System

Abstract

A system is provided.SOLUTION: A system comprising: means for learning information on a person's annual income, family structure, and subscribed insurance and loans; means for acquiring information on various insurance and loan interest rates that change daily from an external source and updating a database; means for generating and providing an optimal insurance, loan review plan, and tax saving advice to a user; means for proposing optimal financial products according to the family structure and the passage of years and supporting procedures; and means for recording a dialogue log with the user and causing a AI model to continuously learn to improve the accuracy of advice.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The technology of the present disclosure relates to a system. [Background technology]

[0002] Patent document 1 discloses a persona chatbot control method performed by at least one processor, the method including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to a description of the chatbot character, encoding the prompt, and inputting the encoded prompt into a language model to generate a chatbot utterance in response to the user utterance. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2022-180282 Summary of the Invention [Problem to be solved by the invention]

[0004] Choosing financial products and managing loans is extremely complex for individuals, requiring specialized knowledge. Making optimal choices in a volatile financial market also requires constantly grasping the latest information and making decisions based on that information. Furthermore, it can be difficult to receive optimal financial advice tailored to changes in an individual's life stage, and not taking the right action at the right time can result in financial disadvantage. There is a need for a system that can solve these problems and consistently provide optimal financial advice to individuals. [Means for solving the problem]

[0005] The present invention provides a system that functions as a personal financial planner (FP). This system includes means for learning information about an individual's annual income, family composition, and insurance and loan enrollments, and for updating the database by obtaining information on various insurance and loan interest rates that change daily from external sources. Furthermore, the system includes means for generating and providing optimal insurance, loan review proposals, and tax-saving advice to the user. The system also includes means for proposing optimal financial products based on family composition and years of age, and supporting the procedures. Furthermore, the system includes means for recording a dialogue log with the user, continuously training an AI model, and improving the accuracy of the advice. This allows users to always receive the latest and most appropriate financial advice, enabling them to optimize their financial situation.

[0006] "User" refers to an individual who uses this system and provides information on annual income, family composition, and insurance and loan enrollment.

[0007] "Annual income" refers to the total income earned by a user in a year, including sources of income such as salary, bonuses, investment income, etc.

[0008] "Family structure" refers to the status of members in the user's household, including relationships with spouses, children, parents, and the like.

[0009] "Insurance" refers to various insurance products such as life insurance, health insurance, and automobile insurance that a user has subscribed to.

[0010] "Loan" refers to borrowed money including the amount borrowed by the user from a financial institution or the like, and includes a home loan, an education loan, a car loan, and the like.

[0011] "External sources" refers to financial institutions and other relevant information providers, such as those providing the latest insurance and loan interest rates, economic news, etc.

[0012] "Database" means an information management system for storing and updating information obtained from users and external sources.

[0013] "Advice" refers to suggestions provided to users regarding financial product selection, loan review, and tax savings.

[0014] "Dialogue log" refers to data that records the content of dialogue between a user and a system.

[0015] "AI model" refers to a predictive model built using machine learning algorithms based on user data and dialogue logs.

[0016] "Algorithm" refers to the calculation procedure by which the system processes user information and external information to generate optimal advice. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10]1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0018] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0019] First, the terms used in the following description will be explained.

[0020] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0021] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0022] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0023] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0024] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0025] [First embodiment]

[0026] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0027] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0028] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0029] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0030] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0031] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0032] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0033] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0034] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0035] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0036] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0037] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0038] This invention is a system that functions as a personal financial planner (FP) for each user. This system collects information on an individual's annual income, family structure, insurance and loan enrollment, and provides optimal advice based on financial information that changes daily. Below, the program processing of this system is described in detail in natural language.

[0039] Collecting and updating information

[0040] Collecting basic information

[0041] A user accesses a dedicated application or web interface and enters information about their annual income, family composition, insurance and loans. This information is received by the device and sent to a server, which stores it in a database and checks for consistency and format.

[0042] Retrieving and updating external information

[0043] The server periodically collects the latest information from external sources such as financial institutions and insurance companies, such as current insurance rates, loan interest rates, and economic news. This information is stored in a database and matched with the user's information.

[0044] Generating Advice

[0045] User information analysis

[0046] The server uses an algorithm to analyze the user information stored in the database and external information it has acquired, and generates advice on the best financial products for the user's current situation and tax-saving strategies.

[0047] Providing advice

[0048] The device provides advice and notifies the user. For example, if mortgage interest rates fall, the server generates a new loan offer and the device sends a notification to the user. When the user confirms the notification, the specific advice details are displayed.

[0049] Continuous learning and improvement

[0050] Collecting and learning from conversation logs

[0051] All interactions between users and the system are recorded and stored on a server. These interaction logs are used as training data for machine learning algorithms. The server periodically analyzes this data and learns new models to improve the accuracy of advice.

[0052] Deploying and applying the model

[0053] Once a new machine learning model is learned, the server deploys it and applies it to future advice generation, improving the accuracy of advice to users and providing more appropriate financial support.

[0054] Specific examples

[0055] Scenario 1: New life insurance proposal

[0056] The user reports the birth of a new child to the system. The device sends this information to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and presents the proposal.

[0057] Scenario 2: Loan Restructuring

[0058] The server retrieves the latest loan interest rate information from an external source and compares it with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan refinancing plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[0059] In this way, the system continuously collects and updates user information and provides optimal advice to optimize the user's financial situation, allowing users to easily select and use the financial products and services that are best suited to them.

[0060] The processing flow will be explained below.

[0061] Collecting and updating information

[0062] Step 1: Initial input of user information

[0063] The user logs into a dedicated application or web interface and enters information about their annual income, family composition, and insurance and loan enrollment.

[0064] The terminal receives the input information and transmits it to the server.

[0065] Step 2: Save user information

[0066] The server stores the received user information in a database.

[0067] The server checks the information for consistency and format validity.

[0068] Step 3: Obtaining external information

[0069] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies.

[0070] The server stores this information in a database and matches it with the user's information.

[0071] Generating Advice

[0072] Step 4: Analyzing User Information

[0073] The server inputs user information stored in a database and external information into an analysis algorithm.

[0074] The server generates optimal insurance, loan review, and tax saving advice based on the user's financial situation.

[0075] Step 5: Generate and save advice

[0076] The server stores the generated advice in a database as a separate plan.

[0077] The server converts the generated advice into JSON format or similar and sends it to the terminal.

[0078] Step 6: User notification and confirmation

[0079] The terminal sends an advice notification to the user, informing them that new advice is available.

[0080] The user acknowledges the notification and opens the application or web interface to view detailed advice.

[0081] Continuous learning and improvement

[0082] Step 7: Collect conversation logs

[0083] The server records all user interaction logs and stores them in a database.

[0084] Step 8: Training the model

[0085] The server periodically analyzes the dialogue logs and uses machine learning algorithms to train new models.

[0086] Step 9: Deploy the improved model

[0087] The server deploys the newly trained model and uses it for future advice generation.

[0088] Specific examples

[0089] Example 1: Proposing a new life insurance policy

[0090] Step 1: The user enters that a new child has been born.

[0091] The user uses the application to enter changes to their household.

[0092] The terminal receives this information and sends it to the server.

[0093] Step 2: Update the database

[0094] The server stores the new family structure information in the database.

[0095] The server compares the insurance coverage with existing insurance coverage and evaluates the need for new insurance.

[0096] Step 3: Generate and notify advice

[0097] The server generates the optimal life insurance plan and sends it to the terminal.

[0098] The device will send a notification to the user prompting them to confirm the suggestion.

[0099] Example 2: Loan review

[0100] Step 1: The server obtains external information.

[0101] The server retrieves the latest loan interest rate information from an external source.

[0102] The server stores this information in a database.

[0103] Step 2: Reevaluate the loan

[0104] The server compares the new interest rate information with the user's current loan terms.

[0105] If the server finds favorable terms, it generates a new loan plan.

[0106] Step 3: Generate and notify advice

[0107] The server transmits the generated loan plan to the terminal.

[0108] The device will then send a notification to the user prompting them to confirm the new loan offer.

[0109] This allows users to always receive the latest and most appropriate financial advice, optimizing their financial situation.

[0110] Example 1

[0111] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0112] Currently, many individuals struggle with selecting and managing complex financial products. It is particularly challenging to centrally manage information such as annual income, family composition, insurance and loan enrollment, and appropriately handle financial information that changes daily. Furthermore, to propose optimal financial products based on users' life events and provide prompt and accurate advice, highly accurate support utilizing dialogue history analysis and machine learning is required. A system that solves these challenges and enables users to efficiently select and manage optimal financial products is needed.

[0113] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0114] In this invention, the server includes a means for learning information on an individual's income, family structure, and insurance and loan enrollment, a means for acquiring information on various insurance and loan interest rates that change daily from external sources and updating the database, and a means for generating and providing optimal insurance and loan review proposals and tax saving advice to the user, thereby enabling the user to efficiently select and review optimal financial products based on the latest financial information.

[0115] "Income" is all the economic benefits that an individual or household receives over a given period of time.

[0116] "Family structure" refers to information about the number of family members who make up a household and their relationships.

[0117] "Insurance" is a financial product in which a fixed insurance premium is paid based on a contract, and insurance benefits can be received in the event of an accident or disaster.

[0118] A "loan" is a contract to borrow money from a financial institution on the condition that it will be repaid over a certain period of time.

[0119] "External sources" refer to information providers and information sources that exist outside the system, including financial institutions, insurance companies, and economic news.

[0120] A "database" is a system that efficiently stores and manages collected data and allows it to be searched and updated as needed.

[0121] An "artificial intelligence model" is a model that is built based on machine learning algorithms and makes decisions and predictions based on data.

[0122] "Life events" are events that have a significant impact on an individual's life, such as marriage, childbirth, transfer, and retirement.

[0123] "Financial products" are products such as insurance, loans, stocks, bonds, and investment trusts offered by financial institutions, which are intended for asset management and risk hedging.

[0124] "Dialogue history" is a record of communication between the user and the system, and is used to improve services in the future.

[0125] MODE FOR CARRYING OUT THE INVENTION

[0126] This invention is a system that functions as a personal financial planner (FP) for each user. The system collects information on an individual's income, family structure, insurance and loan enrollment, and provides optimal advice based on financial information that changes daily.

[0127] Collecting and updating information

[0128] Collecting basic information

[0129] A user accesses a dedicated application or web interface and enters information about their income, family composition, insurance and loans. This information is received by the device and sent to a server, which stores it in a database and checks for consistency and format.

[0130] Examples of hardware and software used:

[0131] Hardware: User's personal computer, smartphone

[0132] Software: web browsers, mobile applications, server-side database management systems (e.g., MySQL®)

[0133] Examples:

[0134] A user fills out a web form with information about their income (6 million yen), family composition (spouse and two children), and their life insurance and mortgage. This information is sent to the server and stored in a database.

[0135] Retrieving and updating external information

[0136] The server periodically contacts external sources, such as financial institutions and insurance companies, to obtain up-to-date information on current insurance rates, loan interest rates, economic news, etc. This information is stored in a database and matched with the user's information.

[0137] Examples of hardware and software used:

[0138] Hardware: Server

[0139] Software: External API, data collection scripts (e.g., Python)

[0140] Examples:

[0141] The server retrieves the latest loan interest rate information from the financial institution's API, stores it in a database, and compares it with the user's information.

[0142] Generating Advice

[0143] User information analysis

[0144] The server uses a generative AI model to analyze the user's information stored in the database and external information it has acquired, thereby generating advice on optimal financial products and tax-saving strategies.

[0145] Examples of hardware and software used:

[0146] Hardware: Server

[0147] Software: Data analysis tools (e.g., scikit-learn), generative AI models

[0148] Examples:

[0149] The server generates advice recommending additional life insurance coverage, taking into account the user's annual income of 6 million yen, family composition, and the latest loan interest rate of 1.2%.

[0150] Providing advice

[0151] The device provides advice and notifies the user. For example, if mortgage interest rates fall, the server generates a new loan offer and the device sends a notification to the user. When the user confirms the notification, the specific advice details are displayed.

[0152] Examples of hardware and software used:

[0153] Hardware: User's personal computer, smartphone

[0154] Software: Mobile applications, notification systems (e.g., Firebase Cloud Messaging)

[0155] Examples:

[0156] A notification will appear on the user's device saying, "We recommend you take out additional life insurance," and when they click on the details, specific insurance plans and benefits will be displayed.

[0157] Continuous learning and improvement

[0158] Collecting and learning from conversation logs

[0159] All interactions between the user and the system are recorded and stored on a server. These interaction logs are used as training data for machine learning algorithms. The server analyzes this data and learns new generative AI models to improve the accuracy of advice.

[0160] Examples of hardware and software used:

[0161] Hardware: Server

[0162] Software: Machine learning algorithms (e.g., TENSORFLOW®)

[0163] Examples:

[0164] If a user asks the system, "Please tell me more about the insurance plan," this conversation history is saved on the server and used to generate future advice.

[0165] Deploying and applying the model

[0166] Once a new generative AI model is learned, the server deploys it and applies it to future advice generation.

[0167] Examples:

[0168] The server learns a new generative AI model to improve the accuracy of life insurance advice, and uses this model for future advice generation.

[0169] Prompt Sentence Examples

[0170] Here are some example prompts to input to the generative AI model:

[0171] User: I have a baby and would like to review my life insurance.

[0172] Server: We're offering a new life insurance plan. Here are the details.

[0173] In this way, the information input by the user and the server's response based on that information can be presented in a natural conversational format.

[0174] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0175] Step 1: Gather basic information

[0176] 1. A user accesses an application or web interface and enters information about their income, family status, and insurance and loan coverage.

[0177] 2. The device sends the information entered by the user to the server via the backend API.

[0178] 3. The server stores the received information in a database, checking the consistency and format of the entered data.

[0179] Input: User's income, family structure, insurance information, loan information

[0180] Data processing: Check income, family structure, insurance information, loan information and save it in the database

[0181] Output: Save results to database

[0182] Specific behavior:

[0183] A user fills out a web form with information about their income of 6 million yen, their family composition (spouse and two children), and their life insurance and mortgage insurance policies.

[0184] The device sends the input information to the server via the backend API.

[0185] The server stores the received information in a database and records it as "annual income of 6 million yen," "spouse and two children," "life insurance insured," and "mortgage held."

[0186] Step 2: Retrieving and updating external information

[0187] 1. The server periodically contacts external sources such as financial institutions and insurance companies to obtain up-to-date information such as current insurance rates, loan interest rates, and economic news.

[0188] 2. The server stores the acquired information in a database and matches it with the user's information.

[0189] Input: Latest financial information obtained from external sources

[0190] Data processing: The latest financial information is stored in a database and compared with user information.

[0191] Output: Results saved to the database and results of matching with user information

[0192] Specific behavior:

[0193] Suppose the latest mortgage interest rate the server retrieves from the financial institution's API is 1.2% per annum.

[0194] The server stores this interest rate information in a database and compares it with the user's current loan interest rate.

[0195] Step 3: Analyze user information

[0196] 1. The server performs analysis using specific algorithms (such as linear regression or decision trees) based on the user's basic information and external information stored in the database.

[0197] 2. The server generates advice such as reviewing financial products that are best suited to the user's situation and tax-saving measures.

[0198] Input: User's basic information, external information

[0199] Data arithmetic: Analyzing data using algorithms

[0200] Output: Generated advice

[0201] Specific behavior:

[0202] The server takes into consideration the user's annual income of 6 million yen, family composition, life insurance policies, and the latest mortgage interest rate of 1.2%, and generates advice recommending additional life insurance.

[0203] Step 4: Providing advice

[0204] 1. The terminal notifies the user of the advice received from the server.

[0205] 2. The user checks the notification and sees the detailed advice on the device screen.

[0206] Input: Server-generated advice

[0207] Data processing: Notify the user of the advice content

[0208] Output: Notification to the user

[0209] Specific behavior:

[0210] A notification will appear on the user's device saying, "We recommend you take out additional life insurance," and when they click on the details, specific insurance plans and benefits will be displayed.

[0211] Step 5: Collect and learn from conversation logs

[0212] 1. The server records all interactions between the user and the system.

[0213] 2. The server periodically analyzes this data as training data for machine learning algorithms.

[0214] 3. The server trains new machine learning models to improve the accuracy of advice.

[0215] Input: Dialogue history

[0216] Data processing: Analyze dialogue history as training data

[0217] Output: The training results of the new machine learning model

[0218] Specific behavior:

[0219] If a user asks the system, "Please tell me more about the insurance plan," this conversation history is saved on the server and used to generate future advice.

[0220] Step 6: Deploy and apply the model

[0221] 1. The server trains a new generative AI model and deploys it to the system.

[0222] 2. The server applies the new model to future advice generation.

[0223] Input: A new generative AI model

[0224] Data processing: Deploy the new model to the system

[0225] Output: Apply the new model to future advice generation

[0226] Specific behavior:

[0227] The server learns a new generative AI model to improve the accuracy of life insurance advice, and uses this model for future advice generation.

[0228] (Application example 1)

[0229] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0230] Today's consumers often find it difficult to choose from a complex range of financial products and services. Finding the best financial plan for their individual circumstances and keeping up with constantly changing financial information can be a particularly challenging task. Furthermore, there is a lack of tools to analyze users' spending patterns and income information in real time and provide appropriate advice, making it difficult to manage assets efficiently.

[0231] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0232] In this invention, the server includes: means for learning information on an individual's annual income, family composition, and insurance and loan enrollments; means for obtaining information on various insurance and loan interest rates that changes daily from external sources and updating the database; means for generating and providing optimal insurance, loan review proposals, and tax-saving advice to the user; means for proposing optimal financial products based on family composition and years of service and supporting the procedures; means for recording a dialogue log with the user and continuously training an AI model to improve the accuracy of the advice; means for analyzing the user's spending patterns and income information and providing financial advice and savings proposals in real time; and means for obtaining the latest financial data from external sources, calculating income and expenditure balances, and generating advice. This enables users to easily select optimal financial products and services from complex financial products and realizes efficient asset management.

[0233] "Annual income" is the total amount of income an individual receives in a year.

[0234] "Family composition" refers to the composition of members in the household to which an individual belongs, and includes, for example, information on whether or not the individual has a spouse and children.

