A system, method, and program for predicting a user's remaining lifespan.

Abstract

To provide a system that predicts a user's remaining lifespan using a new approach. [Solution] The present invention provides a system for predicting a user's remaining lifespan, the system comprising: receiving means for receiving data relating to a user, the data including data relating to the user's personality and data relating to the user's health; derivation means for deriving the probability that the user is willing to use an AI agent that provides health guidance, based on the data relating to the personality; and prediction means for predicting the user's remaining lifespan, based on the health data and the probability.

Description

Technical Field

[0006] , , , ,

[0001] The present invention relates to a system, method, and program for predicting a user's life expectancy.

Background Art

[0002] Many people are concerned about their health and are also interested in their life expectancy, that is, how much longer they can live. For example, it is known to generate prediction data of healthy life expectancy from the results of a personal health checkup (Patent Document 1).

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] An object of the present invention is to provide a system or the like for predicting a user's life expectancy with a new approach.

Means for Solving the Problems

[0005] In the present invention, the life expectancy of a user is predicted using data related to the user's personality. Specifically, in the present invention, the life expectancy of a user is predicted based on data related to health and data related to personality.

[0006] The present invention provides, for example, the following. (Item 1) A system for predicting a user's life expectancy, the system comprising: Receiving means for receiving data related to a user, the data including data related to the user's personality and data related to the user's health, the receiving means; A derivation means for deriving the probability that the user is willing to use an AI agent to provide health guidance, based on the aforementioned personality data, A prediction means for predicting the user's remaining lifespan based on the aforementioned health data and the aforementioned probability. A system equipped with these features. (Item 2) The derivation means is, Based on the aforementioned personality data and / or health data, determine a type of AI agent specific to the user. The system described above, further configured to perform the derivation, which includes deriving the probability that the user is willing to use the determined AI agent. (Item 3) The prediction means is a system according to any one of the above items, which predicts the remaining lifespan based on the health data, the probability, and the characteristics of the determined AI agent. (Item 4) The system described in any one of the above items includes, for example, data relating to the personality, at least one of the user's behavioral history, the user's SNS usage history, the user's purchase history, and a personality assessment. (Item 5) The prediction means is a system according to any one of the above items, which predicts the user's remaining lifespan using a machine learning model that has learned the relationship between data on the target's health, the probability that the target is willing to use an AI agent to provide health guidance, and the target's remaining lifespan. (Item 6) The receiving means further receives data representing the user's actual lifespan and data relating to the user's use of the AI ​​agent. The system further comprises adjustment means for adjusting the machine learning model using the data representing the actual lifespan and the data relating to the usage, as described in any one of the above items. (Item 7) A method for predicting a user's remaining lifespan, wherein the method is: Receiving data relating to a user, wherein the data includes data relating to the user's personality and data relating to the user's health. Based on the aforementioned personality data, the probability that the user would be willing to use an AI agent to provide health guidance is derived. To predict the user's remaining lifespan based on the aforementioned health data and the aforementioned probability. A method that includes this. (Item 7A) The method described in item 7, which includes the characteristics described in any one of the above items. (Item 8) A program that predicts a user's remaining lifespan, wherein the program is executed in a system equipped with a processor, and the program is Receiving data relating to a user, wherein the data includes data relating to the user's personality and data relating to the user's health. Based on the aforementioned personality data, the probability that the user would be willing to use an AI agent to provide health guidance is derived. To predict the user's remaining lifespan based on the aforementioned health data and the aforementioned probability. A program that causes the processor to perform the processing including the specified operations. (Item 8A) A program as described in item 8, which includes the characteristics described in any one of the above items. (Item 8B) A non-transient, computer-readable storage medium for storing the programs described in item 8 or item 8A. [Effects of the Invention]

[0007] According to the present invention, the remaining life of a user can be predicted by a new approach, and the accuracy of remaining life prediction can be improved. Further, the predicted remaining life reflects the personality of the user, and the predicted remaining life can be used for various personalization purposes. For example, the predicted remaining life can be used for tuning an AI agent.

Brief Description of Drawings

[0008] [Figure 1] A diagram showing an example of a flow for realizing a new method for predicting the remaining life of a user [Figure 2] A diagram showing an example of the configuration of a system 100 for predicting the remaining life of a user [Figure 3] A diagram showing an example of the specific configuration of a system 100 for predicting the remaining life of a user [Figure 4A] A diagram showing an example of the configuration of the processor unit 120 [Figure 4B] A diagram showing an example of another configuration of the processor unit 120 [Figure 5] A flowchart showing an example of a process 500 in a system 100 for predicting the remaining life of a user

