stage to which the user u1 belongs is predicted to be the social desire 33, the interest item prediction unit 104 acquires the interest graph 121 corresponding to the social desire 33 (for example, the interest graph A shown in FIG. 5A).

[0064] In S64, the interest item prediction unit 104 waits for the occurrence of a life event in the user u1. As described above, the interest item prediction unit 104 can detect the occurrence of a life event in the user u1 based on the behavior history (search history, purchase history, etc.) included in the user characteristics of the user u1 in the web service. When the life event occurs (Yes in S64), in S65, the interest item prediction unit 104 predicts and determines an item in which the user u1 is interested based on the interest graph 121 acquired in S63 according to the life event that has occurred. For example, the interest item prediction unit 104 may predict that the items in which the user u1, who has experienced the life event of "getting married" (life event 501), is interested are "travel", "mobile connection", and "insurance" based on the interest graph 121A shown in FIG. 5A. The interest item prediction unit 104 outputs information on one or more items in which the user u1 is interested that have been predicted (determined) to the recommendation unit 105.

[0065] In S66, the recommendation unit 105 performs control for recommending to the user u1 one or more items predicted in S65 that the user u1 is interested in. For example, the recommendation unit 105 recommends the one or more items to the user u1 by providing the user u1 with an advertisement related to the one or more items. The advertisement may be created by the recommendation unit 105, or may be selected by the recommendation unit 105 from one or more advertisements created in advance.

[0066] In this way, the information processing device according to this embodiment predicts the user's desires from a plurality of predetermined psychological desire stages using user characteristics including the user's behavior history. Furthermore, the information processing device determines one or more items in which the user is interested according to a life event that has occurred in the user based on the predicted desire stage, and recommends the item. This makes it possible to determine and recommend an appropriate item in line with the user's desires. This corresponds to determining and recommending an item in line with a desire based on a deep psychology that the user himself is not aware of. In other words, according to this embodiment, it is possible to determine and recommend an item in line with the user's fundamental desires.

[0067] The interest item prediction unit 104 may obtain the interest graph 121 constructed regardless of the desire stage to which the user u1 belongs, and may predict and determine items in which the user u1 is interested according to the detected life events. At this time, the recommendation unit 105 may determine an item according to the desire stage to which the user u1 belongs from the predicted items, and perform the recommendation.

[0068] [Hardware configuration of information processing device 10] FIG. 7 is a block diagram showing an example of a hardware configuration of the information processing device 10 according to the present embodiment. The information processing apparatus 10 according to the present embodiment can be implemented on any single or multiple computers, mobile devices, or any other processing platform. 7, the information processing device 10 is illustrated as being implemented in a single computer, but the information processing device 10 according to the present embodiment may be implemented in a computer system including multiple computers. The multiple computers may be connected to each other via a wired or wireless network so as to be able to communicate with each other.

[0069] 7, the information processing device 10 may include a CPU 71, a ROM 72, a RAM 73, a HDD 74, an input unit 75, a display unit 76, a communication I / F 77, and a system bus 78. The information processing device 10 may also include an external memory. A CPU (Central Processing Unit) 71 generally controls the operation of the information processing device 10, and controls each of the components (72 to 77) via a system bus 78, which is a data transmission path.

[0070] The ROM (Read Only Memory) 72 is a non-volatile memory that stores control programs and the like necessary for the CPU 71 to execute processing. Note that the programs may be stored in a non-volatile memory such as an HDD (Hard Disk Drive) 74 or an SSD (Solid State Drive) or in an external memory such as a removable storage medium (not shown). The RAM (Random Access Memory) 73 is a volatile memory and functions as a main memory, a work area, etc. of the CPU 71. That is, the CPU 71 loads necessary programs, etc. from the ROM 72 to the RAM 73 when executing processing, and realizes various functional operations by executing the programs, etc. The learning model storage unit 110 and the graph storage unit 120 shown in FIG. 2 can be configured by the RAM 73.

[0071] The HDD 74 stores, for example, various data and various information required when the CPU 71 performs processing using a program. The HDD 74 also stores, for example, various data and various information obtained when the CPU 71 performs processing using a program. The input unit 75 is composed of a keyboard and a pointing device such as a mouse. The display unit 76 is configured with a monitor such as a liquid crystal display (LCD), etc. The display unit 76 may be configured in combination with the input unit 75 to function as a GUI (Graphical User Interface).

[0072] The communication I / F 77 is an interface that controls communication between the information processing device 10 and an external device. The communication I / F 77 provides an interface with the network and executes communication with an external device via the network. Various data, various parameters, etc. are transmitted and received between the external device and the communication I / F 77. In this embodiment, the communication I / F 77 may execute communication via a wired LAN (Local Area Network) or a dedicated line conforming to a communication standard such as Ethernet (registered trademark). However, the network available in this embodiment is not limited to this, and may be configured as a wireless network. This wireless network includes wireless PANs (Personal Area Networks) such as Bluetooth (registered trademark), ZigBee (registered trademark), and UWB (Ultra Wide Band). It also includes wireless LANs (Local Area Networks) such as Wi-Fi (Wireless Fidelity) (registered trademark), and wireless MANs (Metropolitan Area Networks) such as WiMAX (registered trademark). It also includes wireless WANs (Wide Area Networks) such as LTE / 3G, 4G, and 5G. Note that the network only needs to connect each device so that they can communicate with each other, and the communication standard, scale, and configuration are not limited to those described above.