[0235] "Insurance" generally refers to financial products that individuals and companies contract to reduce risk, and includes life insurance, medical insurance, and automobile insurance.

[0236] A "loan" is a financial product that individuals or companies borrow from financial institutions and repay with a certain interest rate.

[0237] "External sources" refers to information sources provided by third parties such as financial institutions and insurance companies.

[0238] A "database" is a software system for efficiently storing, managing, and retrieving information.

[0239] "Advice" refers to specific advice or suggestions given to an individual.

[0240] "Spending patterns" are information that shows the trends and distribution of how individuals spend their money.

[0241] "Income information" refers to data about an individual's source and amount of income.

[0242] "Real-time" refers to the temporal characteristics of processing data and providing results almost immediately.

[0243] An "AI model" is an algorithm or program that uses machine learning or artificial intelligence to perform a specific task.

[0244] A "prompt" is an instruction or question that is input to a generative AI model and influences the output that the model generates.

[0245] MODE FOR CARRYING OUT THE INVENTION

[0246] This invention is a system that functions as a personal financial planner (FP) for each user, analyzing the user's income and expenditure information in real time and providing optimal financial advice based on that information. This system is implemented using the following hardware and software.

[0247] Collecting and updating information

[0248] Collecting basic information

[0249] Users use a dedicated application to enter information such as annual income, family composition, insurance and loan enrollment. This information is received by a device such as a smartphone and sent to a server, which stores the information in a database and checks for consistency and format.

[0250] Retrieving and updating external information

[0251] The server periodically collects information from external sources such as financial institutions and insurance companies, such as the latest insurance rates, loan interest rates, and economic news, which is then stored in a database and matched with the user's information.

[0252] Generating Advice

[0253] User information analysis

[0254] The server uses an algorithm to analyze the user information stored in the database and external information it has acquired, and generates advice on the best financial products for the user's current situation and tax-saving strategies.

[0255] Providing advice

[0256] The device provides advice and notifies the user. For example, if mortgage interest rates fall, the server generates a new loan offer and the device sends a notification to the user. When the user confirms the notification, the specific advice details are displayed.

[0257] Continuous learning and improvement

[0258] Collecting and learning from conversation logs

[0259] All interactions between users and the system are recorded and stored on a server. These interaction logs are used as training data for machine learning algorithms. The server periodically analyzes this data and learns new models to improve the accuracy of advice.

[0260] Deploying and applying the model

[0261] Once a new machine learning model is learned, the server deploys it and applies it to future advice generation, improving the accuracy of advice to users and providing more appropriate financial support.

[0262] Specific examples

[0263] Scenario 1: New life insurance proposal

[0264] The user reports the birth of a new child to the system. The device sends this information to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and presents the proposal.

[0265] Scenario 2: Loan Restructuring

[0266] The server retrieves the latest loan interest rate information from an external source and compares it with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan refinancing plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[0267] Prompt Sentence Examples

[0268] We collected information such as the user's income of 5 million yen, monthly expenses of 150,000 yen, 50,000 yen, and 30,000 yen, as well as family composition. The latest loan interest rate was obtained from an API as 4.0%. Analyze the user's income and expenditure balance and generate optimal financial advice based on an interest rate lower than the current loan interest rate of 4.5%.

[0269] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0270] Step 1:

[0271] The user uses a dedicated application to input basic information (annual income, family composition, insurance and loan information). This is the input information. The device sends this information to the server, which stores the received information in a database and performs formal checks. Through this process, the user's basic information is stored in the database.

[0272] Step 2:

[0273] The server periodically retrieves information such as the latest insurance rates, loan interest rates, and economic news from external sources (financial institutions and insurance companies). This information becomes the input of external financial data. The server stores the retrieved information in a database and compares it with the user's existing data. This updates the database with the latest financial information.

[0274] Step 3:

[0275] The server performs an algorithmic analysis based on user information and external financial information. At this time, the user's income and expenditure balance, insurance, and loan status are calculated. The input is the user's basic information and external financial information, and the output is the generated advice. Specifically, the system compares the user's income and expenses and creates a proposal for reviewing the optimal financial product based on fluctuations in loan interest rates.

[0276] Step 4:

[0277] The terminal receives advice from the server and provides it to the user in the form of a notification. At this time, the user is shown specific details of the advice. For example, the device may notify the user that mortgage interest rates have dropped and propose a new loan plan. It may also propose a life insurance plan that reflects changes in the user's family structure. This allows the user to receive the latest financial advice in real time.

[0278] Step 5:

[0279] All interactions between the user and the system are recorded and stored on the server. This adds the interaction log to a database. This interaction log serves as training data for future advice generation processes. The server uses machine learning algorithms to analyze this data and improve the accuracy of the advice.

[0280] Step 6:

[0281] Once a new machine learning model is trained, the server deploys it in real time and applies it to the next and subsequent advice generation. This process enables the system to always use the latest information and learning models to provide users with optimal financial advice.

[0282] Step 7:

[0283] For example, when a user reports the birth of a new child to the system, the device sends this information to the server. The server updates the database with the change in family structure and generates a new life insurance plan. The plan is then notified to the user via the device, and the proposed plan is presented in detail.

[0284] Prompt Sentence Examples

[0285] We collected information such as the user's income of 5 million yen, monthly expenses of 150,000 yen, 50,000 yen, and 30,000 yen, as well as family composition. The latest loan interest rate was obtained from an API as 4.0%. Analyze the user's income and expenditure balance and generate optimal financial advice based on an interest rate lower than the current loan interest rate of 4.5%.

[0286] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0287] This invention provides more accurate advice and services by combining a system that functions as a personal financial planner (FP) with an emotion engine that recognizes the user's emotions. This system collects information on an individual's annual income, family composition, and insurance and loan enrollment, and not only provides optimal advice based on financial information that changes daily, but also recognizes the user's emotions and takes an approach based on that emotional data. The program processing of this system is described in detail below in natural language.

[0288] Collecting and updating information

[0289] Collecting basic information

[0290] A user accesses a dedicated application or web interface and enters information about their annual income, family composition, insurance and loans. This information is received by the device and sent to a server, which stores it in a database and checks for consistency and format.

[0291] Retrieving and updating external information

[0292] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news from financial institutions and insurance companies through their APIs. This information is stored in a database and matched with user information.

[0293] Collecting emotional information

[0294] When the device interacts with the user, it activates the emotion engine and collects dialogue logs, voice data, and facial expression data, which are then sent to a server for emotion analysis.

[0295] Generating Advice

[0296] User information analysis

[0297] The server inputs the user information stored in the database and external information into an analytical algorithm, which generates advice on the optimal insurance, loan review, and tax saving measures for the user's current situation.

[0298] Emotional information analysis

[0299] The server analyzes the collected emotional data to recognize the user's current emotional state, and adjusts the content and presentation of advice based on the user's emotional state.

[0300] Generating and saving advice

[0301] The server saves the generated advice in a database as an individual plan, converts the generated advice into JSON format, etc., and sends it to the device.

[0302] Providing advice

[0303] The device sends an advice notification to the user informing them that new advice is available, and the user acknowledges the notification and opens the application or web interface to view the detailed advice.

[0304] Continuous learning and improvement

[0305] Collection and storage of conversation logs

[0306] All interactions between users and the system are recorded and stored on a server, and these interaction logs are used as training data for machine learning algorithms.

[0307] Model training

[0308] The server periodically analyzes the dialogue logs and uses machine learning algorithms to train new models, including emotional data, to improve advice accuracy and user satisfaction.

[0309] Deploying and applying the model

[0310] Once a new machine learning model is learned, the server deploys it and applies it to future advice generation, improving the accuracy of advice to users and providing more appropriate financial support.

[0311] Specific examples

[0312] Example 1: Proposing a new life insurance policy

[0313] scenario

[0314] The user reports the birth of a new child to the system. The device sends this information to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and asks them to confirm the proposal.

[0315] Use of emotion engine

[0316] When a user reviews a new suggestion, the device activates an emotion engine to analyze the user's reaction. For example, if the user is feeling stressed, the server will provide customized advice to reduce stress.

[0317] Example 2: Loan review

[0318] scenario

[0319] The server retrieves the latest loan interest rates from external sources and compares them with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[0320] Use of emotion engine

[0321] When a user receives a notification, the emotion engine analyzes the user's facial expression and tone of voice. If the user indicates anxiety, the server provides additional information to provide further explanation or reassurance.

[0322] In this way, the system continuously collects and updates user information and provides optimal advice, optimizing the user's financial situation and providing personalized support that takes into account their emotional state.

[0323] The processing flow will be explained below.

[0324] Collecting and updating information

[0325] Step 1: Initial input of user information

[0326] The user logs into a dedicated application or web interface and enters information about their annual income, family composition, and insurance and loan enrollment.

[0327] The terminal receives the input information and transmits it to the server.

[0328] Step 2: Save user information

[0329] The server stores the received user information in a database.

[0330] The server checks the information for consistency and format validity.

[0331] Step 3: Obtaining external information

[0332] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies.

[0333] The server stores this information in a database and matches it with the user's information.

[0334] Generating Advice

[0335] Step 4: Analyzing User Information

[0336] The server inputs user information stored in a database and external information into an analysis algorithm.

[0337] The server generates advice on the optimal insurance, loan review, and tax saving measures for the user's current situation.

[0338] Step 5: Collecting emotional information

[0339] When the device interacts with the user, it activates an emotion engine and collects dialogue logs, voice data, and facial expression data.

[0340] The device transmits the collected emotion data to a server.

[0341] Step 6: Emotional Analysis

[0342] The server analyzes the collected emotional data to recognize the user's current emotional state.

[0343] The server adjusts the content and presentation of advice based on the emotional state.

[0344] Step 7: Generate and save advice

[0345] The server stores the generated advice in a database as a separate plan.

[0346] The server converts the generated advice into JSON format or similar and sends it to the terminal.

[0347] Step 8: User notification and confirmation

[0348] The terminal sends an advice notification to the user, informing them that new advice is available.

[0349] The user acknowledges the notification and opens the application or web interface to view detailed advice.

[0350] Continuous learning and improvement

[0351] Step 9: Collect and store conversation logs

[0352] The server records all user interaction logs and stores them in a database.

[0353] Step 10: Train the model

[0354] The server periodically analyzes the dialogue logs and uses machine learning algorithms to train new models.

[0355] The server also learns from emotional data, improving the accuracy of advice and user satisfaction.

[0356] Step 11: Deploy and apply the model

[0357] The server deploys the newly trained model and applies it to future advice generation.

[0358] The server will provide better financial support based on the new model.

[0359] Specific examples

[0360] Example 1: Proposing a new life insurance policy

[0361] Step 1: Enter changes to family composition

[0362] A user reports the birth of a new child to the system.

[0363] The terminal receives this information and sends it to the server.

[0364] Step 2: Update the database

[0365] The server stores the new family structure information in the database.

[0366] The server compares the insurance coverage with existing insurance coverage and evaluates the need for new insurance.

[0367] Step 3: Collecting emotion data

[0368] The device activates the emotion engine when a user reports something and collects dialogue logs and voice data.

[0369] The device transmits the emotion data to the server.

[0370] Step 4: Analyze the sentiment data

[0371] The server analyzes the collected emotional data to understand how the user feels about the birth of the new child.

[0372] The server sets the content of advice in an approach that matches the user's emotions.

[0373] Step 5: Generate and notify advice

[0374] The server generates the optimal life insurance plan and sends it to the terminal.

[0375] The device will send a notification to the user prompting them to confirm the suggestion.

[0376] Example 2: Loan review

[0377] Step 1: Obtaining external information

[0378] The server retrieves the latest loan interest rate information from an external source.

[0379] The server stores this information in a database.

[0380] Step 2: Reevaluate the loan

[0381] The server compares the new interest rate information with the user's current loan terms.

[0382] If the server finds favorable terms, it generates a new loan plan.

[0383] Step 3: Collecting emotion data

[0384] The device activates an emotion engine when the user interacts with it, and collects voice data and facial expression data.

[0385] The device transmits the emotion data to the server.

[0386] Step 4: Analyze the sentiment data

[0387] The server analyzes the emotional data and determines the emotional state of the user when they receive the notification.

[0388] The server provides additional information to allay the user's concerns and doubts.

[0389] Step 5: Generate and notify advice

[0390] The server transmits the generated loan plan to the terminal.

[0391] The device will then send a notification to the user prompting them to confirm the new loan offer.

[0392] This allows users to always receive the latest and most appropriate financial advice, and also enjoy personalized support that takes into account their emotional state.

[0393] Example 2

[0394] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0395] In modern society, personal financial information is diverse, making it difficult to obtain accurate advice tailored to daily changing market information and individual circumstances. Furthermore, there is a lack of personalized support that takes into account users' emotions and stress levels, resulting in reduced user satisfaction and convenience. Furthermore, mechanisms for continuous learning and model improvement to improve the accuracy of financial advice have not been effectively implemented. There is a need for a system that can resolve these issues and provide more accurate financial advice.

[0396] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0397] In this invention, the server includes: means for inputting information on an individual's annual income, family composition, and insurance and loan enrollments; means for transmitting and storing the information input by the user to a database; means for acquiring information on various insurance and loan interest rates from external sources and periodically updating the database; means for collecting emotional information such as a dialogue log, voice data, and facial expression data using an emotion engine; means for generating optimal insurance and loan review proposals and tax-saving advice based on the collected user information and emotional data; means for storing the generated advice in a database and transmitting it to a terminal; means for adjusting the content and expression of the advice based on the emotional data and providing personalized information to the user; means for sending new advice and financial product plan change notifications to the user; means for recording a dialogue log with the user, periodically training a machine learning algorithm to improve the accuracy of the advice; and means for deploying a new machine learning model and applying it thereafter. This enables the provision of highly accurate personalized advice tailored to an individual's financial situation and emotional state.

[0398] "Individual annual income" is the total income earned by a user in a year.

[0399] "Family structure" is information that indicates the number of members in the user's household and their relationships.

[0400] "Insurance" refers to insurance contracts such as life insurance, medical insurance, and non-life insurance that the user has subscribed to.

[0401] A "loan" is a mortgage, car loan, personal loan, or other debt agreement taken out by a user.

[0402] A "database" is a part of a computer system where collected information is stored and managed.

[0403] "External Source" is an institution or data source that provides information from outside the system, such as a financial institution or insurance company.

[0404] An "emotion engine" is software or a system that analyzes dialogue logs, voice data, and facial expression data to identify a user's emotional state.

[0405] A "dialogue log" is data that records the content of conversations between a user and a system.

[0406] A "machine learning algorithm" is an algorithm that analyzes large amounts of data, recognizes patterns, and makes future predictions and classifications.

[0407] An "AI model" is a mathematical model built on machine learning algorithms for making predictions and classifications.

[0408] "Personalized information" is information or advice that is customized based on a user's individual situation or emotional state.

[0409] A "notification" is a message that informs the user that new information or advice is available.

[0410] A "Plan Change Notification" is a message informing a user of a proposed change to their financial product or insurance plan.

[0411] "Means" refers to the functions or processes used to achieve a particular purpose.

[0412] Collecting basic information

[0413] The user accesses a dedicated application or web interface and enters information such as annual income, family composition, and insurance and loan enrollment. The device receives this information and temporarily stores it. The device then sends the saved user information to a server, which stores it in a database and checks its consistency and format. This ensures that the user information is accurately registered in the database.

[0414] Retrieving and updating external information

[0415] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies. The server stores the information obtained from external sources in a database and compares it with user information. This ensures that the database always contains up-to-date information, forming the basis for generating accurate advice.

[0416] Collecting and analyzing emotional information

[0417] When a user begins interacting with the system, the device activates the emotion engine. The device collects emotional information, such as dialogue logs, voice data, and facial expression data, in real time and sends it to the server. The server then analyzes the received emotional data using a machine learning algorithm to identify the user's current emotional state. This allows the system to provide optimal advice based on the user's emotional state.

[0418] Generating and Serving Advice

[0419] The server uses analytical algorithms to generate optimal financial advice based on user information in the database and external information. The generated advice is stored in the database, converted to JSON format, etc., and sent to the device. The device sends an advice notification to the user, informing them that new advice is available. The user confirms the notification and views the detailed advice in the application or web interface.

[0420] Continuous training and model deployment

[0421] The server records dialogue logs with the user and stores them in a database. The dialogue logs and emotion data are periodically analyzed, and a new model is trained using a machine learning algorithm. The new trained model is deployed and applied to future advice generation. This improves advice accuracy and user satisfaction.

[0422] Specific examples

[0423] Example 1: New life insurance proposal

[0424] The user reports the birth of a new child to the system via a web interface. The device sends this information to the server, which updates the database with the change in family structure. The server generates a new life insurance plan taking the change in family structure into account. The device notifies the user of the new insurance plan proposal and asks them to confirm the proposal. As the user confirms the proposal, the device activates an emotion engine to analyze the user's reaction in real time. The server provides customized advice to reduce the user's stress.

[0425] Example 2: Loan Restructuring

[0426] The server retrieves the latest loan interest rate information from an external source and compares it with the user's current loan interest rate. If there is a favorable change in interest rate, the server generates a new loan plan and the device notifies the user. The user receives the notification, reviews the proposal, and decides whether to proceed. The device activates an emotion engine to analyze the user's facial expressions and tone of voice. If the user shows signs of anxiety, the server provides additional information to provide further explanation or reassurance.

[0427] Example prompt sentence:

[0428] "Please let us know about the birth of your new child and provide appropriate life insurance recommendations."

[0429] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0430] Step 1: Gather basic information

[0431] User initiated

[0432] Users access a dedicated application or web interface and enter information about their annual income, family composition, and insurance and loan coverage.

[0433] Input: Annual income, family composition, insurance, and loan information entered by the user.

[0434] Output: User information received by the device.

[0435] Terminal Processing

[0436] The device receives the information entered by the user and temporarily stores it.

[0437] Input: Information entered by the user.

[0438] Output: Data in a format that can be sent to the server.

[0439] Server Processing

[0440] The device sends the stored user information to the server, which stores this information in a database and checks its consistency and format.

[0441] Input: User information received from the device.

[0442] Output: User information stored in a database, data checked for consistency and format.