Modes for Carrying Out the Invention

[0009] (Definition) In this specification, "personality data" refers to data that directly or indirectly represents a person's character and / or personality. Personality data includes subjective data obtained from information subjectively entered by the person, and objective data obtained objectively from the person. Subjective data includes, for example, psychological test data and personality assessment data, and can be obtained through psychological tests, personality assessments, questionnaires, etc. Objective data includes, for example, behavioral history (especially smart device usage history), purchase history (especially online shopping purchase history), SNS usage history, vital information, etc., and can be obtained from mobile devices such as smartphones, IoT devices, networks, etc. More specifically, personality data may include the following data: Data based on the Big Five personality theory. The Big Five theory is a model that evaluates five traits: extraversion, agreeableness, conscientiousness, emotional intensity, and openness, and can be measured using self-report questionnaires or psychological tests. Data based on the MBTI (Myers-Briggs Type Indicator). The MBTI is a tool that classifies individuals' personalities into 16 types, which are comprised of four dimensions: extraversion and introversion, sensing and intuition, thinking and feeling, and judging and perceiving. Data obtained through text analysis. By analyzing text data such as social media posts and essays, personality traits can be estimated, and tools such as IBM's Personality Insights can be used for text analysis. Data based on an Egogram or Digram assessment. Egogram and Digram assessments are psychological tests that measure ego states and analyze an individual's emotions and behavioral patterns. Data obtained through self-report questionnaires. Self-report questionnaires are a type of questionnaire in which respondents answer questions about their own feelings and behaviors, and can measure psychological characteristics and stress tolerance. Data obtained through behavioral observation. Through behavioral observation by others, characteristics related to a person's personality can be understood. Physiological data, including heart rate and stress hormone levels, can be measured by wearable devices, IoT devices, etc.

[0010] In this specification, "life expectancy" includes not only life expectancy in the narrow sense, which is the expected number of years remaining until death, but also the concept of healthy life expectancy, which is the expected number of years one can live a normal life without health problems.

[0011] In this specification, “AI agent” means an autonomously operating artificial intelligence (AI). An AI agent is designed to act as a substitute (i.e., an agent) for a human. In particular, the AI ​​agent covered by the present invention is designed to provide at least health guidance to a user. For example, one AI agent may be designed to provide health guidance to a user with strict guidance (e.g., by interfering with the user's behavior), and another AI agent may be designed to provide health guidance to a user with lenient guidance (e.g., by letting the user's behavior be left to their own devices). Health guidance by an AI agent includes, but is not limited to, guidance on the content and quantity of food, guidance on the content and timing of exercise, guidance on the amount and timing of sleep, guidance on the timing of sleep, guidance on the timing of rest, etc.

[0012] Embodiments of the present invention will be described below with reference to the drawings.

[0013] 1. Predicting the user's remaining lifespan The inventors of this invention have developed a new method for predicting a user's life expectancy. This new method predicts a user's life expectancy by using data related to the user's personality in addition to health-related data such as health checkup results, which have been used in the past. As a result of diligent research, the inventors of this invention have found that data related to the user's personality, which may not seem related to life expectancy, can be used to predict life expectancy.

[0014] The inventors of this invention believe that a person's character, particularly their personality or temperament, contributes at least partially to whether they can maintain their health and, consequently, extend their lifespan. For example, an introverted person who frequently uses online shopping may have a longer lifespan. Similarly, an extroverted person who frequently uses social media may have a longer lifespan. For example, highly responsible individuals tend to lead regular lives, pay attention to their health, and maintain healthy lifestyle habits, thus having a lower risk of cardiovascular disease and chronic illness. For example, optimistic individuals are better able to manage stress and strengthen their immune systems, leading to a lower risk of disease and often a longer lifespan. Optimists are also thought to have a greater ability to cope positively with difficult situations and maintain their health. For example, sociable individuals can build good relationships, which can contribute to stress reduction and the maintenance of physical and mental health, resulting in less loneliness, better psychological health, and a longer lifespan. For example, individuals with high stress tolerance are more likely to maintain physical and mental health and have a lower risk of chronic illness, which may have a positive impact on their lifespan. For example, individuals with high self-efficacy tend to proactively cope with difficult situations and take actions to maintain their health, which is thought to contribute to a longer life.

[0015] Furthermore, the inventors of this invention considered that whether or not a person uses an AI agent that provides health guidance may also affect whether or not they can maintain their health and, consequently, extend their life expectancy. For example, a person with a certain personality may have an extended life expectancy by using an AI agent that provides health guidance using a method suited to their personality. For example, a person with a certain personality may have an extended life expectancy by using an AI agent that provides health guidance using a method unsuitable for their personality. For example, a person with a certain personality may have an extended life expectancy by not using an AI agent that provides health guidance.

[0016] A new method for predicting a user's remaining lifespan uses data on the user's personality to estimate whether or not the user is someone who would use an AI agent to provide health guidance. This estimation, along with data on the user's health, is then used to predict the user's remaining lifespan.

[0017] Figure 1 shows an example of a workflow for implementing a new method for predicting a user's remaining lifespan. In this example, system 100 will predict the user's remaining lifespan.

[0018] In step S1, data related to user U's health is provided to system 100. For example, the results of a health checkup conducted by user U at hospital H are provided to system 100. Although Figure 1 shows that data related to user U's health is input to system 100 from hospital H, the system is not limited to this, and user U may also provide data related to user U's health to system 100 via a terminal device or the like.

[0019] Health data typically includes, but is not limited to, data obtained during health checkups, such as physical measurements like height, weight, and BMI; non-invasive measurements like blood pressure and heart rate; and invasive measurements like blood glucose levels, cholesterol levels, and liver function values. For example, health data also includes information obtained through interviews (e.g., current illness, past medical history, and medication history).