[0073] At least some of the functions of each element of the information processing device 10 shown in Fig. 2 can be realized by the CPU 71 executing a program. However, at least some of the functions of each element of the information processing device 10 shown in Fig. 2 may be operated as dedicated hardware. In this case, the dedicated hardware operates under the control of the CPU 71.

[0074] [Hardware configuration of user device 11] The hardware configuration of the user device 11 shown in Fig. 1 may be the same as that of Fig. 7. That is, the user device 11 may include a CPU 71, a ROM 72, a RAM 73, a HDD 74, an input unit 75, a display unit 76, a communication I / F 77, and a system bus 78. The user device 11 can display various information provided by the information processing device 10 on the display unit 76, and perform processing corresponding to an input operation received from the user u1 via a GUI (configured by the input unit 75 and the display unit 76).

[0075] Although specific embodiments have been described above, the embodiments are merely illustrative and are not intended to limit the scope of the present invention. The apparatus and method described in this specification can be embodied in forms other than those described above. Furthermore, the above-described embodiments can be appropriately omitted, substituted, and modified without departing from the scope of the present invention. Such omitted, substituted, and modified forms are included in the scope of the claims and their equivalents, and belong to the technical scope of the present invention.

[0076] The disclosure of this embodiment includes the following configuration. [1] An information processing device having: an acquisition means for acquiring user characteristics of a user; a prediction means for predicting, based on the user characteristics and using a learning model, which desire stage the user belongs to among a plurality of preset desire stages; a detection means for detecting a life event that has occurred to the user; and a determination means for determining one or more items in which the user is interested, based on the predicted desire stage and the detected life events.

[0077] [2] The information processing device according to [1], wherein the user characteristics include the user's behavioral history in a web service.

[0078] [3] An information processing device according to [1] or [2], wherein the one or more items in which the user is interested include one or more of one or more products and one or more services.

[0079] [4] The information processing device described in any of [1] to [3], wherein the multiple desire stages are composed of hierarchical stages representing human psychological desires.

[0080] [5] The information processing device according to any one of [1] to [4], wherein the detection means detects the user's life events based on the user's behavioral history in a web service.

[0081] [6] An information processing device described in any of [1] to [5], wherein the determination means determines the one or more items in which the user is interested based on a directed graph from each of a plurality of life events to one or more items defined for each of the plurality of desire stages.

[0082] [7] An information processing device according to any one of [1] to [6], further comprising a recommendation means for recommending to the user the one or more items in which the user is interested.

[0083] [8] The information processing device according to any one of [7], wherein the recommendation means provides the user with advertisements relating to the one or more items in which the user is interested. [Explanation of symbols]

[0084] u1: user, 10: information processing device, 11: user device, 101: user feature acquisition unit, 102: desire stage prediction unit, 103: interest graph management unit, 104: interest item prediction unit, 105: recommendation unit, 110: learning model storage unit, 111: user feature prediction model, 112: desire stage prediction model, 120: graph storage unit, 121: interest graph

Claims

1. An acquisition means for acquiring user characteristics of a user including a behavior history of the user in a web service; A prediction means for predicting a desire stage to which the user belongs among the plurality of desire stages by using a learning model trained to output a probability that the user belongs to a plurality of predetermined desire stages using the user characteristics as input; A detection means for detecting a life event occurring to the user; determining means for determining one or more items in which the user is interested based on the predicted desire stage and the detected life events; having The information processing device is characterized in that the determination means determines one or more items connected to the detected life event in a graph connected from each of a plurality of life events to one or more items defined for each of the plurality of desire stages, as the one or more items in which the user is interested.

2. the one or more items in which the user is interested include one or more of one or more products and one or more services; 2. The information processing apparatus according to claim 1,

3. 2. The information processing device according to claim 1, wherein the plurality of need stages are configured as hierarchical stages representing human psychological needs.

4. The detection means detects a life event of the user based on a behavior history of the user in a web service.

2. The information processing apparatus according to claim 1,

5. The system further comprises a recommendation means for recommending to the user the one or more items in which the user is interested.

2. The information processing apparatus according to claim 1,

6. the recommendation means providing advertisements to the user regarding the one or more items in which the user is interested; 6. The information processing apparatus according to claim 5,

7. An information processing method executed by an information processing device, An acquisition step of acquiring user characteristics of the user including a behavior history of the user in a web service; a prediction step of predicting a desire stage to which the user belongs among the plurality of desire stages by using a learning model trained to output a probability corresponding to a plurality of predetermined desire stages using the user characteristics as input; A detection step of detecting a life event that has occurred to the user; determining one or more items in which the user is interested based on the predicted desire stage and the detected life events; having An information processing method characterized in that in the determination step, one or more items connected to the detected life event in a graph connected from each of a plurality of life events to one or more items defined for each of the plurality of desire stages are determined to be the one or more items in which the user is interested.

8. An information processing program for causing a computer to execute information processing, the information processing program comprising: An acquisition process of acquiring user characteristics of the user including a behavior history of the user in a web service; A prediction process for predicting a desire stage to which the user belongs among the plurality of desire stages by using a learning model trained to output a probability corresponding to a plurality of predetermined desire stages using the user features as input; A detection process for detecting a life event occurring to the user; determining one or more items in which the user is interested based on the predicted desire stage and the detected life events; The determination process includes determining, as the one or more items in which the user is interested, one or more items connected to the detected life event in a graph connected to one or more items from each of a plurality of life events defined for each of the plurality of desire stages. Information processing program.

Images