[0443] Step 2: Retrieving and updating external information

[0444] Server activation

[0445] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies.

[0446] Input: API request.

[0447] Output: Financial information obtained from external sources.

[0448] Server Processing

[0449] The server stores the information obtained from external sources in a database and matches it with user information.

[0450] Input: Financial information from external sources.

[0451] Output: The external information updated in the database.

[0452] Step 3: Collecting and analyzing emotional information

[0453] User initiated

[0454] When the user starts interacting with the system, the device activates the emotion engine.

[0455] Input: User interaction begins.

[0456] Output: Launch of emotion engine.

[0457] Terminal Processing

[0458] The device collects emotional information such as dialogue logs, voice data, and facial expression data in real time and sends it to the server.

[0459] Input: Dialogue logs, voice data, facial expression data.

[0460] Output: Emotion data sent to the server.

[0461] Server Processing

[0462] The server analyzes the received emotional data using machine learning algorithms to identify the user's current emotional state.

[0463] Input: Emotion data received from the device.

[0464] Output: Parsed emotional state data.

[0465] Step 4: Generating and serving advice

[0466] Server activation

[0467] The server uses analytical algorithms to generate optimal financial advice based on user information in the database and external information.

[0468] Input: User information in the database and external information.

[0469] Output: The generated financial advice.

[0470] Server Processing

[0471] The generated advice is stored in a database, converted to JSON format, etc., and sent to the terminal.

[0472] Input: Generated financial advice.

[0473] Output: Advice data in a format that can be sent to a terminal.

[0474] Terminal activation

[0475] The device sends an advice notification to the user informing them that new advice is available, and the user can view the notification and the detailed advice in the application or web interface.

[0476] Input: Advice data from the server.

[0477] Output: Notify the user and provide further advice.

[0478] Step 5: Continue training and deploy the model

[0479] Server activation

[0480] The server records the user interaction log and stores it in a database.

[0481] Input: User interaction log.

[0482] Output: Interaction logs stored in a database.

[0483] Server Processing

[0484] Dialogue logs and sentiment data are analyzed periodically, and new models are trained using machine learning algorithms.

[0485] Input: Dialogue logs and emotion data.

[0486] Output: A new trained machine learning model.

[0487] Server Processing

[0488] The new trained model is deployed and applied to future advice generation.

[0489] Input: A trained machine learning model.

[0490] Output: The new deployed machine learning model.

[0491] (Application example 2)

[0492] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0493] Conventional financial planner (FP) systems provide optimal advice based on a user's annual income, family structure, insurance, and loan information. However, because they do not take the user's emotional state into account, the receptivity and degree of personalization of the advice is insufficient. This poses a challenge in that they are unable to respond appropriately in situations where the user feels anxious or stressed. Furthermore, there is a need for a system that can quickly respond to changes in the user's daily consumption behavior and emotions.

[0494] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0495] In this invention, the server includes means for learning information on an individual's annual income, family composition, and insurance and loan enrollments, means for acquiring daily changing economic information from external sources and updating the database, means for generating and providing optimal insurance and loan review proposals and tax saving advice to the user, means for analyzing the user's facial expression data and voice data and recognizing the emotional state, means for adjusting the content and presentation of the advice based on the user's emotional data, and means for recording a dialogue log with the user and continuously training the AI ​​model to improve the accuracy of the advice. This enables personalized, highly accurate advice that takes into account both the user's economic situation and emotional state.

[0496] "Individual annual income" refers to the total income earned by a particular individual over a certain period of time.

[0497] "Family structure" refers to the composition of members and their relationships within a particular household.

[0498] "Insurance" is a contract that provides financial compensation for risks to persons or property.

[0499] A "loan" is a financial contract in which you promise to repay a borrowed amount under certain conditions.

[0500] "Means of learning" refers to the function of collecting, analyzing, and memorizing information.

[0501] "External source" refers to an information source provided from outside the system.

[0502] "Means for updating the database" refers to a function for updating existing information to the latest information.

[0503] "Optimal insurance and loan review proposals" refer to proposals for changes to insurance and loans that are most suitable for the user's current situation.

[0504] "Tax advice" means instructions or suggestions for minimizing your tax burden.

[0505] "Means of providing" refers to the function for delivering information and advice to users.

[0506] "Financial products" are products traded on the market, including stocks, bonds, investment trusts, etc.

[0507] "Means to support procedures" refers to support functions that allow users to smoothly carry out the necessary procedures.

[0508] An "interaction log" is a history of interactions between a user and a system.

[0509] An "AI model" is a data analysis model based on artificial intelligence.

[0510] "Means for recognizing emotional states" refers to a function that analyzes and recognizes emotions from input data such as the user's facial expressions and voice.

[0511] "Facial expression data and voice data" refers to digital information related to the user's facial expressions and voice.

[0512] "Means for adjusting the content and presentation of advice based on emotional data" refers to a function for changing the content and presentation of advice provided according to the user's emotional state.

[0513] This invention is a system that combines a personal financial planner (FP) system with an emotion recognition engine to provide highly accurate advice to users. Specifically, the system is configured using the following hardware and software:

[0514] System Components

[0515] 1. User data collection

[0516] The device (smartphone, tablet, smart glasses, etc.) provides an interface to collect information about the user's annual income, family composition, insurance, and loans. This information is sent to a server and stored in a database.

[0517] 2. Obtaining information from external sources

[0518] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies. This information is also stored in a database and compared with user information.

[0519] 3. Collecting emotional information

[0520] The device's built-in camera and microphone are used to collect the user's facial expression and voice data, which is then analyzed by an emotion recognition engine to recognize the user's emotional state.

[0521] 4. Generating Advice

[0522] The server uses user information stored in a database, external information, and emotional data to generate optimal advice, including insurance and loan restructuring and tax-saving strategies. It also takes into account the user's emotional state and adjusts the content and presentation of the advice.

[0523] 5. Providing Advice and Notification

[0524] The generated advice is converted to JSON format or similar and sent to the terminal. The terminal then sends an advice notification to the user informing them that new advice is available. The user can then confirm the notification and open the interface to view the detailed advice.

[0525] 6. Continuous learning and improvement

[0526] All user interaction logs are recorded and stored on the server. These interaction logs are used by the machine learning algorithm to learn new models and improve the accuracy of advice. Once a new model is learned, the server deploys it and applies it to future advice generation.

[0527] Specific examples

[0528] Example 1: Proposing a new life insurance policy

[0529] The user reports the birth of a new child to the system. This information is sent to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and asks them to confirm the proposal.

[0530] Use of Emotion Recognition

[0531] When a user confirms a new suggestion, the device activates an emotion recognition engine to analyze the user's reaction. For example, if the user is feeling stressed, the server will provide customized advice to reduce stress.

[0532] Example 2: Loan review

[0533] The server retrieves the latest loan interest rates from external sources and compares them with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[0534] Use of Emotion Recognition

[0535] When a user receives a notification, an emotion recognition engine analyzes the user's facial expressions and tone of voice. If the user indicates anxiety, the server provides additional information to provide further explanation or reassurance.

[0536] Prompt Sentence Examples

[0537] "Give me an example of a user opening a finance app and checking their latest finances and sentiment data."

[0538] "Give me an example of a system where your personal financial planner provides advice based on today's spending behavior and sentiment data."

[0539] In this way, the system enables personalized, highly accurate advice that takes into account both the user's financial situation and emotional state.

[0540] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0541] Step 1:

[0542] Collection of User Information

[0543] Input: User's annual income, family composition, insurance and loan information.

[0544] How it works: The user opens a dedicated application on a device such as a smartphone or tablet and enters the necessary information, which is then sent from the device to the server.

[0545] Output: The user information sent to the server.

[0546] Step 2:

[0547] Obtaining external information

[0548] Input: The API endpoint of the financial institution or insurance company.

[0549] How it works: The server periodically calls the API to retrieve information such as the latest insurance rates, loan interest rates, and economic news.

[0550] Output: A database containing up-to-date financial information.

[0551] Step 3:

[0552] Collecting emotional information

[0553] Input: User's facial expression data and voice data.

[0554] How it works: When a user uses the device, the camera and microphone are activated to collect the necessary emotional data. The emotion recognition engine analyzes this data to recognize the user's emotional state.

[0555] Output: Emotion data sent to the server.

[0556] Step 4:

[0557] Data analysis and advice generation

[0558] Input: User information, external financial information, sentiment data.

[0559] How it works: The server feeds all the information stored in the database into an analytical algorithm to generate advice on the best insurance, loan review, tax savings, etc. The content and presentation of the advice is adjusted based on the emotional data.

[0560] Output: The generated advice plan.

[0561] Step 5:

[0562] Notification and provision of advice

[0563] Input: The generated advice plan.

[0564] How it works: The server converts the advice plan into a format such as JSON and sends it to the device. The device then sends an advice notification to the user informing them that new advice is available. The user confirms the notification and opens the application to view the detailed advice.

[0565] Output: Advice sent to the user's device.

[0566] Step 6:

[0567] Interaction logging and continuous learning

[0568] Input: Log of interactions between the user and the system.

[0569] How it works: All user interactions with the system are logged and stored on a server. These logs are used as training data for machine learning algorithms to learn new AI models. Once a new model is learned, the server deploys it and applies it to future advice generation.

[0570] Output: The updated AI model.

[0571] This series of processing steps allows the system to take into account both the user's financial situation and emotional state and provide highly accurate and personalized advice.

[0572] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0573] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0574] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0575] [Second embodiment]

[0576] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0577] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0578] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0579] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0580] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0581] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0582] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0583] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0584] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0585] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0586] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0587] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0588] This invention is a system that functions as a personal financial planner (FP) for each user. This system collects information on an individual's annual income, family structure, insurance and loan enrollment, and provides optimal advice based on financial information that changes daily. Below, the program processing of this system is described in detail in natural language.

[0589] Collecting and updating information

[0590] Collecting basic information

[0591] A user accesses a dedicated application or web interface and enters information about their annual income, family composition, insurance and loans. This information is received by the device and sent to a server, which stores it in a database and checks for consistency and format.

[0592] Retrieving and updating external information

[0593] The server periodically collects the latest information from external sources such as financial institutions and insurance companies, such as current insurance rates, loan interest rates, and economic news. This information is stored in a database and matched with the user's information.

[0594] Generating Advice

[0595] User information analysis

[0596] The server uses an algorithm to analyze the user information stored in the database and external information it has acquired, and generates advice on the best financial products for the user's current situation and tax-saving strategies.

[0597] Providing advice

[0598] The device provides advice and notifies the user. For example, if mortgage interest rates fall, the server generates a new loan offer and the device sends a notification to the user. When the user confirms the notification, the specific advice details are displayed.

[0599] Continuous learning and improvement

[0600] Collecting and learning from conversation logs

[0601] All interactions between users and the system are recorded and stored on a server. These interaction logs are used as training data for machine learning algorithms. The server periodically analyzes this data and learns new models to improve the accuracy of advice.

[0602] Deploying and applying the model

[0603] Once a new machine learning model is learned, the server deploys it and applies it to future advice generation, improving the accuracy of advice to users and providing more appropriate financial support.

[0604] Specific examples

[0605] Scenario 1: New life insurance proposal

[0606] The user reports the birth of a new child to the system. The device sends this information to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and presents the proposal.

[0607] Scenario 2: Loan Restructuring

[0608] The server retrieves the latest loan interest rate information from an external source and compares it with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan refinancing plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[0609] In this way, the system continuously collects and updates user information and provides optimal advice to optimize the user's financial situation, allowing users to easily select and use the financial products and services that are best suited to them.

[0610] The processing flow will be explained below.

[0611] Collecting and updating information

[0612] Step 1: Initial input of user information

[0613] The user logs into a dedicated application or web interface and enters information about their annual income, family composition, and insurance and loan enrollment.

[0614] The terminal receives the input information and transmits it to the server.

[0615] Step 2: Save user information

[0616] The server stores the received user information in a database.

[0617] The server checks the information for consistency and format validity.

[0618] Step 3: Obtaining external information

[0619] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies.

[0620] The server stores this information in a database and matches it with the user's information.

[0621] Generating Advice

[0622] Step 4: Analyzing User Information

[0623] The server inputs user information stored in a database and external information into an analysis algorithm.

[0624] The server generates optimal insurance, loan review, and tax saving advice based on the user's financial situation.

[0625] Step 5: Generate and save advice

[0626] The server stores the generated advice in a database as a separate plan.

[0627] The server converts the generated advice into JSON format or similar and sends it to the terminal.

[0628] Step 6: User notification and confirmation

[0629] The terminal sends an advice notification to the user, informing them that new advice is available.

[0630] The user acknowledges the notification and opens the application or web interface to view detailed advice.

[0631] Continuous learning and improvement

[0632] Step 7: Collect conversation logs

[0633] The server records all user interaction logs and stores them in a database.

[0634] Step 8: Training the model

[0635] The server periodically analyzes the dialogue logs and uses machine learning algorithms to train new models.

[0636] Step 9: Deploy the improved model

[0637] The server deploys the newly trained model and uses it for future advice generation.

[0638] Specific examples

[0639] Example 1: Proposing a new life insurance policy

[0640] Step 1: The user enters that a new child has been born.

[0641] The user uses the application to enter changes to their household.

[0642] The terminal receives this information and sends it to the server.

[0643] Step 2: Update the database

[0644] The server stores the new family structure information in the database.

[0645] The server compares the insurance coverage with existing insurance coverage and evaluates the need for new insurance.

[0646] Step 3: Generate and notify advice

[0647] The server generates the optimal life insurance plan and sends it to the terminal.

[0648] The device will send a notification to the user prompting them to confirm the suggestion.

[0649] Example 2: Loan review

[0650] Step 1: The server obtains external information.

[0651] The server retrieves the latest loan interest rate information from an external source.

[0652] The server stores this information in a database.

[0653] Step 2: Reevaluate the loan

[0654] The server compares the new interest rate information with the user's current loan terms.

[0655] If the server finds favorable terms, it generates a new loan plan.

[0656] Step 3: Generate and notify advice

[0657] The server transmits the generated loan plan to the terminal.

[0658] The device will then send a notification to the user prompting them to confirm the new loan offer.

[0659] This allows users to always receive the latest and most appropriate financial advice, optimizing their financial situation.

[0660] Example 1

[0661] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0662] Currently, many individuals struggle with selecting and managing complex financial products. It is particularly challenging to centrally manage information such as annual income, family composition, insurance and loan enrollment, and appropriately handle financial information that changes daily. Furthermore, to propose optimal financial products based on users' life events and provide prompt and accurate advice, highly accurate support utilizing dialogue history analysis and machine learning is required. A system that solves these challenges and enables users to efficiently select and manage optimal financial products is needed.

[0663] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0664] In this invention, the server includes a means for learning information on an individual's income, family structure, and insurance and loan enrollment, a means for acquiring information on various insurance and loan interest rates that change daily from external sources and updating the database, and a means for generating and providing optimal insurance and loan review proposals and tax saving advice to the user, thereby enabling the user to efficiently select and review optimal financial products based on the latest financial information.

[0665] "Income" is all the economic benefits that an individual or household receives over a given period of time.

[0666] "Family structure" refers to information about the number of family members who make up a household and their relationships.

[0667] "Insurance" is a financial product in which a fixed insurance premium is paid based on a contract, and insurance benefits can be received in the event of an accident or disaster.

[0668] A "loan" is a contract to borrow money from a financial institution on the condition that it will be repaid over a certain period of time.

[0669] "External sources" refer to information providers and information sources that exist outside the system, including financial institutions, insurance companies, and economic news.

[0670] A "database" is a system that efficiently stores and manages collected data and allows it to be searched and updated as needed.

[0671] An "artificial intelligence model" is a model that is built based on machine learning algorithms and makes decisions and predictions based on data.

[0672] "Life events" are events that have a significant impact on an individual's life, such as marriage, childbirth, transfer, and retirement.

[0673] "Financial products" are products such as insurance, loans, stocks, bonds, and investment trusts offered by financial institutions, which are intended for asset management and risk hedging.

[0674] "Dialogue history" is a record of communication between the user and the system, and is used to improve services in the future.

[0675] MODE FOR CARRYING OUT THE INVENTION

[0676] This invention is a system that functions as a personal financial planner (FP) for each user. The system collects information on an individual's income, family structure, insurance and loan enrollment, and provides optimal advice based on financial information that changes daily.

[0677] Collecting and updating information

[0678] Collecting basic information

[0679] A user accesses a dedicated application or web interface and enters information about their income, family composition, insurance and loans. This information is received by the device and sent to a server, which stores it in a database and checks for consistency and format.

[0680] Examples of hardware and software used:

[0681] Hardware: User's personal computer, smartphone

[0682] Software: web browsers, mobile applications, server-side database management systems (e.g., MySQL)

[0683] Examples:

[0684] A user fills out a web form with information about their income (6 million yen), family composition (spouse and two children), and their life insurance and mortgage. This information is sent to the server and stored in a database.

[0685] Retrieving and updating external information

[0686] The server periodically contacts external sources, such as financial institutions and insurance companies, to obtain up-to-date information on current insurance rates, loan interest rates, economic news, etc. This information is stored in a database and matched with the user's information.

[0687] Examples of hardware and software used:

[0688] Hardware: Server

[0689] Software: External API, data collection scripts (e.g., Python)

[0690] Examples:

[0691] The server retrieves the latest loan interest rate information from the financial institution's API, stores it in a database, and compares it with the user's information.

[0692] Generating Advice

[0693] User information analysis

[0694] The server uses a generative AI model to analyze the user's information stored in the database and external information it has acquired, thereby generating advice on optimal financial products and tax-saving strategies.

[0695] Examples of hardware and software used:

[0696] Hardware: Server

[0697] Software: Data analysis tools (e.g., scikit-learn), generative AI models

[0698] Examples:

[0699] The server generates advice recommending additional life insurance coverage, taking into account the user's annual income of 6 million yen, family composition, and the latest loan interest rate of 1.2%.

[0700] Providing advice

[0701] The device provides advice and notifies the user. For example, if mortgage interest rates fall, the server generates a new loan offer and the device sends a notification to the user. When the user confirms the notification, the specific advice details are displayed.