[0020] In step S2, data relating to user U's personality is provided to system 100. For example, data relating to user U's personality stored or distributed on network cloud C is provided to system 100. Although Figure 1 shows that data relating to user U's personality is input to system 100 from network cloud C, the system is not limited to this, and user U may also provide data relating to user U's personality to system 100 via a terminal device or the like.

[0021] System 100 predicts User U's remaining lifespan based on the provided data. Specifically, System 100 derives the probability that User U will accept health guidance from AI Agent A based on data about the user's personality, and then predicts the remaining lifespan based on that probability and health data. For example, if User U is likely to accept health guidance from AI Agent A, the probability will be close to 1. On the other hand, if User U is likely not to accept health guidance from AI Agent A, the probability will be close to 0. For example, a user who likes new things might be happy to use an AI agent that provides health guidance, while a confident user might be reluctant to use an AI agent that provides health guidance. Based on such user tendencies and statistics, System 100 derives the probability that User U will accept health guidance from AI Agent A, that is, the probability that User U will not be reluctant to use an AI agent that provides health guidance.

[0022] Compared to predicting life expectancy based solely on user U's health data, predicting life expectancy based on the probability that user U is willing to use an AI agent to provide health guidance is expected to improve prediction accuracy. This is because even the user's personality or character can be used in the prediction.

[0023] System 100, for example, includes a machine learning model that can be used to predict a user's remaining lifespan. The machine learning model may be a dedicated model specifically designed for predicting user U's remaining lifespan, or it may be a general-purpose model that can be used to predict the remaining lifespan of any user. Preferably, the machine learning model may be a dedicated model that has been trained or transferred-learned using user U's data.

[0024] System 100 may determine which AI agent user U should use and derive the probability that user U will accept health guidance from the determined specific AI agent.

[0025] In step S3, information including the predicted life expectancy is provided to user U. User U can view the information including the predicted life expectancy via their terminal device. The information including the predicted life expectancy may be, for example, direct information indicating how many years user U has left to live, or it may be indirect information derived based on the predicted life expectancy. The indirect information may be, for example, information indicating products or services (e.g., financial products, especially insurance products) that can be determined based on the predicted life expectancy. The indirect information may be, for example, medical information (e.g., treatment plan) that can be determined based on the predicted life expectancy, or health advice that can be determined based on the predicted life expectancy. In step S3, in addition to or instead of the information, a product designed based on the predicted life expectancy may be provided to user U.

[0026] In step S3, user U may also be provided with information indicating which AI agent user U should use. For example, user U may refer to the provided information and information including predicted life expectancy to decide whether to receive health guidance using the AI ​​agent determined by system 100, to receive health guidance using an AI agent other than the one determined by system 100, or not to receive health guidance from an AI agent at all.

[0027] For example, in step S4, user U receives health guidance using AI agent A, which has been determined to be the AI ​​agent to be used. AI agent A can advise user U on things like what kind of meals to eat and when, how much exercise to do and when, and how much sleep to get. Whether or not user U followed the advice can be recorded, for example, via the user's wearable device or smartphone. The recorded information can be stored, for example, on the user's terminal device.

[0028] In step S5, the system 100 may be provided with a record of whether user U used AI agent A or followed the advice given by AI agent A. For example, a record stored on the user's terminal device may be provided to the system 100.

[0029] System 100 may update its machine learning model using records of whether user U used AI agent A or followed advice from AI agent A. For example, the machine learning model can be retrained using the life expectancy predicted from only the user's health data after a predetermined period, and records of whether user U used AI agent A or followed advice from AI agent A during that predetermined period. Alternatively, for example, after the user dies of old age or illness, the machine learning model can be retrained using the actual lifespan and records of whether user U used AI agent A or followed advice from AI agent A.

[0030] As many users utilize System 100 and a large amount of data is accumulated, this data can be used for retraining, thereby improving the accuracy of the machine learning model.

[0031] The system 100 described above can be implemented by a system for predicting the user's remaining lifespan, which will be described later.

[0032] 2. System configuration for predicting the user's remaining lifespan Figure 2 shows an example of the configuration of system 100 for predicting a user's remaining lifespan.

[0033] System 100 is connected to the database unit 200. System 100 is also connected to at least one user terminal device 300 via network N. Furthermore, system 100 is connected to at least one server device 400 via network N.

[0034] Although Figure 2 shows three user terminal devices 300, the number of user terminal devices 300 is not limited to these. Any number of user terminal devices 300 can be connected to the system 100 via the network N.

[0035] Furthermore, although two server devices 400 are shown in Figure 2, the number of server devices 400 is not limited to these. Any number of server devices 400 can be connected to system 100 via network N.

[0036] Here, the server device 400 is a device capable of handling user data. The server device 400 has its own database, and these databases store user data. For example, one of the server devices 400 may be a device capable of handling user health data, and could be a server device installed in, for example, a medical institution. For example, one of the server devices 400 may be a device capable of handling user personality data, and could be, for example, an SNS management server.

[0037] Network N can be any type of network. Network N may be, for example, the Internet or a LAN. Network N may be a wired network or a wireless network.