[0702] Examples of hardware and software used:

[0703] Hardware: User's personal computer, smartphone

[0704] Software: Mobile applications, notification systems (e.g., Firebase Cloud Messaging)

[0705] Examples:

[0706] A notification will appear on the user's device saying, "We recommend you take out additional life insurance," and when they click on the details, specific insurance plans and benefits will be displayed.

[0707] Continuous learning and improvement

[0708] Collecting and learning from conversation logs

[0709] All interactions between the user and the system are recorded and stored on a server. These interaction logs are used as training data for machine learning algorithms. The server analyzes this data and learns new generative AI models to improve the accuracy of advice.

[0710] Examples of hardware and software used:

[0711] Hardware: Server

[0712] Software: Machine learning algorithms (e.g., TensorFlow)

[0713] Examples:

[0714] If a user asks the system, "Please tell me more about the insurance plan," this conversation history is saved on the server and used to generate future advice.

[0715] Deploying and applying the model

[0716] Once a new generative AI model is learned, the server deploys it and applies it to future advice generation.

[0717] Examples:

[0718] The server learns a new generative AI model to improve the accuracy of life insurance advice, and uses this model for future advice generation.

[0719] Prompt Sentence Examples

[0720] Here are some example prompts to input to the generative AI model:

[0721] User: I have a baby and would like to review my life insurance.

[0722] Server: We're offering a new life insurance plan. Here are the details.

[0723] In this way, the information input by the user and the server's response based on that information can be presented in a natural conversational format.

[0724] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0725] Step 1: Gather basic information

[0726] 1. A user accesses an application or web interface and enters information about their income, family status, and insurance and loan coverage.

[0727] 2. The device sends the information entered by the user to the server via the backend API.

[0728] 3. The server stores the received information in a database, checking the consistency and format of the entered data.

[0729] Input: User's income, family structure, insurance information, loan information

[0730] Data processing: Check income, family structure, insurance information, loan information and save it in the database

[0731] Output: Save results to database

[0732] Specific behavior:

[0733] A user fills out a web form with information about their income of 6 million yen, their family composition (spouse and two children), and their life insurance and mortgage insurance policies.

[0734] The device sends the input information to the server via the backend API.

[0735] The server stores the received information in a database and records it as "annual income of 6 million yen," "spouse and two children," "life insurance insured," and "mortgage held."

[0736] Step 2: Retrieving and updating external information

[0737] 1. The server periodically contacts external sources such as financial institutions and insurance companies to obtain up-to-date information such as current insurance rates, loan interest rates, and economic news.

[0738] 2. The server stores the acquired information in a database and matches it with the user's information.

[0739] Input: Latest financial information obtained from external sources

[0740] Data processing: The latest financial information is stored in a database and compared with user information.

[0741] Output: Results saved to the database and results of matching with user information

[0742] Specific behavior:

[0743] Suppose the latest mortgage interest rate the server retrieves from the financial institution's API is 1.2% per annum.

[0744] The server stores this interest rate information in a database and compares it with the user's current loan interest rate.

[0745] Step 3: Analyze user information

[0746] 1. The server performs analysis using specific algorithms (such as linear regression or decision trees) based on the user's basic information and external information stored in the database.

[0747] 2. The server generates advice such as reviewing financial products that are best suited to the user's situation and tax-saving measures.

[0748] Input: User's basic information, external information

[0749] Data arithmetic: Analyzing data using algorithms

[0750] Output: Generated advice

[0751] Specific behavior:

[0752] The server takes into consideration the user's annual income of 6 million yen, family composition, life insurance policies, and the latest mortgage interest rate of 1.2%, and generates advice recommending additional life insurance.

[0753] Step 4: Providing advice

[0754] 1. The terminal notifies the user of the advice received from the server.

[0755] 2. The user checks the notification and sees the detailed advice on the device screen.

[0756] Input: Server-generated advice

[0757] Data processing: Notify the user of the advice content

[0758] Output: Notification to the user

[0759] Specific behavior:

[0760] A notification will appear on the user's device saying, "We recommend you take out additional life insurance," and when they click on the details, specific insurance plans and benefits will be displayed.

[0761] Step 5: Collect and learn from conversation logs

[0762] 1. The server records all interactions between the user and the system.

[0763] 2. The server periodically analyzes this data as training data for machine learning algorithms.

[0764] 3. The server trains new machine learning models to improve the accuracy of advice.

[0765] Input: Dialogue history

[0766] Data processing: Analyze dialogue history as training data

[0767] Output: The training results of the new machine learning model

[0768] Specific behavior:

[0769] If a user asks the system, "Please tell me more about the insurance plan," this conversation history is saved on the server and used to generate future advice.

[0770] Step 6: Deploy and apply the model

[0771] 1. The server trains a new generative AI model and deploys it to the system.

[0772] 2. The server applies the new model to future advice generation.

[0773] Input: A new generative AI model

[0774] Data processing: Deploy the new model to the system

[0775] Output: Apply the new model to future advice generation

[0776] Specific behavior:

[0777] The server learns a new generative AI model to improve the accuracy of life insurance advice, and uses this model for future advice generation.

[0778] (Application example 1)

[0779] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0780] Today's consumers often find it difficult to choose from a complex range of financial products and services. Finding the best financial plan for their individual circumstances and keeping up with constantly changing financial information can be a particularly challenging task. Furthermore, there is a lack of tools to analyze users' spending patterns and income information in real time and provide appropriate advice, making it difficult to manage assets efficiently.

[0781] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0782] In this invention, the server includes: means for learning information on an individual's annual income, family composition, and insurance and loan enrollments; means for obtaining information on various insurance and loan interest rates that changes daily from external sources and updating the database; means for generating and providing optimal insurance, loan review proposals, and tax-saving advice to the user; means for proposing optimal financial products based on family composition and years of service and supporting the procedures; means for recording a dialogue log with the user and continuously training an AI model to improve the accuracy of the advice; means for analyzing the user's spending patterns and income information and providing financial advice and savings proposals in real time; and means for obtaining the latest financial data from external sources, calculating income and expenditure balances, and generating advice. This enables users to easily select optimal financial products and services from complex financial products and realizes efficient asset management.

[0783] "Annual income" is the total amount of income an individual receives in a year.

[0784] "Family composition" refers to the composition of members in the household to which an individual belongs, and includes, for example, information on whether or not the individual has a spouse and children.

[0785] "Insurance" generally refers to financial products that individuals and companies contract to reduce risk, and includes life insurance, medical insurance, and automobile insurance.

[0786] A "loan" is a financial product that individuals or companies borrow from financial institutions and repay with a certain interest rate.

[0787] "External sources" refers to information sources provided by third parties such as financial institutions and insurance companies.

[0788] A "database" is a software system for efficiently storing, managing, and retrieving information.

[0789] "Advice" refers to specific advice or suggestions given to an individual.

[0790] "Spending patterns" are information that shows the trends and distribution of how individuals spend their money.

[0791] "Income information" refers to data about an individual's source and amount of income.

[0792] "Real-time" refers to the temporal characteristics of processing data and providing results almost immediately.

[0793] An "AI model" is an algorithm or program that uses machine learning or artificial intelligence to perform a specific task.

[0794] A "prompt" is an instruction or question that is input to a generative AI model and influences the output that the model generates.

[0795] MODE FOR CARRYING OUT THE INVENTION

[0796] This invention is a system that functions as a personal financial planner (FP) for each user, analyzing the user's income and expenditure information in real time and providing optimal financial advice based on that information. This system is implemented using the following hardware and software.

[0797] Collecting and updating information

[0798] Collecting basic information

[0799] Users use a dedicated application to enter information such as annual income, family composition, insurance and loan enrollment. This information is received by a device such as a smartphone and sent to a server, which stores the information in a database and checks for consistency and format.

[0800] Retrieving and updating external information

[0801] The server periodically collects information from external sources such as financial institutions and insurance companies, such as the latest insurance rates, loan interest rates, and economic news, which is then stored in a database and matched with the user's information.

[0802] Generating Advice

[0803] User information analysis

[0804] The server uses an algorithm to analyze the user information stored in the database and external information it has acquired, and generates advice on the best financial products for the user's current situation and tax-saving strategies.

[0805] Providing advice

[0806] The device provides advice and notifies the user. For example, if mortgage interest rates fall, the server generates a new loan offer and the device sends a notification to the user. When the user confirms the notification, the specific advice details are displayed.

[0807] Continuous learning and improvement

[0808] Collecting and learning from conversation logs

[0809] All interactions between users and the system are recorded and stored on a server. These interaction logs are used as training data for machine learning algorithms. The server periodically analyzes this data and learns new models to improve the accuracy of advice.

[0810] Deploying and applying the model

[0811] Once a new machine learning model is learned, the server deploys it and applies it to future advice generation, improving the accuracy of advice to users and providing more appropriate financial support.

[0812] Specific examples

[0813] Scenario 1: New life insurance proposal

[0814] The user reports the birth of a new child to the system. The device sends this information to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and presents the proposal.

[0815] Scenario 2: Loan Restructuring

[0816] The server retrieves the latest loan interest rate information from an external source and compares it with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan refinancing plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[0817] Prompt Sentence Examples

[0818] We collected information such as the user's income of 5 million yen, monthly expenses of 150,000 yen, 50,000 yen, and 30,000 yen, as well as family composition. The latest loan interest rate was obtained from an API as 4.0%. Analyze the user's income and expenditure balance and generate optimal financial advice based on an interest rate lower than the current loan interest rate of 4.5%.

[0819] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0820] Step 1:

[0821] The user uses a dedicated application to input basic information (annual income, family composition, insurance and loan information). This is the input information. The device sends this information to the server, which stores the received information in a database and performs formal checks. Through this process, the user's basic information is stored in the database.

[0822] Step 2:

[0823] The server periodically retrieves information such as the latest insurance rates, loan interest rates, and economic news from external sources (financial institutions and insurance companies). This information becomes the input of external financial data. The server stores the retrieved information in a database and compares it with the user's existing data. This updates the database with the latest financial information.

[0824] Step 3:

[0825] The server performs an algorithmic analysis based on user information and external financial information. At this time, the user's income and expenditure balance, insurance, and loan status are calculated. The input is the user's basic information and external financial information, and the output is the generated advice. Specifically, the system compares the user's income and expenses and creates a proposal for reviewing the optimal financial product based on fluctuations in loan interest rates.

[0826] Step 4:

[0827] The terminal receives advice from the server and provides it to the user in the form of a notification. At this time, the user is shown specific details of the advice. For example, the device may notify the user that mortgage interest rates have dropped and propose a new loan plan. It may also propose a life insurance plan that reflects changes in the user's family structure. This allows the user to receive the latest financial advice in real time.

[0828] Step 5:

[0829] All interactions between the user and the system are recorded and stored on the server. This adds the interaction log to a database. This interaction log serves as training data for future advice generation processes. The server uses machine learning algorithms to analyze this data and improve the accuracy of the advice.

[0830] Step 6:

[0831] Once a new machine learning model is trained, the server deploys it in real time and applies it to the next and subsequent advice generation. This process enables the system to always use the latest information and learning models to provide users with optimal financial advice.

[0832] Step 7:

[0833] For example, when a user reports the birth of a new child to the system, the device sends this information to the server. The server updates the database with the change in family structure and generates a new life insurance plan. The plan is then notified to the user via the device, and the proposed plan is presented in detail.

[0834] Prompt Sentence Examples

[0835] We collected information such as the user's income of 5 million yen, monthly expenses of 150,000 yen, 50,000 yen, and 30,000 yen, as well as family composition. The latest loan interest rate was obtained from an API as 4.0%. Analyze the user's income and expenditure balance and generate optimal financial advice based on an interest rate lower than the current loan interest rate of 4.5%.

[0836] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0837] This invention provides more accurate advice and services by combining a system that functions as a personal financial planner (FP) with an emotion engine that recognizes the user's emotions. This system collects information on an individual's annual income, family composition, and insurance and loan enrollment, and not only provides optimal advice based on financial information that changes daily, but also recognizes the user's emotions and takes an approach based on that emotional data. The program processing of this system is described in detail below in natural language.

[0838] Collecting and updating information

[0839] Collecting basic information

[0840] A user accesses a dedicated application or web interface and enters information about their annual income, family composition, insurance and loans. This information is received by the device and sent to a server, which stores it in a database and checks for consistency and format.

[0841] Retrieving and updating external information

[0842] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news from financial institutions and insurance companies through their APIs. This information is stored in a database and matched with user information.

[0843] Collecting emotional information

[0844] When the device interacts with the user, it activates the emotion engine and collects dialogue logs, voice data, and facial expression data, which are then sent to a server for emotion analysis.

[0845] Generating Advice

[0846] User information analysis

[0847] The server inputs the user information stored in the database and external information into an analytical algorithm, which generates advice on the optimal insurance, loan review, and tax saving measures for the user's current situation.

[0848] Emotional information analysis

[0849] The server analyzes the collected emotional data to recognize the user's current emotional state, and adjusts the content and presentation of advice based on the user's emotional state.

[0850] Generating and saving advice

[0851] The server saves the generated advice in a database as an individual plan, converts the generated advice into JSON format, etc., and sends it to the device.

[0852] Providing advice

[0853] The device sends an advice notification to the user informing them that new advice is available, and the user acknowledges the notification and opens the application or web interface to view the detailed advice.

[0854] Continuous learning and improvement

[0855] Collection and storage of conversation logs

[0856] All interactions between users and the system are recorded and stored on a server, and these interaction logs are used as training data for machine learning algorithms.

[0857] Model training

[0858] The server periodically analyzes the dialogue logs and uses machine learning algorithms to train new models, including emotional data, to improve advice accuracy and user satisfaction.

[0859] Deploying and applying the model

[0860] Once a new machine learning model is learned, the server deploys it and applies it to future advice generation, improving the accuracy of advice to users and providing more appropriate financial support.

[0861] Specific examples

[0862] Example 1: Proposing a new life insurance policy

[0863] scenario

[0864] The user reports the birth of a new child to the system. The device sends this information to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and asks them to confirm the proposal.

[0865] Use of emotion engine

[0866] When a user reviews a new suggestion, the device activates an emotion engine to analyze the user's reaction. For example, if the user is feeling stressed, the server will provide customized advice to reduce stress.

[0867] Example 2: Loan review

[0868] scenario

[0869] The server retrieves the latest loan interest rates from external sources and compares them with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[0870] Use of emotion engine

[0871] When a user receives a notification, the emotion engine analyzes the user's facial expression and tone of voice. If the user indicates anxiety, the server provides additional information to provide further explanation or reassurance.

[0872] In this way, the system continuously collects and updates user information and provides optimal advice, optimizing the user's financial situation and providing personalized support that takes into account their emotional state.

[0873] The processing flow will be explained below.

[0874] Collecting and updating information

[0875] Step 1: Initial input of user information

[0876] The user logs into a dedicated application or web interface and enters information about their annual income, family composition, and insurance and loan enrollment.

[0877] The terminal receives the input information and transmits it to the server.

[0878] Step 2: Save user information

[0879] The server stores the received user information in a database.

[0880] The server checks the information for consistency and format validity.

[0881] Step 3: Obtaining external information

[0882] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies.

[0883] The server stores this information in a database and matches it with the user's information.

[0884] Generating Advice

[0885] Step 4: Analyzing User Information

[0886] The server inputs user information stored in a database and external information into an analysis algorithm.

[0887] The server generates advice on the optimal insurance, loan review, and tax saving measures for the user's current situation.

[0888] Step 5: Collecting emotional information

[0889] When the device interacts with the user, it activates an emotion engine and collects dialogue logs, voice data, and facial expression data.

[0890] The device transmits the collected emotion data to a server.

[0891] Step 6: Emotional Analysis

[0892] The server analyzes the collected emotional data to recognize the user's current emotional state.

[0893] The server adjusts the content and presentation of advice based on the emotional state.

[0894] Step 7: Generate and save advice

[0895] The server stores the generated advice in a database as a separate plan.

[0896] The server converts the generated advice into JSON format or similar and sends it to the terminal.

[0897] Step 8: User notification and confirmation

[0898] The terminal sends an advice notification to the user, informing them that new advice is available.

[0899] The user acknowledges the notification and opens the application or web interface to view detailed advice.

[0900] Continuous learning and improvement

[0901] Step 9: Collect and store conversation logs

[0902] The server records all user interaction logs and stores them in a database.

[0903] Step 10: Train the model

[0904] The server periodically analyzes the dialogue logs and uses machine learning algorithms to train new models.

[0905] The server also learns from emotional data, improving the accuracy of advice and user satisfaction.

[0906] Step 11: Deploy and apply the model

[0907] The server deploys the newly trained model and applies it to future advice generation.

[0908] The server will provide better financial support based on the new model.

[0909] Specific examples

[0910] Example 1: Proposing a new life insurance policy

[0911] Step 1: Enter changes to family composition

[0912] A user reports the birth of a new child to the system.

[0913] The terminal receives this information and sends it to the server.

[0914] Step 2: Update the database

[0915] The server stores the new family structure information in the database.

[0916] The server compares the insurance coverage with existing insurance coverage and evaluates the need for new insurance.

[0917] Step 3: Collecting emotion data

[0918] The device activates the emotion engine when a user reports something and collects dialogue logs and voice data.

[0919] The device transmits the emotion data to the server.

[0920] Step 4: Analyze the sentiment data

[0921] The server analyzes the collected emotional data to understand how the user feels about the birth of the new child.

[0922] The server sets the content of advice in an approach that matches the user's emotions.

[0923] Step 5: Generate and notify advice

[0924] The server generates the optimal life insurance plan and sends it to the terminal.

[0925] The device will send a notification to the user prompting them to confirm the suggestion.

[0926] Example 2: Loan review

[0927] Step 1: Obtaining external information

[0928] The server retrieves the latest loan interest rate information from an external source.

[0929] The server stores this information in a database.

[0930] Step 2: Reevaluate the loan

[0931] The server compares the new interest rate information with the user's current loan terms.

[0932] If the server finds favorable terms, it generates a new loan plan.

[0933] Step 3: Collecting emotion data

[0934] The device activates an emotion engine when the user interacts with it, and collects voice data and facial expression data.

[0935] The device transmits the emotion data to the server.

[0936] Step 4: Analyze the sentiment data

[0937] The server analyzes the emotional data and determines the emotional state of the user when they receive the notification.

[0938] The server provides additional information to allay the user's concerns and doubts.

[0939] Step 5: Generate and notify advice

[0940] The server transmits the generated loan plan to the terminal.

[0941] The device will then send a notification to the user prompting them to confirm the new loan offer.

[0942] This allows users to always receive the latest and most appropriate financial advice, and also enjoy personalized support that takes into account their emotional state.

[0943] Example 2

[0944] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0945] In modern society, personal financial information is diverse, making it difficult to obtain accurate advice tailored to daily changing market information and individual circumstances. Furthermore, there is a lack of personalized support that takes into account users' emotions and stress levels, resulting in reduced user satisfaction and convenience. Furthermore, mechanisms for continuous learning and model improvement to improve the accuracy of financial advice have not been effectively implemented. There is a need for a system that can resolve these issues and provide more accurate financial advice.

[0946] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0947] In this invention, the server includes: means for inputting information on an individual's annual income, family composition, and insurance and loan enrollments; means for transmitting and storing the information input by the user to a database; means for acquiring information on various insurance and loan interest rates from external sources and periodically updating the database; means for collecting emotional information such as a dialogue log, voice data, and facial expression data using an emotion engine; means for generating optimal insurance and loan review proposals and tax-saving advice based on the collected user information and emotional data; means for storing the generated advice in a database and transmitting it to a terminal; means for adjusting the content and expression of the advice based on the emotional data and providing personalized information to the user; means for sending new advice and financial product plan change notifications to the user; means for recording a dialogue log with the user, periodically training a machine learning algorithm to improve the accuracy of the advice; and means for deploying a new machine learning model and applying it thereafter. This enables the provision of highly accurate personalized advice tailored to an individual's financial situation and emotional state.

[0948] "Individual annual income" is the total income earned by a user in a year.

[0949] "Family structure" is information that indicates the number of members in the user's household and their relationships.

[0950] "Insurance" refers to insurance contracts such as life insurance, medical insurance, and non-life insurance that the user has subscribed to.

[0951] A "loan" is a mortgage, car loan, personal loan, or other debt agreement taken out by a user.

[0952] A "database" is a part of a computer system where collected information is stored and managed.

[0953] "External Source" is an institution or data source that provides information from outside the system, such as a financial institution or insurance company.

[0954] An "emotion engine" is software or a system that analyzes dialogue logs, voice data, and facial expression data to identify a user's emotional state.

[0955] A "dialogue log" is data that records the content of conversations between a user and a system.

[0956] A "machine learning algorithm" is an algorithm that analyzes large amounts of data, recognizes patterns, and makes future predictions and classifications.

[0957] An "AI model" is a mathematical model built on machine learning algorithms for making predictions and classifications.

[0958] "Personalized information" is information or advice that is customized based on a user's individual situation or emotional state.

[0959] A "notification" is a message that informs the user that new information or advice is available.

[0960] A "Plan Change Notification" is a message informing a user of a proposed change to their financial product or insurance plan.

[0961] "Means" refers to the functions or processes used to achieve a particular purpose.

[0962] Collecting basic information

[0963] The user accesses a dedicated application or web interface and enters information such as annual income, family composition, and insurance and loan enrollment. The device receives this information and temporarily stores it. The device then sends the saved user information to a server, which stores it in a database and checks its consistency and format. This ensures that the user information is accurately registered in the database.

[0964] Retrieving and updating external information

[0965] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies. The server stores the information obtained from external sources in a database and compares it with user information. This ensures that the database always contains up-to-date information, forming the basis for generating accurate advice.

[0966] Collecting and analyzing emotional information

[0967] When a user begins interacting with the system, the device activates the emotion engine. The device collects emotional information, such as dialogue logs, voice data, and facial expression data, in real time and sends it to the server. The server then analyzes the received emotional data using a machine learning algorithm to identify the user's current emotional state. This allows the system to provide optimal advice based on the user's emotional state.

[0968] Generating and Serving Advice

[0969] The server uses analytical algorithms to generate optimal financial advice based on user information in the database and external information. The generated advice is stored in the database, converted to JSON format, etc., and sent to the device. The device sends an advice notification to the user, informing them that new advice is available. The user confirms the notification and views the detailed advice in the application or web interface.

[0970] Continuous training and model deployment

[0971] The server records dialogue logs with the user and stores them in a database. The dialogue logs and emotion data are periodically analyzed, and a new model is trained using a machine learning algorithm. The new trained model is deployed and applied to future advice generation. This improves advice accuracy and user satisfaction.

[0972] Specific examples

[0973] Example 1: New life insurance proposal

[0974] The user reports the birth of a new child to the system via a web interface. The device sends this information to the server, which updates the database with the change in family structure. The server generates a new life insurance plan taking the change in family structure into account. The device notifies the user of the new insurance plan proposal and asks them to confirm the proposal. As the user confirms the proposal, the device activates an emotion engine to analyze the user's reaction in real time. The server provides customized advice to reduce the user's stress.

[0975] Example 2: Loan Restructuring

[0976] The server retrieves the latest loan interest rate information from an external source and compares it with the user's current loan interest rate. If there is a favorable change in interest rate, the server generates a new loan plan and the device notifies the user. The user receives the notification, reviews the proposal, and decides whether to proceed. The device activates an emotion engine to analyze the user's facial expressions and tone of voice. If the user shows signs of anxiety, the server provides additional information to provide further explanation or reassurance.

[0977] Example prompt sentence:

[0978] "Please let us know about the birth of your new child and provide appropriate life insurance recommendations."

[0979] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0980] Step 1: Gather basic information

[0981] User initiated

[0982] Users access a dedicated application or web interface and enter information about their annual income, family composition, and insurance and loan coverage.

[0983] Input: Annual income, family composition, insurance, and loan information entered by the user.

[0984] Output: User information received by the device.

[0985] Terminal Processing

[0986] The device receives the information entered by the user and temporarily stores it.

[0987] Input: Information entered by the user.

[0988] Output: Data in a format that can be sent to the server.

[0989] Server Processing

[0990] The device sends the stored user information to the server, which stores this information in a database and checks its consistency and format.

[0991] Input: User information received from the device.

[0992] Output: User information stored in a database, data checked for consistency and format.

[0993] Step 2: Retrieving and updating external information

[0994] Server activation

[0995] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies.

[0996] Input: API request.

[0997] Output: Financial information obtained from external sources.

[0998] Server Processing

[0999] The server stores the information obtained from external sources in a database and matches it with user information.

[1000] Input: Financial information from external sources.

[1001] Output: The external information updated in the database.

[1002] Step 3: Collecting and analyzing emotional information

[1003] User initiated

[1004] When the user starts interacting with the system, the device activates the emotion engine.

[1005] Input: User interaction begins.

[1006] Output: Launch of emotion engine.

[1007] Terminal Processing

[1008] The device collects emotional information such as dialogue logs, voice data, and facial expression data in real time and sends it to the server.

[1009] Input: Dialogue logs, voice data, facial expression data.

[1010] Output: Emotion data sent to the server.

[1011] Server Processing

[1012] The server analyzes the received emotional data using machine learning algorithms to identify the user's current emotional state.

[1013] Input: Emotion data received from the device.

[1014] Output: Parsed emotional state data.

[1015] Step 4: Generating and serving advice

[1016] Server activation

[1017] The server uses analytical algorithms to generate optimal financial advice based on user information in the database and external information.

[1018] Input: User information in the database and external information.

[1019] Output: The generated financial advice.

[1020] Server Processing

[1021] The generated advice is stored in a database, converted to JSON format, etc., and sent to the terminal.

[1022] Input: Generated financial advice.

[1023] Output: Advice data in a format that can be sent to a terminal.

[1024] Terminal activation

[1025] The device sends an advice notification to the user informing them that new advice is available, and the user can view the notification and the detailed advice in the application or web interface.

[1026] Input: Advice data from the server.

[1027] Output: Notify the user and provide further advice.

[1028] Step 5: Continue training and deploy the model

[1029] Server activation

[1030] The server records the user interaction log and stores it in a database.

[1031] Input: User interaction log.

[1032] Output: Interaction logs stored in a database.

[1033] Server Processing

[1034] Dialogue logs and sentiment data are analyzed periodically, and new models are trained using machine learning algorithms.

[1035] Input: Dialogue logs and emotion data.

[1036] Output: A new trained machine learning model.

[1037] Server Processing

[1038] The new trained model is deployed and applied to future advice generation.

[1039] Input: A trained machine learning model.

[1040] Output: The new deployed machine learning model.

[1041] (Application example 2)

[1042] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[1043] Conventional financial planner (FP) systems provide optimal advice based on a user's annual income, family structure, insurance, and loan information. However, because they do not take the user's emotional state into account, the receptivity and degree of personalization of the advice is insufficient. This poses a challenge in that they are unable to respond appropriately in situations where the user feels anxious or stressed. Furthermore, there is a need for a system that can quickly respond to changes in the user's daily consumption behavior and emotions.

[1044] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1045] In this invention, the server includes means for learning information on an individual's annual income, family composition, and insurance and loan enrollments, means for acquiring daily changing economic information from external sources and updating the database, means for generating and providing optimal insurance and loan review proposals and tax saving advice to the user, means for analyzing the user's facial expression data and voice data and recognizing the emotional state, means for adjusting the content and presentation of the advice based on the user's emotional data, and means for recording a dialogue log with the user and continuously training the AI ​​model to improve the accuracy of the advice. This enables personalized, highly accurate advice that takes into account both the user's economic situation and emotional state.

[1046] "Individual annual income" refers to the total income earned by a particular individual over a certain period of time.

[1047] "Family structure" refers to the composition of members and their relationships within a particular household.

[1048] "Insurance" is a contract that provides financial compensation for risks to persons or property.

[1049] A "loan" is a financial contract in which you promise to repay a borrowed amount under certain conditions.

[1050] "Means of learning" refers to the function of collecting, analyzing, and memorizing information.

[1051] "External source" refers to an information source provided from outside the system.

[1052] "Means for updating the database" refers to a function for updating existing information to the latest information.

[1053] "Optimal insurance and loan review proposals" refer to proposals for changes to insurance and loans that are most suitable for the user's current situation.

[1054] "Tax advice" means instructions or suggestions for minimizing your tax burden.

[1055] "Means of providing" refers to the function for delivering information and advice to users.

[1056] "Financial products" are products traded on the market, including stocks, bonds, investment trusts, etc.

[1057] "Means to support procedures" refers to support functions that allow users to smoothly carry out the necessary procedures.

[1058] An "interaction log" is a history of interactions between a user and a system.

[1059] An "AI model" is a data analysis model based on artificial intelligence.

[1060] "Means for recognizing emotional states" refers to a function that analyzes and recognizes emotions from input data such as the user's facial expressions and voice.

[1061] "Facial expression data and voice data" refers to digital information related to the user's facial expressions and voice.

[1062] "Means for adjusting the content and presentation of advice based on emotional data" refers to a function for changing the content and presentation of advice provided according to the user's emotional state.

[1063] This invention is a system that combines a personal financial planner (FP) system with an emotion recognition engine to provide highly accurate advice to users. Specifically, the system is configured using the following hardware and software:

[1064] System Components

[1065] 1. User data collection

[1066] The device (smartphone, tablet, smart glasses, etc.) provides an interface to collect information about the user's annual income, family composition, insurance, and loans. This information is sent to a server and stored in a database.

[1067] 2. Obtaining information from external sources

[1068] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies. This information is also stored in a database and compared with user information.

[1069] 3. Collecting emotional information

[1070] The device's built-in camera and microphone are used to collect the user's facial expression and voice data, which is then analyzed by an emotion recognition engine to recognize the user's emotional state.

[1071] 4. Generating Advice

[1072] The server uses user information stored in a database, external information, and emotional data to generate optimal advice, including insurance and loan restructuring and tax-saving strategies. It also takes into account the user's emotional state and adjusts the content and presentation of the advice.

[1073] 5. Providing Advice and Notification

[1074] The generated advice is converted to JSON format or similar and sent to the terminal. The terminal then sends an advice notification to the user informing them that new advice is available. The user can then confirm the notification and open the interface to view the detailed advice.

[1075] 6. Continuous learning and improvement

[1076] All user interaction logs are recorded and stored on the server. These interaction logs are used by the machine learning algorithm to learn new models and improve the accuracy of advice. Once a new model is learned, the server deploys it and applies it to future advice generation.

[1077] Specific examples

[1078] Example 1: Proposing a new life insurance policy

[1079] The user reports the birth of a new child to the system. This information is sent to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and asks them to confirm the proposal.

[1080] Use of Emotion Recognition

[1081] When a user confirms a new suggestion, the device activates an emotion recognition engine to analyze the user's reaction. For example, if the user is feeling stressed, the server will provide customized advice to reduce stress.

[1082] Example 2: Loan review

[1083] The server retrieves the latest loan interest rates from external sources and compares them with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[1084] Use of Emotion Recognition

[1085] When a user receives a notification, an emotion recognition engine analyzes the user's facial expressions and tone of voice. If the user indicates anxiety, the server provides additional information to provide further explanation or reassurance.

[1086] Prompt Sentence Examples

[1087] "Give me an example of a user opening a finance app and checking their latest finances and sentiment data."

[1088] "Give me an example of a system where your personal financial planner provides advice based on today's spending behavior and sentiment data."

[1089] In this way, the system enables personalized, highly accurate advice that takes into account both the user's financial situation and emotional state.

[1090] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1091] Step 1:

[1092] Collection of User Information

[1093] Input: User's annual income, family composition, insurance and loan information.

[1094] How it works: The user opens a dedicated application on a device such as a smartphone or tablet and enters the necessary information, which is then sent from the device to the server.

[1095] Output: The user information sent to the server.

[1096] Step 2:

[1097] Obtaining external information

[1098] Input: The API endpoint of the financial institution or insurance company.

[1099] How it works: The server periodically calls the API to retrieve information such as the latest insurance rates, loan interest rates, and economic news.

[1100] Output: A database containing up-to-date financial information.

[1101] Step 3:

[1102] Collecting emotional information

[1103] Input: User's facial expression data and voice data.

[1104] How it works: When a user uses the device, the camera and microphone are activated to collect the necessary emotional data. The emotion recognition engine analyzes this data to recognize the user's emotional state.

[1105] Output: Emotion data sent to the server.

[1106] Step 4:

[1107] Data analysis and advice generation

[1108] Input: User information, external financial information, sentiment data.

[1109] How it works: The server feeds all the information stored in the database into an analytical algorithm to generate advice on the best insurance, loan review, tax savings, etc. The content and presentation of the advice is adjusted based on the emotional data.

[1110] Output: The generated advice plan.

[1111] Step 5:

[1112] Notification and provision of advice

[1113] Input: The generated advice plan.

[1114] How it works: The server converts the advice plan into a format such as JSON and sends it to the device. The device then sends an advice notification to the user informing them that new advice is available. The user confirms the notification and opens the application to view the detailed advice.

[1115] Output: Advice sent to the user's device.

[1116] Step 6:

[1117] Interaction logging and continuous learning

[1118] Input: Log of interactions between the user and the system.

[1119] How it works: All user interactions with the system are logged and stored on a server. These logs are used as training data for machine learning algorithms to learn new AI models. Once a new model is learned, the server deploys it and applies it to future advice generation.

[1120] Output: The updated AI model.

[1121] This series of processing steps allows the system to take into account both the user's financial situation and emotional state and provide highly accurate and personalized advice.

[1122] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1123] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1124] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[1125] [Third embodiment]

[1126] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1127] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1128] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1129] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[1130] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1131] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1132] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1133] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1134] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1135] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1136] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1137] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[1138] This invention is a system that functions as a personal financial planner (FP) for each user. This system collects information on an individual's annual income, family structure, insurance and loan enrollment, and provides optimal advice based on financial information that changes daily. Below, the program processing of this system is described in detail in natural language.

[1139] Collecting and updating information

[1140] Collecting basic information

[1141] A user accesses a dedicated application or web interface and enters information about their annual income, family composition, insurance and loans. This information is received by the device and sent to a server, which stores it in a database and checks for consistency and format.

[1142] Retrieving and updating external information

[1143] The server periodically collects the latest information from external sources such as financial institutions and insurance companies, such as current insurance rates, loan interest rates, and economic news. This information is stored in a database and matched with the user's information.

[1144] Generating Advice

[1145] User information analysis

[1146] The server uses an algorithm to analyze the user information stored in the database and external information it has acquired, and generates advice on the best financial products for the user's current situation and tax-saving strategies.

[1147] Providing advice

[1148] The device provides advice and notifies the user. For example, if mortgage interest rates fall, the server generates a new loan offer and the device sends a notification to the user. When the user confirms the notification, the specific advice details are displayed.

[1149] Continuous learning and improvement

[1150] Collecting and learning from conversation logs

[1151] All interactions between users and the system are recorded and stored on a server. These interaction logs are used as training data for machine learning algorithms. The server periodically analyzes this data and learns new models to improve the accuracy of advice.

[1152] Deploying and applying the model

[1153] Once a new machine learning model is learned, the server deploys it and applies it to future advice generation, improving the accuracy of advice to users and providing more appropriate financial support.

[1154] Specific examples

[1155] Scenario 1: New life insurance proposal

[1156] The user reports the birth of a new child to the system. The device sends this information to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and presents the proposal.

[1157] Scenario 2: Loan Restructuring

[1158] The server retrieves the latest loan interest rate information from an external source and compares it with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan refinancing plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[1159] In this way, the system continuously collects and updates user information and provides optimal advice to optimize the user's financial situation, allowing users to easily select and use the financial products and services that are best suited to them.

[1160] The processing flow will be explained below.

[1161] Collecting and updating information

[1162] Step 1: Initial input of user information

[1163] The user logs into a dedicated application or web interface and enters information about their annual income, family composition, and insurance and loan enrollment.

[1164] The terminal receives the input information and transmits it to the server.

[1165] Step 2: Save user information

[1166] The server stores the received user information in a database.

[1167] The server checks the information for consistency and format validity.

[1168] Step 3: Obtaining external information

[1169] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies.

[1170] The server stores this information in a database and matches it with the user's information.

[1171] Generating Advice

[1172] Step 4: Analyzing User Information

[1173] The server inputs user information stored in a database and external information into an analysis algorithm.