[0038] One example of system 100 is a computer (e.g., a server) installed in a provider that provides a service to predict a user's remaining lifespan, but it is not limited to this. System 100 may also be a computer (e.g., a server) installed in a provider that provides a service to propose products or services to users, in which case system 100 can propose products or services that are suitable for the user based on the predicted remaining lifespan. More specifically, system 100 may be a computer installed in a provider that provides a service to propose financial products to users, in which case system 100 can provide financial products that are suitable for the user based on the predicted remaining lifespan. By proposing financial products based on the predicted remaining lifespan, the proposed financial products may take into account the risks related to the user's remaining lifespan.

[0039] An example of a user terminal device 300 is a computer (e.g., a terminal device) used by a user who is a consumer of a product or service, but is not limited to this. Here, the computer (server device or terminal device) can be any type of computer. For example, the terminal device can be any type of terminal device such as a smartphone, tablet, personal computer, smart glasses, or smartwatch.

[0040] The database unit 200 stores various types of information used to build machine learning models that can be used in system 100.

[0041] Figure 3 shows an example of a specific configuration of system 100 for predicting a user's remaining lifespan.

[0042] System 100 comprises an interface unit 110, a processor unit 120, and a memory unit 130.

[0043] The interface unit 110 exchanges information with the outside of the system 100. The processor unit 120 of the system 100 can receive information from the outside of the system 100 via the interface unit 110 and can transmit information to the outside of the system 100. The interface unit 110 can exchange information in any format.

[0044] The interface unit 110 includes, for example, an input unit that enables information to be input to the system 100. The manner in which the input unit enables information to be input to the system 100 is not limited. For example, if the input unit is a receiver, the receiver may input information by receiving it from outside the system 100 via a network. Alternatively, if the input unit is a data reading device, it may input information by reading it from a storage medium connected to the system 100.

[0045] The interface unit 110 includes, for example, an output unit that enables information to be output from the system 100. The mode in which the output unit enables information to be output from the system 100 is not limited. For example, if the output unit is a transmitter, the information may be output by the transmitter transmitting information to the outside of the system 100 via a network. Alternatively, if the output unit is a data writing device, the information may be output by writing the information to a storage medium connected to the system 100.

[0046] System 100 can, for example, transmit information to and / or receive information from the database unit 200 via the interface unit 110. System 100 can, for example, transmit information to and / or receive information from the user terminal device 300 via the interface unit 110. System 100 can, for example, transmit information to and / or receive information from the server device 400 via the interface unit 110.

[0047] System 100 can receive user data, for example, via the interface unit 110. User data includes data on the user's health and data on the user's personality. User data may also include data representing the user's actual lifespan and data on the user's use of the AI ​​agent. System 100 can transmit information, for example, via the interface unit 110, including the predicted lifespan of the user. System 100 can transmit a modified trained model to the database unit 200, for example, via the interface unit 110.

[0048] The processor unit 120 executes the processing of the system 100 and controls the operation of the entire system 100. The processor unit 120 reads the program stored in the memory unit 130 and executes the program. This makes it possible to make the system 100 function as a system that executes desired steps. The processor unit 120 may be implemented by a single processor or by multiple processors.

[0049] The memory unit 130 stores programs necessary for executing the processes of the system 100, as well as data necessary for executing those programs. The memory unit 130 may also store a program (for example, a program that implements the process shown in Figure 5, described later) that causes the processor unit 120 to perform processing to predict the user's remaining lifespan. Here, it is not specified how the program is stored in the memory unit 130. For example, the program may be pre-installed in the memory unit 130. Alternatively, the program may be stored in a non-transient computer-readable storage medium and installed by reading the storage medium. Alternatively, the program may be installed in the memory unit 130 by being downloaded via a network. In this case, the type of network is not specified. The memory unit 130 can be implemented using any storage means.

[0050] The database unit 200 stores various types of information used to build machine learning models that can be used in system 100. For example, the database unit 200 stores data related to personality, health, and life expectancy obtained from multiple users in the past, all linked together.

[0051] In the examples shown in Figures 2 and 3, the database unit 200 is located outside the system 100, but the present invention is not limited thereto. It is also possible to provide at least a portion of the database unit 200 inside the system 100. In this case, at least a portion of the database unit 200 may be implemented by the same storage means as the storage means implementing the memory unit 130, or by storage means different from the storage means implementing the memory unit 130. In any case, at least a portion of the database unit 200 is configured as a storage unit for the system 100. The configuration of the database unit 200 is not limited to a specific hardware configuration. For example, the database unit 200 may be composed of a single hardware component or multiple hardware components. For example, the database unit 200 may be configured as an external hard disk drive for the system 100, as cloud storage connected via a network, or as a distributed network utilizing blockchain technology or the like.

[0052] For example, data relating to the user's health and / or personality is stored in a database unit 200 configured as a decentralized network utilizing blockchain technology, and in this case, the data relating to the user's health and / or personality becomes virtually impossible to tamper with.

[0053] Figure 4A shows an example of the configuration of the processor unit 120.

[0054] The processor unit 120 includes a receiving means 121, a derivation means 122, and a prediction means 123.

[0055] The receiving means 121 is configured to receive data about the user. The data about the user includes data about the user's health and data about the user's personality.

[0056] User health data typically includes, but is not limited to, data obtained during health checkups, such as physical measurements like height, weight, and BMI; non-invasive measurements like blood pressure and heart rate; and invasive measurements like blood glucose levels, cholesterol levels, and liver function values. For example, health data also includes information obtained through interviews (e.g., current medical history, past medical history, and medication history). Health data may be received from medical institutions such as hospitals, or from the user's terminal device.

[0057] Data relating to a user's personality may include, for example, behavioral history (especially smart device usage history), purchase history (especially online shopping purchase history), social media usage history, and personality assessments. Data relating to a user's personality may be received, for example, from a network cloud or from the user's terminal device.

[0058] The data received by the receiving means 121 is passed to the derivation means 122.

[0059] The derivation means 122 derives the probability that a user is willing to use an AI agent that provides health guidance, based on data about the user's personality. The derivation means 122 may, for example, derive the probability using a rule-based method or by using a machine learning model.

[0060] In one example, the derivation means 122 can use a machine learning model to derive the probability that a user is willing to use an AI agent that provides health guidance. The machine learning model used learns the relationship between data on the personalities of multiple users and data indicating whether each of the multiple users willingly uses an AI agent for health guidance. Specifically, the set of training data for training the machine learning model (input training data, output training data) may be (data on the personality of the first user, whether the first user willingly uses an AI agent for health guidance), (data on the personality of the second user, whether the second user willingly uses an AI agent for health guidance), (data on the personality of the third user, whether the third user willingly uses an AI agent for health guidance), etc. The data indicating whether a user willingly uses an AI agent for health guidance may be binary ("1" indicating use, "0" indicating not use) or multi-valued (for example, a multi-level value indicating the degree to which they are willing to use it).

[0061] When data about a user's personality is input into a machine learning model that has undergone such training, the probability that the user would be willing to use an AI agent to provide health guidance is output. Whether or not a user is willing to use an AI agent to provide health guidance largely depends on the user's personality and / or character, and this probability can be estimated with high accuracy by using data about the user's personality.

[0062] Here, the AI ​​agent for health guidance may be a general-purpose AI agent, or it may be a specific AI agent from among several types of AI agents.

[0063] For example, the derivation means 122 may determine a type of AI agent specific to the user (e.g., an AI agent that is compatible with the user's personality, an AI agent that is suited to the user's health condition, etc.) and derive the probability that the user is willing to use the determined AI agent. The derivation means 122 can determine a type of AI agent specific to the user based on, for example, data on personality and / or data on health. For example, the derivation means 122 can determine the user's personality from the personality data and / or their health condition from the health data, and determine an AI agent that is compatible with that personality and / or health condition, for example, using a rule-based approach or machine learning. For example, an AI agent that provides frequent praise and health guidance may be determined for a user who thrives on praise, and an AI agent that provides strict health guidance may be determined for a user who thrives in challenging environments. For example, an AI agent that provides little intervention may be determined for a user with good health, and an AI agent that provides a lot of intervention may be determined for a user with poor health. For example, the database unit 200 may store which AI agent is suitable for each of multiple personalities and / or health states, and the derivation means 122 can refer to this information to determine the user's specific type of AI agent.

[0064] In another example, the derivation means 122 may derive a score representing the user's characteristics from personality data, and based on that score, derive the probability that the user would be willing to use an AI agent that provides health guidance. For example, instead of the personality data described above, the derivation means 122 could derive the probability that the user would be willing to use an AI agent that provides health guidance based on a score representing the user's characteristics.

[0065] The score representing user U's characteristics can represent a "digital twin." A "digital twin" is a virtual or digital replica of an object in real space. In other words, a user's "digital twin" reflects what kind of person the user is, or their personality, in virtual or digital space. That is, the score representing user U's characteristics can represent what kind of person the user is, or their personality. In particular, it represents what kind of person the user is from a personality perspective.

[0066] In a score that represents a user's characteristics, for example, points are assigned to multiple items, and the person's personality is represented based on the scores of each of these items. For example, a person who is quick-tempered but honest may have a high score on the "honesty" item or a low score on the "liar" item, and a high score on the "short temper" item or a low score on the "patience" item among the items related to "personality."

[0067] For example, the database unit 200 may store concepts related to the parameters used to calculate the score, along with various other information.

[0068] For example, regarding "personality," the database unit 200 may store concepts related to personality in association with various types of information. For instance, the concept of "altruistic" (or "transpersonal") in relation to "personality" may be associated with keywords such as "volunteering," "donating," "caring," and "consulting," as well as statuses such as "frequently participates in volunteer work," "frequently seeks consultation," and "has experience donating." For example, the concept of "SDGs" (Sustainable Development Goals) in relation to "personality" may be associated with keywords related to the 17 goals and / or 169 targets (e.g., "equality," "environmental protection," etc.), as well as statuses such as "takes action related to the 17 goals and / or 169 targets" and "holds beliefs related to the 17 goals and / or 169 targets."

[0069] Such associations can be made, for example, using artificial intelligence (AI) capable of performing semantic searches, that is, AI that has learned the correlations between keywords.

[0070] This artificial intelligence learns the correlations between keywords from a large amount of text. For example, by syntactically analyzing a text, this AI extracts multiple keywords within the text and identifies the relationships between those keywords. For instance, if a certain keyword is used in conjunction with another keyword in many texts, this AI learns those keywords as having a strong correlation. Having learned the correlations between keywords in this way, the AI ​​can output keywords that correlate with the input keyword based on the learned correlations.