[1174] The server generates optimal insurance, loan review, and tax saving advice based on the user's financial situation.

[1175] Step 5: Generate and save advice

[1176] The server stores the generated advice in a database as a separate plan.

[1177] The server converts the generated advice into JSON format or similar and sends it to the terminal.

[1178] Step 6: User notification and confirmation

[1179] The terminal sends an advice notification to the user, informing them that new advice is available.

[1180] The user acknowledges the notification and opens the application or web interface to view detailed advice.

[1181] Continuous learning and improvement

[1182] Step 7: Collect conversation logs

[1183] The server records all user interaction logs and stores them in a database.

[1184] Step 8: Training the model

[1185] The server periodically analyzes the dialogue logs and uses machine learning algorithms to train new models.

[1186] Step 9: Deploy the improved model

[1187] The server deploys the newly trained model and uses it for future advice generation.

[1188] Specific examples

[1189] Example 1: Proposing a new life insurance policy

[1190] Step 1: The user enters that a new child has been born.

[1191] The user uses the application to enter changes to their household.

[1192] The terminal receives this information and sends it to the server.

[1193] Step 2: Update the database

[1194] The server stores the new family structure information in the database.

[1195] The server compares the insurance coverage with existing insurance coverage and evaluates the need for new insurance.

[1196] Step 3: Generate and notify advice

[1197] The server generates the optimal life insurance plan and sends it to the terminal.

[1198] The device will send a notification to the user prompting them to confirm the suggestion.

[1199] Example 2: Loan review

[1200] Step 1: The server obtains external information.

[1201] The server retrieves the latest loan interest rate information from an external source.

[1202] The server stores this information in a database.

[1203] Step 2: Reevaluate the loan

[1204] The server compares the new interest rate information with the user's current loan terms.

[1205] If the server finds favorable terms, it generates a new loan plan.

[1206] Step 3: Generate and notify advice

[1207] The server transmits the generated loan plan to the terminal.

[1208] The device will then send a notification to the user prompting them to confirm the new loan offer.

[1209] This allows users to always receive the latest and most appropriate financial advice, optimizing their financial situation.

[1210] Example 1

[1211] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1212] Currently, many individuals struggle with selecting and managing complex financial products. It is particularly challenging to centrally manage information such as annual income, family composition, insurance and loan enrollment, and appropriately handle financial information that changes daily. Furthermore, to propose optimal financial products based on users' life events and provide prompt and accurate advice, highly accurate support utilizing dialogue history analysis and machine learning is required. A system that solves these challenges and enables users to efficiently select and manage optimal financial products is needed.

[1213] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1214] In this invention, the server includes a means for learning information on an individual's income, family structure, and insurance and loan enrollment, a means for acquiring information on various insurance and loan interest rates that change daily from external sources and updating the database, and a means for generating and providing optimal insurance and loan review proposals and tax saving advice to the user, thereby enabling the user to efficiently select and review optimal financial products based on the latest financial information.

[1215] "Income" is all the economic benefits that an individual or household receives over a given period of time.

[1216] "Family structure" refers to information about the number of family members who make up a household and their relationships.

[1217] "Insurance" is a financial product in which a fixed insurance premium is paid based on a contract, and insurance benefits can be received in the event of an accident or disaster.

[1218] A "loan" is a contract to borrow money from a financial institution on the condition that it will be repaid over a certain period of time.

[1219] "External sources" refer to information providers and information sources that exist outside the system, including financial institutions, insurance companies, and economic news.

[1220] A "database" is a system that efficiently stores and manages collected data and allows it to be searched and updated as needed.

[1221] An "artificial intelligence model" is a model that is built based on machine learning algorithms and makes decisions and predictions based on data.

[1222] "Life events" are events that have a significant impact on an individual's life, such as marriage, childbirth, transfer, and retirement.

[1223] "Financial products" are products such as insurance, loans, stocks, bonds, and investment trusts offered by financial institutions, which are intended for asset management and risk hedging.

[1224] "Dialogue history" is a record of communication between the user and the system, and is used to improve services in the future.

[1225] MODE FOR CARRYING OUT THE INVENTION

[1226] This invention is a system that functions as a personal financial planner (FP) for each user. The system collects information on an individual's income, family structure, insurance and loan enrollment, and provides optimal advice based on financial information that changes daily.

[1227] Collecting and updating information

[1228] Collecting basic information

[1229] A user accesses a dedicated application or web interface and enters information about their income, family composition, insurance and loans. This information is received by the device and sent to a server, which stores it in a database and checks for consistency and format.

[1230] Examples of hardware and software used:

[1231] Hardware: User's personal computer, smartphone

[1232] Software: web browsers, mobile applications, server-side database management systems (e.g., MySQL)

[1233] Examples:

[1234] A user fills out a web form with information about their income (6 million yen), family composition (spouse and two children), and their life insurance and mortgage. This information is sent to the server and stored in a database.

[1235] Retrieving and updating external information

[1236] The server periodically contacts external sources, such as financial institutions and insurance companies, to obtain up-to-date information on current insurance rates, loan interest rates, economic news, etc. This information is stored in a database and matched with the user's information.

[1237] Examples of hardware and software used:

[1238] Hardware: Server

[1239] Software: External API, data collection scripts (e.g., Python)

[1240] Examples:

[1241] The server retrieves the latest loan interest rate information from the financial institution's API, stores it in a database, and compares it with the user's information.

[1242] Generating Advice

[1243] User information analysis

[1244] The server uses a generative AI model to analyze the user's information stored in the database and external information it has acquired, thereby generating advice on optimal financial products and tax-saving strategies.

[1245] Examples of hardware and software used:

[1246] Hardware: Server

[1247] Software: Data analysis tools (e.g., scikit-learn), generative AI models

[1248] Examples:

[1249] The server generates advice recommending additional life insurance coverage, taking into account the user's annual income of 6 million yen, family composition, and the latest loan interest rate of 1.2%.

[1250] Providing advice

[1251] The device provides advice and notifies the user. For example, if mortgage interest rates fall, the server generates a new loan offer and the device sends a notification to the user. When the user confirms the notification, the specific advice details are displayed.

[1252] Examples of hardware and software used:

[1253] Hardware: User's personal computer, smartphone

[1254] Software: Mobile applications, notification systems (e.g., Firebase Cloud Messaging)

[1255] Examples:

[1256] A notification will appear on the user's device saying, "We recommend you take out additional life insurance," and when they click on the details, specific insurance plans and benefits will be displayed.

[1257] Continuous learning and improvement

[1258] Collecting and learning from conversation logs

[1259] All interactions between the user and the system are recorded and stored on a server. These interaction logs are used as training data for machine learning algorithms. The server analyzes this data and learns new generative AI models to improve the accuracy of advice.

[1260] Examples of hardware and software used:

[1261] Hardware: Server

[1262] Software: Machine learning algorithms (e.g., TensorFlow)

[1263] Examples:

[1264] If a user asks the system, "Please tell me more about the insurance plan," this conversation history is saved on the server and used to generate future advice.

[1265] Deploying and applying the model

[1266] Once a new generative AI model is learned, the server deploys it and applies it to future advice generation.

[1267] Examples:

[1268] The server learns a new generative AI model to improve the accuracy of life insurance advice, and uses this model for future advice generation.

[1269] Prompt Sentence Examples

[1270] Here are some example prompts to input to the generative AI model:

[1271] User: I have a baby and would like to review my life insurance.

[1272] Server: We're offering a new life insurance plan. Here are the details.

[1273] In this way, the information input by the user and the server's response based on that information can be presented in a natural conversational format.

[1274] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1275] Step 1: Gather basic information

[1276] 1. A user accesses an application or web interface and enters information about their income, family status, and insurance and loan coverage.

[1277] 2. The device sends the information entered by the user to the server via the backend API.

[1278] 3. The server stores the received information in a database, checking the consistency and format of the entered data.

[1279] Input: User's income, family structure, insurance information, loan information

[1280] Data processing: Check income, family structure, insurance information, loan information and save it in the database

[1281] Output: Save results to database

[1282] Specific behavior:

[1283] A user fills out a web form with information about their income of 6 million yen, their family composition (spouse and two children), and their life insurance and mortgage insurance policies.

[1284] The device sends the input information to the server via the backend API.

[1285] The server stores the received information in a database and records it as "annual income of 6 million yen," "spouse and two children," "life insurance insured," and "mortgage held."

[1286] Step 2: Retrieving and updating external information

[1287] 1. The server periodically contacts external sources such as financial institutions and insurance companies to obtain up-to-date information such as current insurance rates, loan interest rates, and economic news.

[1288] 2. The server stores the acquired information in a database and matches it with the user's information.

[1289] Input: Latest financial information obtained from external sources

[1290] Data processing: The latest financial information is stored in a database and compared with user information.

[1291] Output: Results saved to the database and results of matching with user information

[1292] Specific behavior:

[1293] Suppose the latest mortgage interest rate the server retrieves from the financial institution's API is 1.2% per annum.

[1294] The server stores this interest rate information in a database and compares it with the user's current loan interest rate.

[1295] Step 3: Analyze user information

[1296] 1. The server performs analysis using specific algorithms (such as linear regression or decision trees) based on the user's basic information and external information stored in the database.

[1297] 2. The server generates advice such as reviewing financial products that are best suited to the user's situation and tax-saving measures.

[1298] Input: User's basic information, external information

[1299] Data arithmetic: Analyzing data using algorithms

[1300] Output: Generated advice

[1301] Specific behavior:

[1302] The server takes into consideration the user's annual income of 6 million yen, family composition, life insurance policies, and the latest mortgage interest rate of 1.2%, and generates advice recommending additional life insurance.

[1303] Step 4: Providing advice

[1304] 1. The terminal notifies the user of the advice received from the server.

[1305] 2. The user checks the notification and sees the detailed advice on the device screen.

[1306] Input: Server-generated advice

[1307] Data processing: Notify the user of the advice content

[1308] Output: Notification to the user

[1309] Specific behavior:

[1310] A notification will appear on the user's device saying, "We recommend you take out additional life insurance," and when they click on the details, specific insurance plans and benefits will be displayed.

[1311] Step 5: Collect and learn from conversation logs

[1312] 1. The server records all interactions between the user and the system.

[1313] 2. The server periodically analyzes this data as training data for machine learning algorithms.

[1314] 3. The server trains new machine learning models to improve the accuracy of advice.

[1315] Input: Dialogue history

[1316] Data processing: Analyze dialogue history as training data

[1317] Output: The training results of the new machine learning model

[1318] Specific behavior:

[1319] If a user asks the system, "Please tell me more about the insurance plan," this conversation history is saved on the server and used to generate future advice.

[1320] Step 6: Deploy and apply the model

[1321] 1. The server trains a new generative AI model and deploys it to the system.

[1322] 2. The server applies the new model to future advice generation.

[1323] Input: A new generative AI model

[1324] Data processing: Deploy the new model to the system

[1325] Output: Apply the new model to future advice generation

[1326] Specific behavior:

[1327] The server learns a new generative AI model to improve the accuracy of life insurance advice, and uses this model for future advice generation.

[1328] (Application example 1)

[1329] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1330] Today's consumers often find it difficult to choose from a complex range of financial products and services. Finding the best financial plan for their individual circumstances and keeping up with constantly changing financial information can be a particularly challenging task. Furthermore, there is a lack of tools to analyze users' spending patterns and income information in real time and provide appropriate advice, making it difficult to manage assets efficiently.

[1331] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1332] In this invention, the server includes: means for learning information on an individual's annual income, family composition, and insurance and loan enrollments; means for obtaining information on various insurance and loan interest rates that changes daily from external sources and updating the database; means for generating and providing optimal insurance, loan review proposals, and tax-saving advice to the user; means for proposing optimal financial products based on family composition and years of service and supporting the procedures; means for recording a dialogue log with the user and continuously training an AI model to improve the accuracy of the advice; means for analyzing the user's spending patterns and income information and providing financial advice and savings proposals in real time; and means for obtaining the latest financial data from external sources, calculating income and expenditure balances, and generating advice. This enables users to easily select optimal financial products and services from complex financial products and realizes efficient asset management.

[1333] "Annual income" is the total amount of income an individual receives in a year.

[1334] "Family composition" refers to the composition of members in the household to which an individual belongs, and includes, for example, information on whether or not the individual has a spouse and children.

[1335] "Insurance" generally refers to financial products that individuals and companies contract to reduce risk, and includes life insurance, medical insurance, and automobile insurance.

[1336] A "loan" is a financial product that individuals or companies borrow from financial institutions and repay with a certain interest rate.

[1337] "External sources" refers to information sources provided by third parties such as financial institutions and insurance companies.

[1338] A "database" is a software system for efficiently storing, managing, and retrieving information.

[1339] "Advice" refers to specific advice or suggestions given to an individual.

[1340] "Spending patterns" are information that shows the trends and distribution of how individuals spend their money.

[1341] "Income information" refers to data about an individual's source and amount of income.

[1342] "Real-time" refers to the temporal characteristics of processing data and providing results almost immediately.

[1343] An "AI model" is an algorithm or program that uses machine learning or artificial intelligence to perform a specific task.

[1344] A "prompt" is an instruction or question that is input to a generative AI model and influences the output that the model generates.

[1345] MODE FOR CARRYING OUT THE INVENTION

[1346] This invention is a system that functions as a personal financial planner (FP) for each user, analyzing the user's income and expenditure information in real time and providing optimal financial advice based on that information. This system is implemented using the following hardware and software.

[1347] Collecting and updating information

[1348] Collecting basic information

[1349] Users use a dedicated application to enter information such as annual income, family composition, insurance and loan enrollment. This information is received by a device such as a smartphone and sent to a server, which stores the information in a database and checks for consistency and format.

[1350] Retrieving and updating external information

[1351] The server periodically collects information from external sources such as financial institutions and insurance companies, such as the latest insurance rates, loan interest rates, and economic news, which is then stored in a database and matched with the user's information.

[1352] Generating Advice

[1353] User information analysis

[1354] The server uses an algorithm to analyze the user information stored in the database and external information it has acquired, and generates advice on the best financial products for the user's current situation and tax-saving strategies.

[1355] Providing advice

[1356] The device provides advice and notifies the user. For example, if mortgage interest rates fall, the server generates a new loan offer and the device sends a notification to the user. When the user confirms the notification, the specific advice details are displayed.

[1357] Continuous learning and improvement

[1358] Collecting and learning from conversation logs

[1359] All interactions between users and the system are recorded and stored on a server. These interaction logs are used as training data for machine learning algorithms. The server periodically analyzes this data and learns new models to improve the accuracy of advice.

[1360] Deploying and applying the model

[1361] Once a new machine learning model is learned, the server deploys it and applies it to future advice generation, improving the accuracy of advice to users and providing more appropriate financial support.

[1362] Specific examples

[1363] Scenario 1: New life insurance proposal

[1364] The user reports the birth of a new child to the system. The device sends this information to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and presents the proposal.

[1365] Scenario 2: Loan Restructuring

[1366] The server retrieves the latest loan interest rate information from an external source and compares it with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan refinancing plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[1367] Prompt Sentence Examples

[1368] We collected information such as the user's income of 5 million yen, monthly expenses of 150,000 yen, 50,000 yen, and 30,000 yen, as well as family composition. The latest loan interest rate was obtained from an API as 4.0%. Analyze the user's income and expenditure balance and generate optimal financial advice based on an interest rate lower than the current loan interest rate of 4.5%.

[1369] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1370] Step 1:

[1371] The user uses a dedicated application to input basic information (annual income, family composition, insurance and loan information). This is the input information. The device sends this information to the server, which stores the received information in a database and performs formal checks. Through this process, the user's basic information is stored in the database.

[1372] Step 2:

[1373] The server periodically retrieves information such as the latest insurance rates, loan interest rates, and economic news from external sources (financial institutions and insurance companies). This information becomes the input of external financial data. The server stores the retrieved information in a database and compares it with the user's existing data. This updates the database with the latest financial information.

[1374] Step 3:

[1375] The server performs an algorithmic analysis based on user information and external financial information. At this time, the user's income and expenditure balance, insurance, and loan status are calculated. The input is the user's basic information and external financial information, and the output is the generated advice. Specifically, the system compares the user's income and expenses and creates a proposal for reviewing the optimal financial product based on fluctuations in loan interest rates.

[1376] Step 4:

[1377] The terminal receives advice from the server and provides it to the user in the form of a notification. At this time, the user is shown specific details of the advice. For example, the device may notify the user that mortgage interest rates have dropped and propose a new loan plan. It may also propose a life insurance plan that reflects changes in the user's family structure. This allows the user to receive the latest financial advice in real time.

[1378] Step 5:

[1379] All interactions between the user and the system are recorded and stored on the server. This adds the interaction log to a database. This interaction log serves as training data for future advice generation processes. The server uses machine learning algorithms to analyze this data and improve the accuracy of the advice.

[1380] Step 6:

[1381] Once a new machine learning model is trained, the server deploys it in real time and applies it to the next and subsequent advice generation. This process enables the system to always use the latest information and learning models to provide users with optimal financial advice.

[1382] Step 7:

[1383] For example, when a user reports the birth of a new child to the system, the device sends this information to the server. The server updates the database with the change in family structure and generates a new life insurance plan. The plan is then notified to the user via the device, and the proposed plan is presented in detail.

[1384] Prompt Sentence Examples

[1385] We collected information such as the user's income of 5 million yen, monthly expenses of 150,000 yen, 50,000 yen, and 30,000 yen, as well as family composition. The latest loan interest rate was obtained from an API as 4.0%. Analyze the user's income and expenditure balance and generate optimal financial advice based on an interest rate lower than the current loan interest rate of 4.5%.

[1386] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1387] This invention provides more accurate advice and services by combining a system that functions as a personal financial planner (FP) with an emotion engine that recognizes the user's emotions. This system collects information on an individual's annual income, family composition, and insurance and loan enrollment, and not only provides optimal advice based on financial information that changes daily, but also recognizes the user's emotions and takes an approach based on that emotional data. The program processing of this system is described in detail below in natural language.

[1388] Collecting and updating information

[1389] Collecting basic information

[1390] A user accesses a dedicated application or web interface and enters information about their annual income, family composition, insurance and loans. This information is received by the device and sent to a server, which stores it in a database and checks for consistency and format.

[1391] Retrieving and updating external information

[1392] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news from financial institutions and insurance companies through their APIs. This information is stored in a database and matched with user information.

[1393] Collecting emotional information

[1394] When the device interacts with the user, it activates the emotion engine and collects dialogue logs, voice data, and facial expression data, which are then sent to a server for emotion analysis.

[1395] Generating Advice

[1396] User information analysis

[1397] The server inputs the user information stored in the database and external information into an analytical algorithm, which generates advice on the optimal insurance, loan review, and tax saving measures for the user's current situation.

[1398] Emotional information analysis

[1399] The server analyzes the collected emotional data to recognize the user's current emotional state, and adjusts the content and presentation of advice based on the user's emotional state.

[1400] Generating and saving advice

[1401] The server saves the generated advice in a database as an individual plan, converts the generated advice into JSON format, etc., and sends it to the device.

[1402] Providing advice

[1403] The device sends an advice notification to the user informing them that new advice is available, and the user acknowledges the notification and opens the application or web interface to view the detailed advice.

[1404] Continuous learning and improvement

[1405] Collection and storage of conversation logs

[1406] All interactions between users and the system are recorded and stored on a server, and these interaction logs are used as training data for machine learning algorithms.

[1407] Model training

[1408] The server periodically analyzes the dialogue logs and uses machine learning algorithms to train new models, including emotional data, to improve advice accuracy and user satisfaction.

[1409] Deploying and applying the model

[1410] Once a new machine learning model is learned, the server deploys it and applies it to future advice generation, improving the accuracy of advice to users and providing more appropriate financial support.

[1411] Specific examples

[1412] Example 1: Proposing a new life insurance policy

[1413] scenario

[1414] The user reports the birth of a new child to the system. The device sends this information to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and asks them to confirm the proposal.

[1415] Use of emotion engine

[1416] When a user reviews a new suggestion, the device activates an emotion engine to analyze the user's reaction. For example, if the user is feeling stressed, the server will provide customized advice to reduce stress.

[1417] Example 2: Loan review

[1418] scenario

[1419] The server retrieves the latest loan interest rates from external sources and compares them with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[1420] Use of emotion engine

[1421] When a user receives a notification, the emotion engine analyzes the user's facial expression and tone of voice. If the user indicates anxiety, the server provides additional information to provide further explanation or reassurance.

[1422] In this way, the system continuously collects and updates user information and provides optimal advice, optimizing the user's financial situation and providing personalized support that takes into account their emotional state.

[1423] The processing flow will be explained below.

[1424] Collecting and updating information

[1425] Step 1: Initial input of user information

[1426] The user logs into a dedicated application or web interface and enters information about their annual income, family composition, and insurance and loan enrollment.

[1427] The terminal receives the input information and transmits it to the server.

[1428] Step 2: Save user information

[1429] The server stores the received user information in a database.

[1430] The server checks the information for consistency and format validity.

[1431] Step 3: Obtaining external information

[1432] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies.

[1433] The server stores this information in a database and matches it with the user's information.

[1434] Generating Advice

[1435] Step 4: Analyzing User Information

[1436] The server inputs user information stored in a database and external information into an analysis algorithm.

[1437] The server generates advice on the optimal insurance, loan review, and tax saving measures for the user's current situation.

[1438] Step 5: Collecting emotional information

[1439] When the device interacts with the user, it activates an emotion engine and collects dialogue logs, voice data, and facial expression data.

[1440] The device transmits the collected emotion data to a server.

[1441] Step 6: Emotional Analysis

[1442] The server analyzes the collected emotional data to recognize the user's current emotional state.

[1443] The server adjusts the content and presentation of advice based on the emotional state.

[1444] Step 7: Generate and save advice

[1445] The server stores the generated advice in a database as a separate plan.

[1446] The server converts the generated advice into JSON format or similar and sends it to the terminal.

[1447] Step 8: User notification and confirmation

[1448] The terminal sends an advice notification to the user, informing them that new advice is available.

[1449] The user acknowledges the notification and opens the application or web interface to view detailed advice.

[1450] Continuous learning and improvement

[1451] Step 9: Collect and store conversation logs

[1452] The server records all user interaction logs and stores them in a database.

[1453] Step 10: Train the model

[1454] The server periodically analyzes the dialogue logs and uses machine learning algorithms to train new models.

[1455] The server also learns from emotional data, improving the accuracy of advice and user satisfaction.

[1456] Step 11: Deploy and apply the model

[1457] The server deploys the newly trained model and applies it to future advice generation.

[1458] The server will provide better financial support based on the new model.

[1459] Specific examples

[1460] Example 1: Proposing a new life insurance policy

[1461] Step 1: Enter changes to family composition

[1462] A user reports the birth of a new child to the system.

[1463] The terminal receives this information and sends it to the server.

[1464] Step 2: Update the database

[1465] The server stores the new family structure information in the database.

[1466] The server compares the insurance coverage with existing insurance coverage and evaluates the need for new insurance.

[1467] Step 3: Collecting emotion data

[1468] The device activates the emotion engine when a user reports something and collects dialogue logs and voice data.

[1469] The device transmits the emotion data to the server.

[1470] Step 4: Analyze the sentiment data

[1471] The server analyzes the collected emotional data to understand how the user feels about the birth of the new child.

[1472] The server sets the content of advice in an approach that matches the user's emotions.

[1473] Step 5: Generate and notify advice

[1474] The server generates the optimal life insurance plan and sends it to the terminal.

[1475] The device will send a notification to the user prompting them to confirm the suggestion.

[1476] Example 2: Loan review

[1477] Step 1: Obtaining external information

[1478] The server retrieves the latest loan interest rate information from an external source.

[1479] The server stores this information in a database.

[1480] Step 2: Reevaluate the loan

[1481] The server compares the new interest rate information with the user's current loan terms.

[1482] If the server finds favorable terms, it generates a new loan plan.

[1483] Step 3: Collecting emotion data

[1484] The device activates an emotion engine when the user interacts with it, and collects voice data and facial expression data.

[1485] The device transmits the emotion data to the server.

[1486] Step 4: Analyze the sentiment data

[1487] The server analyzes the emotional data and determines the emotional state of the user when they receive the notification.

[1488] The server provides additional information to allay the user's concerns and doubts.

[1489] Step 5: Generate and notify advice

[1490] The server transmits the generated loan plan to the terminal.

[1491] The device will then send a notification to the user prompting them to confirm the new loan offer.

[1492] This allows users to always receive the latest and most appropriate financial advice, and also enjoy personalized support that takes into account their emotional state.

[1493] Example 2

[1494] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1495] In modern society, personal financial information is diverse, making it difficult to obtain accurate advice tailored to daily changing market information and individual circumstances. Furthermore, there is a lack of personalized support that takes into account users' emotions and stress levels, resulting in reduced user satisfaction and convenience. Furthermore, mechanisms for continuous learning and model improvement to improve the accuracy of financial advice have not been effectively implemented. There is a need for a system that can resolve these issues and provide more accurate financial advice.

[1496] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1497] In this invention, the server includes: means for inputting information on an individual's annual income, family composition, and insurance and loan enrollments; means for transmitting and storing the information input by the user to a database; means for acquiring information on various insurance and loan interest rates from external sources and periodically updating the database; means for collecting emotional information such as a dialogue log, voice data, and facial expression data using an emotion engine; means for generating optimal insurance and loan review proposals and tax-saving advice based on the collected user information and emotional data; means for storing the generated advice in a database and transmitting it to a terminal; means for adjusting the content and expression of the advice based on the emotional data and providing personalized information to the user; means for sending new advice and financial product plan change notifications to the user; means for recording a dialogue log with the user, periodically training a machine learning algorithm to improve the accuracy of the advice; and means for deploying a new machine learning model and applying it thereafter. This enables the provision of highly accurate personalized advice tailored to an individual's financial situation and emotional state.

[1498] "Individual annual income" is the total income earned by a user in a year.

[1499] "Family structure" is information that indicates the number of members in the user's household and their relationships.

[1500] "Insurance" refers to insurance contracts such as life insurance, medical insurance, and non-life insurance that the user has subscribed to.

[1501] A "loan" is a mortgage, car loan, personal loan, or other debt agreement taken out by a user.

[1502] A "database" is a part of a computer system where collected information is stored and managed.

[1503] "External Source" is an institution or data source that provides information from outside the system, such as a financial institution or insurance company.

[1504] An "emotion engine" is software or a system that analyzes dialogue logs, voice data, and facial expression data to identify a user's emotional state.

[1505] A "dialogue log" is data that records the content of conversations between a user and a system.

[1506] A "machine learning algorithm" is an algorithm that analyzes large amounts of data, recognizes patterns, and makes future predictions and classifications.

[1507] An "AI model" is a mathematical model built on machine learning algorithms for making predictions and classifications.

[1508] "Personalized information" is information or advice that is customized based on a user's individual situation or emotional state.

[1509] A "notification" is a message that informs the user that new information or advice is available.

[1510] A "Plan Change Notification" is a message informing a user of a proposed change to their financial product or insurance plan.

[1511] "Means" refers to the functions or processes used to achieve a particular purpose.

[1512] Collecting basic information

[1513] The user accesses a dedicated application or web interface and enters information such as annual income, family composition, and insurance and loan enrollment. The device receives this information and temporarily stores it. The device then sends the saved user information to a server, which stores it in a database and checks its consistency and format. This ensures that the user information is accurately registered in the database.

[1514] Retrieving and updating external information

[1515] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies. The server stores the information obtained from external sources in a database and compares it with user information. This ensures that the database always contains up-to-date information, forming the basis for generating accurate advice.

[1516] Collecting and analyzing emotional information

[1517] When a user begins interacting with the system, the device activates the emotion engine. The device collects emotional information, such as dialogue logs, voice data, and facial expression data, in real time and sends it to the server. The server then analyzes the received emotional data using a machine learning algorithm to identify the user's current emotional state. This allows the system to provide optimal advice based on the user's emotional state.

[1518] Generating and Serving Advice

[1519] The server uses analytical algorithms to generate optimal financial advice based on user information in the database and external information. The generated advice is stored in the database, converted to JSON format, etc., and sent to the device. The device sends an advice notification to the user, informing them that new advice is available. The user confirms the notification and views the detailed advice in the application or web interface.

[1520] Continuous training and model deployment

[1521] The server records dialogue logs with the user and stores them in a database. The dialogue logs and emotion data are periodically analyzed, and a new model is trained using a machine learning algorithm. The new trained model is deployed and applied to future advice generation. This improves advice accuracy and user satisfaction.

[1522] Specific examples

[1523] Example 1: New life insurance proposal

[1524] The user reports the birth of a new child to the system via a web interface. The device sends this information to the server, which updates the database with the change in family structure. The server generates a new life insurance plan taking the change in family structure into account. The device notifies the user of the new insurance plan proposal and asks them to confirm the proposal. As the user confirms the proposal, the device activates an emotion engine to analyze the user's reaction in real time. The server provides customized advice to reduce the user's stress.

[1525] Example 2: Loan Restructuring

[1526] The server retrieves the latest loan interest rate information from an external source and compares it with the user's current loan interest rate. If there is a favorable change in interest rate, the server generates a new loan plan and the device notifies the user. The user receives the notification, reviews the proposal, and decides whether to proceed. The device activates an emotion engine to analyze the user's facial expressions and tone of voice. If the user shows signs of anxiety, the server provides additional information to provide further explanation or reassurance.

[1527] Example prompt sentence:

[1528] "Please let us know about the birth of your new child and provide appropriate life insurance recommendations."

[1529] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1530] Step 1: Gather basic information

[1531] User initiated

[1532] Users access a dedicated application or web interface and enter information about their annual income, family composition, and insurance and loan coverage.

[1533] Input: Annual income, family composition, insurance, and loan information entered by the user.

[1534] Output: User information received by the device.

[1535] Terminal Processing

[1536] The device receives the information entered by the user and temporarily stores it.

[1537] Input: Information entered by the user.

[1538] Output: Data in a format that can be sent to the server.

[1539] Server Processing

[1540] The device sends the stored user information to the server, which stores this information in a database and checks its consistency and format.

[1541] Input: User information received from the device.

[1542] Output: User information stored in a database, data checked for consistency and format.

[1543] Step 2: Retrieving and updating external information

[1544] Server activation

[1545] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies.

[1546] Input: API request.

[1547] Output: Financial information obtained from external sources.

[1548] Server Processing

[1549] The server stores the information obtained from external sources in a database and matches it with user information.

[1550] Input: Financial information from external sources.

[1551] Output: The external information updated in the database.

[1552] Step 3: Collecting and analyzing emotional information

[1553] User initiated

[1554] When the user starts interacting with the system, the device activates the emotion engine.

[1555] Input: User interaction begins.

[1556] Output: Launch of emotion engine.

[1557] Terminal Processing

[1558] The device collects emotional information such as dialogue logs, voice data, and facial expression data in real time and sends it to the server.

[1559] Input: Dialogue logs, voice data, facial expression data.

[1560] Output: Emotion data sent to the server.

[1561] Server Processing

[1562] The server analyzes the received emotional data using machine learning algorithms to identify the user's current emotional state.

[1563] Input: Emotion data received from the device.

[1564] Output: Parsed emotional state data.

[1565] Step 4: Generating and serving advice

[1566] Server activation

[1567] The server uses analytical algorithms to generate optimal financial advice based on user information in the database and external information.

[1568] Input: User information in the database and external information.

[1569] Output: The generated financial advice.

[1570] Server Processing

[1571] The generated advice is stored in a database, converted to JSON format, etc., and sent to the terminal.

[1572] Input: Generated financial advice.

[1573] Output: Advice data in a format that can be sent to a terminal.

[1574] Terminal activation

[1575] The device sends an advice notification to the user informing them that new advice is available, and the user can view the notification and the detailed advice in the application or web interface.

[1576] Input: Advice data from the server.

[1577] Output: Notify the user and provide further advice.

[1578] Step 5: Continue training and deploy the model

[1579] Server activation

[1580] The server records the user interaction log and stores it in a database.

[1581] Input: User interaction log.

[1582] Output: Interaction logs stored in a database.

[1583] Server Processing

[1584] Dialogue logs and sentiment data are analyzed periodically, and new models are trained using machine learning algorithms.

[1585] Input: Dialogue logs and emotion data.

[1586] Output: A new trained machine learning model.

[1587] Server Processing

[1588] The new trained model is deployed and applied to future advice generation.

[1589] Input: A trained machine learning model.

[1590] Output: The new deployed machine learning model.

[1591] (Application example 2)

[1592] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[1593] Conventional financial planner (FP) systems provide optimal advice based on a user's annual income, family structure, insurance, and loan information. However, because they do not take the user's emotional state into account, the receptivity and degree of personalization of the advice is insufficient. This poses a challenge in that they are unable to respond appropriately in situations where the user feels anxious or stressed. Furthermore, there is a need for a system that can quickly respond to changes in the user's daily consumption behavior and emotions.

[1594] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1595] In this invention, the server includes means for learning information on an individual's annual income, family composition, and insurance and loan enrollments, means for acquiring daily changing economic information from external sources and updating the database, means for generating and providing optimal insurance and loan review proposals and tax saving advice to the user, means for analyzing the user's facial expression data and voice data and recognizing the emotional state, means for adjusting the content and presentation of the advice based on the user's emotional data, and means for recording a dialogue log with the user and continuously training the AI ​​model to improve the accuracy of the advice. This enables personalized, highly accurate advice that takes into account both the user's economic situation and emotional state.

[1596] "Individual annual income" refers to the total income earned by a particular individual over a certain period of time.

[1597] "Family structure" refers to the composition of members and their relationships within a particular household.

[1598] "Insurance" is a contract that provides financial compensation for risks to persons or property.

[1599] A "loan" is a financial contract in which you promise to repay a borrowed amount under certain conditions.

[1600] "Means of learning" refers to the function of collecting, analyzing, and memorizing information.

[1601] "External source" refers to an information source provided from outside the system.

[1602] "Means for updating the database" refers to a function for updating existing information to the latest information.

[1603] "Optimal insurance and loan review proposals" refer to proposals for changes to insurance and loans that are most suitable for the user's current situation.

[1604] "Tax advice" means instructions or suggestions for minimizing your tax burden.

[1605] "Means of providing" refers to the function for delivering information and advice to users.

[1606] "Financial products" are products traded on the market, including stocks, bonds, investment trusts, etc.

[1607] "Means to support procedures" refers to support functions that allow users to smoothly carry out the necessary procedures.

[1608] An "interaction log" is a history of interactions between a user and a system.

[1609] An "AI model" is a data analysis model based on artificial intelligence.

[1610] "Means for recognizing emotional states" refers to a function that analyzes and recognizes emotions from input data such as the user's facial expressions and voice.

[1611] "Facial expression data and voice data" refers to digital information related to the user's facial expressions and voice.

[1612] "Means for adjusting the content and presentation of advice based on emotional data" refers to a function for changing the content and presentation of advice provided according to the user's emotional state.

[1613] This invention is a system that combines a personal financial planner (FP) system with an emotion recognition engine to provide highly accurate advice to users. Specifically, the system is configured using the following hardware and software:

[1614] System Components

[1615] 1. User data collection

[1616] The device (smartphone, tablet, smart glasses, etc.) provides an interface to collect information about the user's annual income, family composition, insurance, and loans. This information is sent to a server and stored in a database.

[1617] 2. Obtaining information from external sources

[1618] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies. This information is also stored in a database and compared with user information.

[1619] 3. Collecting emotional information

[1620] The device's built-in camera and microphone are used to collect the user's facial expression and voice data, which is then analyzed by an emotion recognition engine to recognize the user's emotional state.

[1621] 4. Generating Advice

[1622] The server uses user information stored in a database, external information, and emotional data to generate optimal advice, including insurance and loan restructuring and tax-saving strategies. It also takes into account the user's emotional state and adjusts the content and presentation of the advice.

[1623] 5. Providing Advice and Notification

[1624] The generated advice is converted to JSON format or similar and sent to the terminal. The terminal then sends an advice notification to the user informing them that new advice is available. The user can then confirm the notification and open the interface to view the detailed advice.