[0071] The derivation means 122 can, for example, determine the degree to which personality data correlates with concepts related to "personality," and calculate a score according to the determined degree of correlation. For example, if personality data correlates strongly with a particular concept among the concepts related to "personality," the derivation means 122 can include the features corresponding to that concept in the score or increase the score of the items corresponding to that concept. For example, the more strongly personality data correlates with "altruistic" (or "transpersonal"), the more or more strongly the features corresponding to "altruistic" can be included in the score, or the higher the score of the items corresponding to "altruistic." For example, the more strongly personality data correlates with "SDGs" (or "Sustainable Development Goals"), the more or more strongly the features corresponding to "SDGs" can be included in the score, or the higher the score of the items corresponding to "SDGs."

[0072] The derivation means 122 can, for example, calculate a score representing the user's characteristics using a machine learning model that has learned the relationship between personality data and scores. The machine learning model can be constructed using any machine learning model. The machine learning model may be, for example, a neural network model. The machine learning model may also be, for example, an LLM.

[0073] Machine learning models can be pre-trained using personality data from multiple users. This training process might involve, for example, using pre-acquired personality data to calculate the weight coefficients for each node in the hidden layer of a neural network model.

[0074] The learning process is, for example, supervised learning. Supervised learning is performed using information from multiple users, for example, by using personality data as input training data and scores representing the characteristics of those users as output training data. This makes it possible to build a machine learning model that can correlate personality data with scores.

[0075] By deriving a score, it is possible to reduce the dimensionality while maintaining the trends in personality-related data, thereby reducing the load on subsequent processing.

[0076] The derived probability is passed to the prediction means 123.

[0077] The prediction means 123 is configured to predict the user's remaining lifespan based on data related to the user's health and the probability that the user is willing to use an AI agent to provide health guidance. The derivation means 122 can predict the user's remaining lifespan using, for example, a machine learning model.

[0078] The machine learning model used learns the relationship between data on the health of multiple users, data indicating whether each user willingly uses an AI agent for health guidance, and the life expectancy of each user. Specifically, the training data set for the machine learning model ([input training data], output training data) could be ([health data of the first user, probability that the first user willingly use the AI ​​agent for health guidance], life expectancy of the first user), ([health data of the second user, probability that the second user willingly use the AI ​​agent for health guidance], life expectancy of the second user), ([health data of the third user, probability that the third user willingly use the AI ​​agent for health guidance], life expectancy of the third user), etc. When the user's health data and the probability that the user willingly uses the AI ​​agent for health guidance are input to a machine learning model that has undergone such training, the user's life expectancy is output. A user's life expectancy can be influenced by whether or not they use an AI agent for health guidance (for example, using an AI agent suited to the user may extend their life expectancy, while using an AI agent that is not suited to them may shorten it). Therefore, by utilizing the probability that a user is willing to use an AI agent for health guidance, it is possible to accurately estimate the user's life expectancy.

[0079] In one example, the prediction means 123 can predict the user's remaining lifespan based on data related to the user's health, the probability that the user is willing to use an AI agent to provide health guidance, and the characteristics of the AI ​​agent determined by the derivation means 122.

[0080] In this case, the machine learning model used learns the relationship between data on the health of multiple users, data indicating whether each of the multiple users willingly uses the AI ​​agent determined for them, the characteristics of the AI ​​agent determined for each of the multiple users, and the life expectancy of each of the multiple users. Specifically, the training data set for learning the machine learning model ([input training data], output training data) could be ([data on the health of the first user, the probability that the first user willingly use the AI ​​agent determined for the first user, the characteristics of the AI ​​agent determined for the first user], the life expectancy of the first user), ([data on the health of the second user, the probability that the second user willingly use the AI ​​agent determined for the second user, the characteristics of the AI ​​agent determined for the second user], the life expectancy of the second user), ([data on the health of the third user, the probability that the third user willingly use the AI ​​agent determined for the first user, the characteristics of the AI ​​agent determined for the third user], the life expectancy of the third user)... When data on the user's health, the probability that the user would be willing to use a user-specific AI agent, and the characteristics of the user-specific AI agent are input into a machine learning model that has undergone such training, the user's remaining lifespan is output.

[0081] The predicted life expectancy is output to the outside of system 100 and can be used for any purpose. For example, the predicted life expectancy is provided to the user's terminal device to inform the user of their life expectancy. For example, it can be provided to a provider of goods or services and used to determine a product or service that suits the user's life expectancy. Specifically, as an example, the predicted life expectancy is provided to a financial instruments business operator that provides financial products, and the financial instruments business operator can propose financial products to the user, taking into account the user's life expectancy and its risks. For example, risk-segmented financial products (e.g., lifetime annuities or health insurance) can be offered, in which case the premium may be set based on the predicted life expectancy. This makes it possible to take measures against longevity risk. For example, the predicted life expectancy can be used for marketing and product development, and a target market can be identified based on the predicted life expectancy, making it possible to provide products that suit the consumer's lifestyle. For example, a product may be designed based on the predicted life expectancy and then provided to the user.

[0082] As another example, predicted life expectancy can be provided to medical institutions such as hospitals. Based on this prediction, physicians can determine treatment plans, as well as strategies for palliative care and end-of-life care. Predicted life expectancy can also contribute to the efficient allocation of medical resources and help provide appropriate care to the right patients.

[0083] As yet another example, the predicted life expectancy is provided to healthcare service providers, who then provide personalized services tailored to that predicted life expectancy. For instance, they might offer a health management app or wearable device that, based on the predicted life expectancy, suggests ways to improve health or provides specific advice tailored to the individual's health condition.

[0084] As yet another example, the predicted life expectancy could be provided to the company's management department, which could then develop health management strategies based on that prediction. The company's management department could, for example, score and evaluate the health status or life expectancy of its employees.

[0085] As yet another example, predicted life expectancy can be used to tune an AI agent. Since predicted life expectancy reflects the user's personality, this can lead to the personalization of the AI ​​agent. To ensure personalization, predicted life expectancy may be used in conjunction with data about the user's personality, for example. For example, a personalized AI agent could provide the following: Personalized experiences. For example, an AI agent can provide a customized experience based on the user's preferences and behavior, which can improve user satisfaction and increase conversion rates. Marketing efficiency can be improved. For example, AI agents can implement targeted marketing strategies, potentially leading to a better return on advertising spending. Improving customer engagement. For example, AI agents can deepen relationships with customers and improve brand loyalty. Uncovering latent needs. For example, AI agents can identify potential needs and contribute to the development of new products and improvements to services. Applications in the medical field include, for example, AI agents that can help with patient health management and treatment planning. AI agents can also contribute to personalized medicine, potentially improving patient satisfaction and treatment effectiveness. Contribution to preventive medicine. For example, AI agents can provide early intervention and educational programs for specific health risks.

[0086] Figure 4B shows another example of the configuration of the processor unit 120. The processor unit 120 shown in Figure 4B is identical to the processor unit 120 shown in Figure 4A, except that it further comprises adjustment means 124. Therefore, the explanation given above with reference to Figure 4A also applies to Figure 4B, and will not be repeated here.

[0087] The user data received by the receiving means 121 includes data representing the user's actual lifespan and data regarding the user's use of the AI ​​agent.

[0088] The data representing the user's actual lifespan may, for example, represent the period from the user's birth until death or the onset of health problems, or it may represent the period from when the user begins receiving health guidance from the AI ​​agent until death or the onset of health problems, or it may represent the period from when System 100 predicts the remaining lifespan until death or the onset of health problems. The data representing the user's actual lifespan may be received, for example, from a medical institution such as a hospital or an administrative agency that manages family registers, or it may be received from the user's terminal device.

[0089] Data relating to a user's use of an AI agent may include at least one of the following: data indicating whether or not the user used an AI agent; data indicating the type of AI agent the user used; data indicating how often the user used an AI agent; and data indicating whether or not the user followed health guidance provided by the AI ​​agent. Data relating to a user's use of an AI agent may be received, for example, from a network cloud or from the user's terminal device.

[0090] The data received by the receiving means 121 is passed to the adjusting means 124.

[0091] The adjustment means 124 is configured to adjust the machine learning model using data representing the user's actual lifespan and data relating to the user's use of the AI ​​agent. Specifically, the adjustment means 124 retrains the machine learning model using data representing the user's actual lifespan and data relating to the user's use of the AI ​​agent, thereby generating an updated machine learning model.

[0092] As many users utilize system 100, actual results are accumulated, and these results can be used for retraining, potentially improving the accuracy of the machine learning model used by prediction means 123.

[0093] The prediction means 123 can predict the user's remaining lifespan using a modified machine learning model. This allows the prediction means 123 to predict the remaining lifespan with higher accuracy.

[0094] In the examples shown in Figures 4A and 4B above, each component of the processor unit 120 is provided within the same processor unit 120, but the present invention is not limited to this. A configuration in which each component of the processor unit 120 is distributed among multiple processor units is also within the scope of the present invention. In this case, the multiple processor units may be located within the same hardware component, or they may be located within separate hardware components that are nearby or far apart.

[0095] Furthermore, each component of the system 100 described above may consist of a single hardware component or multiple hardware components. If it consists of multiple hardware components, the manner in which each hardware component is connected is irrelevant. Each hardware component may be connected wirelessly or by wire. The system 100 of the present invention is not limited to a specific hardware configuration. Configuring the processor unit 120 with analog circuits instead of digital circuits is also within the scope of the present invention. The configuration of the system 100 of the present invention is not limited to those described above insofar as it can realize its functions.

[0096] 3. Processing in a system for predicting a user's remaining lifespan Figure 5 shows an example of a process 500 in system 100 for predicting a user's remaining lifespan. Process 500 may be executed in the processor unit 120.

[0097] In step S501, the receiving means 121 of the processor unit 120 receives data relating to the user. The data relating to the user includes data relating to the user's personality and data relating to the user's health. The receiving means 121 may, for example, receive data relating to the user's health from a medical institution such as a hospital, or it may receive data relating to the user's health from the user's terminal device. The receiving means 121 may, for example, receive data relating to the user's personality from a network cloud, or it may receive data relating to the user's personality from the user's terminal device.

[0098] In step S502, the derivation means 122 of the processor unit 120 derives the probability that the user is willing to use the AI ​​agent that provides health guidance, based on the user's personality data received in step S501. The derivation means 122 can derive the probability, for example, using a rule-based method. Alternatively, the derivation means 122 can derive the probability, for example, using a machine learning model.

[0099] In one example, the derivation means 122 may determine a type of AI agent specific to the user and derive the probability that the user is willing to use the determined AI agent. The derivation means 122 can determine a type of AI agent specific to the user based, for example, on personality data and / or health data received in step S501. The derivation means 122 can determine the user's personality from the personality data and / or their health status from the health data, and then determine an AI agent that is compatible with that personality and / or health status, for example, using a rule-based approach or machine learning.

[0100] The derivation means 122 may derive the probability that a user is willing to use an AI agent that provides health guidance, based on a score representing the user's characteristics instead of the personality data described above.

[0101] In step S503, the prediction means 123 of the processor unit 120 predicts the user's remaining lifespan based on the user's health data received in step S501 and the probability derived in step S502. The derivation means 122 can predict the user's remaining lifespan using, for example, a machine learning model.

[0102] The predicted life expectancy is output to the outside of system 100 and can be used for any purpose. For example, the predicted life expectancy is provided to the user's terminal device to inform the user of their life expectancy. For example, the predicted life expectancy can be provided to a provider of goods or services and used to determine a product or service that suits the user's life expectancy. Specifically, as an example, it can be provided to a financial instruments business operator that provides financial products, and the financial instruments business operator can propose financial products to the user, taking into account the user's life expectancy and its risks. For example, risk-segmented financial products (e.g., lifetime annuities or health insurance) can be offered, in which case the premium may be set based on the predicted life expectancy. This makes it possible to take measures against longevity risk. For example, the predicted life expectancy can be used for marketing and product development, and a target market can be identified based on the predicted life expectancy, making it possible to provide products that suit the consumer's lifestyle. For example, a product may be designed based on the predicted life expectancy and then provided to the user.

[0103] Data regarding user usage of the AI ​​agent, indicating whether or not the user actually used it, and data representing the user's actual lifespan are fed back to system 100. System 100 can then use this data to train a machine learning model. This allows for incremental improvement of the machine learning model's accuracy.

[0104] Referring to Figure 5, the above example illustrates that the processes are carried out in a specific order. However, the order of each process is not limited to what has been described, and can be carried out in any logically possible order.

[0105] Referring to the example described above with reference to Figure 5, the processing of each step shown in Figure 5 can be realized by the processor unit 120 and the program stored in the memory unit 130, but the present invention is not limited thereto. At least one of the processing of each step shown in Figure 5 may be realized by a hardware configuration such as a control circuit.

[0106] The present invention is not limited to the embodiments described above. It is understood that the scope of the present invention should be interpreted solely by the claims. Those skilled in the art will understand that, based on the description of specific preferred embodiments of the present invention, an equivalent scope can be practiced based on the description of the present invention and common technical knowledge. [Industrial applicability]

[0107] This invention is useful as it provides a system for predicting a user's remaining lifespan, etc. [Explanation of Symbols]

[0108] 100 Systems U User C Network Cloud H Hospital A AI agent

Claims

1. A system for predicting a user's remaining lifespan, wherein the system is A receiving means for receiving data relating to a user, wherein the data includes data relating to the user's personality and data relating to the user's health. A derivation means for deriving the probability that the user is willing to use an AI agent to provide health guidance, based on the aforementioned personality data, A prediction means for predicting the user's remaining lifespan based on the aforementioned health data and the aforementioned probability. A system equipped with these features.

2. The derivation means is, Based on the aforementioned personality data and / or health data, determine a type of AI agent specific to the user. The system according to claim 1, further configured to perform the derivation, wherein the derivation includes deriving the probability that the user is willing to use the determined AI agent.

3. The system according to claim 2, wherein the prediction means predicts the remaining lifespan based on the health data, the probability, and the characteristics of the determined AI agent.

4. The system according to claim 1, wherein the data relating to the personality includes at least one of the user's behavioral history, the user's SNS usage history, the user's purchase history, and a personality assessment.

5. The system according to claim 1, wherein the prediction means predicts the user's remaining lifespan using a machine learning model that has learned the relationship between data on the target's health, the probability that the target is willing to use an AI agent to provide health guidance, and the target's remaining lifespan.

6. The receiving means further receives data representing the user's actual lifespan and data relating to the user's use of the AI ​​agent. The system according to claim 5, further comprising adjustment means for adjusting the machine learning model using data representing the actual lifespan and data relating to usage.

7. A method for predicting a user's remaining lifespan, wherein the method is: Receiving data relating to a user, wherein the data includes data relating to the user's personality and data relating to the user's health. Based on the aforementioned personality data, the probability that the user would be willing to use an AI agent to provide health guidance is derived. To predict the user's remaining lifespan based on the aforementioned health data and the aforementioned probability. A method that includes this.

8. A program that predicts a user's remaining lifespan, wherein the program is executed in a system equipped with a processor, and the program is Receiving data relating to a user, wherein the data includes data relating to the user's personality and data relating to the user's health. Based on the aforementioned personality data, the probability that the user would be willing to use an AI agent to provide health guidance is derived. To predict the user's remaining lifespan based on the aforementioned health data and the aforementioned probability. A program that causes the processor to perform the processing including the specified operations.

Images