[1625] 6. Continuous learning and improvement

[1626] All user interaction logs are recorded and stored on the server. These interaction logs are used by the machine learning algorithm to learn new models and improve the accuracy of advice. Once a new model is learned, the server deploys it and applies it to future advice generation.

[1627] Specific examples

[1628] Example 1: Proposing a new life insurance policy

[1629] The user reports the birth of a new child to the system. This information is sent to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and asks them to confirm the proposal.

[1630] Use of Emotion Recognition

[1631] When a user confirms a new suggestion, the device activates an emotion recognition engine to analyze the user's reaction. For example, if the user is feeling stressed, the server will provide customized advice to reduce stress.

[1632] Example 2: Loan review

[1633] The server retrieves the latest loan interest rates from external sources and compares them with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[1634] Use of Emotion Recognition

[1635] When a user receives a notification, an emotion recognition engine analyzes the user's facial expressions and tone of voice. If the user indicates anxiety, the server provides additional information to provide further explanation or reassurance.

[1636] Prompt Sentence Examples

[1637] "Give me an example of a user opening a finance app and checking their latest finances and sentiment data."

[1638] "Give me an example of a system where your personal financial planner provides advice based on today's spending behavior and sentiment data."

[1639] In this way, the system enables personalized, highly accurate advice that takes into account both the user's financial situation and emotional state.

[1640] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1641] Step 1:

[1642] Collection of User Information

[1643] Input: User's annual income, family composition, insurance and loan information.

[1644] How it works: The user opens a dedicated application on a device such as a smartphone or tablet and enters the necessary information, which is then sent from the device to the server.

[1645] Output: The user information sent to the server.

[1646] Step 2:

[1647] Obtaining external information

[1648] Input: The API endpoint of the financial institution or insurance company.

[1649] How it works: The server periodically calls the API to retrieve information such as the latest insurance rates, loan interest rates, and economic news.

[1650] Output: A database containing up-to-date financial information.

[1651] Step 3:

[1652] Collecting emotional information

[1653] Input: User's facial expression data and voice data.

[1654] How it works: When a user uses the device, the camera and microphone are activated to collect the necessary emotional data. The emotion recognition engine analyzes this data to recognize the user's emotional state.

[1655] Output: Emotion data sent to the server.

[1656] Step 4:

[1657] Data analysis and advice generation

[1658] Input: User information, external financial information, sentiment data.

[1659] How it works: The server feeds all the information stored in the database into an analytical algorithm to generate advice on the best insurance, loan review, tax savings, etc. The content and presentation of the advice is adjusted based on the emotional data.

[1660] Output: The generated advice plan.

[1661] Step 5:

[1662] Notification and provision of advice

[1663] Input: The generated advice plan.

[1664] How it works: The server converts the advice plan into a format such as JSON and sends it to the device. The device then sends an advice notification to the user informing them that new advice is available. The user confirms the notification and opens the application to view the detailed advice.

[1665] Output: Advice sent to the user's device.

[1666] Step 6:

[1667] Interaction logging and continuous learning

[1668] Input: Log of interactions between the user and the system.

[1669] How it works: All user interactions with the system are logged and stored on a server. These logs are used as training data for machine learning algorithms to learn new AI models. Once a new model is learned, the server deploys it and applies it to future advice generation.

[1670] Output: The updated AI model.

[1671] This series of processing steps allows the system to take into account both the user's financial situation and emotional state and provide highly accurate and personalized advice.

[1672] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[1673] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1674] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[1675] [Fourth embodiment]

[1676] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1677] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[1678] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[1679] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[1680] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[1681] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[1682] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[1683] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[1684] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[1685] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[1686] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[1687] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[1688] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1689] This invention is a system that functions as a personal financial planner (FP) for each user. This system collects information on an individual's annual income, family structure, insurance and loan enrollment, and provides optimal advice based on financial information that changes daily. Below, the program processing of this system is described in detail in natural language.

[1690] Collecting and updating information

[1691] Collecting basic information

[1692] A user accesses a dedicated application or web interface and enters information about their annual income, family composition, insurance and loans. This information is received by the device and sent to a server, which stores it in a database and checks for consistency and format.

[1693] Retrieving and updating external information

[1694] The server periodically collects the latest information from external sources such as financial institutions and insurance companies, such as current insurance rates, loan interest rates, and economic news. This information is stored in a database and matched with the user's information.

[1695] Generating Advice

[1696] User information analysis

[1697] The server uses an algorithm to analyze the user information stored in the database and external information it has acquired, and generates advice on the best financial products for the user's current situation and tax-saving strategies.

[1698] Providing advice

[1699] The device provides advice and notifies the user. For example, if mortgage interest rates fall, the server generates a new loan offer and the device sends a notification to the user. When the user confirms the notification, the specific advice details are displayed.

[1700] Continuous learning and improvement

[1701] Collecting and learning from conversation logs

[1702] All interactions between users and the system are recorded and stored on a server. These interaction logs are used as training data for machine learning algorithms. The server periodically analyzes this data and learns new models to improve the accuracy of advice.

[1703] Deploying and applying the model

[1704] Once a new machine learning model is learned, the server deploys it and applies it to future advice generation, improving the accuracy of advice to users and providing more appropriate financial support.

[1705] Specific examples

[1706] Scenario 1: New life insurance proposal

[1707] The user reports the birth of a new child to the system. The device sends this information to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and presents the proposal.

[1708] Scenario 2: Loan Restructuring

[1709] The server retrieves the latest loan interest rate information from an external source and compares it with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan refinancing plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[1710] In this way, the system continuously collects and updates user information and provides optimal advice to optimize the user's financial situation, allowing users to easily select and use the financial products and services that are best suited to them.

[1711] The processing flow will be explained below.

[1712] Collecting and updating information

[1713] Step 1: Initial input of user information

[1714] The user logs into a dedicated application or web interface and enters information about their annual income, family composition, and insurance and loan enrollment.

[1715] The terminal receives the input information and transmits it to the server.

[1716] Step 2: Save user information

[1717] The server stores the received user information in a database.

[1718] The server checks the information for consistency and format validity.

[1719] Step 3: Obtaining external information

[1720] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies.

[1721] The server stores this information in a database and matches it with the user's information.

[1722] Generating Advice

[1723] Step 4: Analyzing User Information

[1724] The server inputs user information stored in a database and external information into an analysis algorithm.

[1725] The server generates optimal insurance, loan review, and tax saving advice based on the user's financial situation.

[1726] Step 5: Generate and save advice

[1727] The server stores the generated advice in a database as a separate plan.

[1728] The server converts the generated advice into JSON format or similar and sends it to the terminal.

[1729] Step 6: User notification and confirmation

[1730] The terminal sends an advice notification to the user, informing them that new advice is available.

[1731] The user acknowledges the notification and opens the application or web interface to view detailed advice.

[1732] Continuous learning and improvement

[1733] Step 7: Collect conversation logs

[1734] The server records all user interaction logs and stores them in a database.

[1735] Step 8: Training the model

[1736] The server periodically analyzes the dialogue logs and uses machine learning algorithms to train new models.

[1737] Step 9: Deploy the improved model

[1738] The server deploys the newly trained model and uses it for future advice generation.

[1739] Specific examples

[1740] Example 1: Proposing a new life insurance policy

[1741] Step 1: The user enters that a new child has been born.

[1742] The user uses the application to enter changes to their household.

[1743] The terminal receives this information and sends it to the server.

[1744] Step 2: Update the database

[1745] The server stores the new family structure information in the database.

[1746] The server compares the insurance coverage with existing insurance coverage and evaluates the need for new insurance.

[1747] Step 3: Generate and notify advice

[1748] The server generates the optimal life insurance plan and sends it to the terminal.

[1749] The device will send a notification to the user prompting them to confirm the suggestion.

[1750] Example 2: Loan review

[1751] Step 1: The server obtains external information.

[1752] The server retrieves the latest loan interest rate information from an external source.

[1753] The server stores this information in a database.

[1754] Step 2: Reevaluate the loan

[1755] The server compares the new interest rate information with the user's current loan terms.

[1756] If the server finds favorable terms, it generates a new loan plan.

[1757] Step 3: Generate and notify advice

[1758] The server transmits the generated loan plan to the terminal.

[1759] The device will then send a notification to the user prompting them to confirm the new loan offer.

[1760] This allows users to always receive the latest and most appropriate financial advice, optimizing their financial situation.

[1761] Example 1

[1762] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1763] Currently, many individuals struggle with selecting and managing complex financial products. It is particularly challenging to centrally manage information such as annual income, family composition, insurance and loan enrollment, and appropriately handle financial information that changes daily. Furthermore, to propose optimal financial products based on users' life events and provide prompt and accurate advice, highly accurate support utilizing dialogue history analysis and machine learning is required. A system that solves these challenges and enables users to efficiently select and manage optimal financial products is needed.

[1764] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[1765] In this invention, the server includes a means for learning information on an individual's income, family structure, and insurance and loan enrollment, a means for acquiring information on various insurance and loan interest rates that change daily from external sources and updating the database, and a means for generating and providing optimal insurance and loan review proposals and tax saving advice to the user, thereby enabling the user to efficiently select and review optimal financial products based on the latest financial information.

[1766] "Income" is all the economic benefits that an individual or household receives over a given period of time.

[1767] "Family structure" refers to information about the number of family members who make up a household and their relationships.

[1768] "Insurance" is a financial product in which a fixed insurance premium is paid based on a contract, and insurance benefits can be received in the event of an accident or disaster.

[1769] A "loan" is a contract to borrow money from a financial institution on the condition that it will be repaid over a certain period of time.

[1770] "External sources" refer to information providers and information sources that exist outside the system, including financial institutions, insurance companies, and economic news.

[1771] A "database" is a system that efficiently stores and manages collected data and allows it to be searched and updated as needed.

[1772] An "artificial intelligence model" is a model that is built based on machine learning algorithms and makes decisions and predictions based on data.

[1773] "Life events" are events that have a significant impact on an individual's life, such as marriage, childbirth, transfer, and retirement.

[1774] "Financial products" are products such as insurance, loans, stocks, bonds, and investment trusts offered by financial institutions, which are intended for asset management and risk hedging.

[1775] "Dialogue history" is a record of communication between the user and the system, and is used to improve services in the future.

[1776] MODE FOR CARRYING OUT THE INVENTION

[1777] This invention is a system that functions as a personal financial planner (FP) for each user. The system collects information on an individual's income, family structure, insurance and loan enrollment, and provides optimal advice based on financial information that changes daily.

[1778] Collecting and updating information

[1779] Collecting basic information

[1780] A user accesses a dedicated application or web interface and enters information about their income, family composition, insurance and loans. This information is received by the device and sent to a server, which stores it in a database and checks for consistency and format.

[1781] Examples of hardware and software used:

[1782] Hardware: User's personal computer, smartphone

[1783] Software: web browsers, mobile applications, server-side database management systems (e.g., MySQL)

[1784] Examples:

[1785] A user fills out a web form with information about their income (6 million yen), family composition (spouse and two children), and their life insurance and mortgage. This information is sent to the server and stored in a database.

[1786] Retrieving and updating external information

[1787] The server periodically contacts external sources, such as financial institutions and insurance companies, to obtain up-to-date information on current insurance rates, loan interest rates, economic news, etc. This information is stored in a database and matched with the user's information.

[1788] Examples of hardware and software used:

[1789] Hardware: Server

[1790] Software: External API, data collection scripts (e.g., Python)

[1791] Examples:

[1792] The server retrieves the latest loan interest rate information from the financial institution's API, stores it in a database, and compares it with the user's information.

[1793] Generating Advice

[1794] User information analysis

[1795] The server uses a generative AI model to analyze the user's information stored in the database and external information it has acquired, thereby generating advice on optimal financial products and tax-saving strategies.

[1796] Examples of hardware and software used:

[1797] Hardware: Server

[1798] Software: Data analysis tools (e.g., scikit-learn), generative AI models

[1799] Examples:

[1800] The server generates advice recommending additional life insurance coverage, taking into account the user's annual income of 6 million yen, family composition, and the latest loan interest rate of 1.2%.

[1801] Providing advice

[1802] The device provides advice and notifies the user. For example, if mortgage interest rates fall, the server generates a new loan offer and the device sends a notification to the user. When the user confirms the notification, the specific advice details are displayed.

[1803] Examples of hardware and software used:

[1804] Hardware: User's personal computer, smartphone

[1805] Software: Mobile applications, notification systems (e.g., Firebase Cloud Messaging)

[1806] Examples:

[1807] A notification will appear on the user's device saying, "We recommend you take out additional life insurance," and when they click on the details, specific insurance plans and benefits will be displayed.

[1808] Continuous learning and improvement

[1809] Collecting and learning from conversation logs

[1810] All interactions between the user and the system are recorded and stored on a server. These interaction logs are used as training data for machine learning algorithms. The server analyzes this data and learns new generative AI models to improve the accuracy of advice.

[1811] Examples of hardware and software used:

[1812] Hardware: Server

[1813] Software: Machine learning algorithms (e.g., TensorFlow)

[1814] Examples:

[1815] If a user asks the system, "Please tell me more about the insurance plan," this conversation history is saved on the server and used to generate future advice.

[1816] Deploying and applying the model

[1817] Once a new generative AI model is learned, the server deploys it and applies it to future advice generation.

[1818] Examples:

[1819] The server learns a new generative AI model to improve the accuracy of life insurance advice, and uses this model for future advice generation.

[1820] Prompt Sentence Examples

[1821] Here are some example prompts to input to the generative AI model:

[1822] User: I have a baby and would like to review my life insurance.

[1823] Server: We're offering a new life insurance plan. Here are the details.

[1824] In this way, the information input by the user and the server's response based on that information can be presented in a natural conversational format.

[1825] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1826] Step 1: Gather basic information

[1827] 1. A user accesses an application or web interface and enters information about their income, family status, and insurance and loan coverage.

[1828] 2. The device sends the information entered by the user to the server via the backend API.

[1829] 3. The server stores the received information in a database, checking the consistency and format of the entered data.

[1830] Input: User's income, family structure, insurance information, loan information

[1831] Data processing: Check income, family structure, insurance information, loan information and save it in the database

[1832] Output: Save results to database

[1833] Specific behavior:

[1834] A user fills out a web form with information about their income of 6 million yen, their family composition (spouse and two children), and their life insurance and mortgage insurance policies.

[1835] The device sends the input information to the server via the backend API.

[1836] The server stores the received information in a database and records it as "annual income of 6 million yen," "spouse and two children," "life insurance insured," and "mortgage held."

[1837] Step 2: Retrieving and updating external information

[1838] 1. The server periodically contacts external sources such as financial institutions and insurance companies to obtain up-to-date information such as current insurance rates, loan interest rates, and economic news.

[1839] 2. The server stores the acquired information in a database and matches it with the user's information.

[1840] Input: Latest financial information obtained from external sources

[1841] Data processing: The latest financial information is stored in a database and compared with user information.

[1842] Output: Results saved to the database and results of matching with user information

[1843] Specific behavior:

[1844] Suppose the latest mortgage interest rate the server retrieves from the financial institution's API is 1.2% per annum.

[1845] The server stores this interest rate information in a database and compares it with the user's current loan interest rate.

[1846] Step 3: Analyze user information

[1847] 1. The server performs analysis using specific algorithms (such as linear regression or decision trees) based on the user's basic information and external information stored in the database.

[1848] 2. The server generates advice such as reviewing financial products that are best suited to the user's situation and tax-saving measures.

[1849] Input: User's basic information, external information

[1850] Data arithmetic: Analyzing data using algorithms

[1851] Output: Generated advice

[1852] Specific behavior:

[1853] The server takes into consideration the user's annual income of 6 million yen, family composition, life insurance policies, and the latest mortgage interest rate of 1.2%, and generates advice recommending additional life insurance.

[1854] Step 4: Providing advice

[1855] 1. The terminal notifies the user of the advice received from the server.

[1856] 2. The user checks the notification and sees the detailed advice on the device screen.

[1857] Input: Server-generated advice

[1858] Data processing: Notify the user of the advice content

[1859] Output: Notification to the user

[1860] Specific behavior:

[1861] A notification will appear on the user's device saying, "We recommend you take out additional life insurance," and when they click on the details, specific insurance plans and benefits will be displayed.

[1862] Step 5: Collect and learn from conversation logs

[1863] 1. The server records all interactions between the user and the system.

[1864] 2. The server periodically analyzes this data as training data for machine learning algorithms.

[1865] 3. The server trains new machine learning models to improve the accuracy of advice.

[1866] Input: Dialogue history

[1867] Data processing: Analyze dialogue history as training data

[1868] Output: The training results of the new machine learning model

[1869] Specific behavior:

[1870] If a user asks the system, "Please tell me more about the insurance plan," this conversation history is saved on the server and used to generate future advice.

[1871] Step 6: Deploy and apply the model

[1872] 1. The server trains a new generative AI model and deploys it to the system.

[1873] 2. The server applies the new model to future advice generation.

[1874] Input: A new generative AI model

[1875] Data processing: Deploy the new model to the system

[1876] Output: Apply the new model to future advice generation

[1877] Specific behavior:

[1878] The server learns a new generative AI model to improve the accuracy of life insurance advice, and uses this model for future advice generation.

[1879] (Application example 1)

[1880] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1881] Today's consumers often find it difficult to choose from a complex range of financial products and services. Finding the best financial plan for their individual circumstances and keeping up with constantly changing financial information can be a particularly challenging task. Furthermore, there is a lack of tools to analyze users' spending patterns and income information in real time and provide appropriate advice, making it difficult to manage assets efficiently.

[1882] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1883] In this invention, the server includes: means for learning information on an individual's annual income, family composition, and insurance and loan enrollments; means for obtaining information on various insurance and loan interest rates that changes daily from external sources and updating the database; means for generating and providing optimal insurance, loan review proposals, and tax-saving advice to the user; means for proposing optimal financial products based on family composition and years of service and supporting the procedures; means for recording a dialogue log with the user and continuously training an AI model to improve the accuracy of the advice; means for analyzing the user's spending patterns and income information and providing financial advice and savings proposals in real time; and means for obtaining the latest financial data from external sources, calculating income and expenditure balances, and generating advice. This enables users to easily select optimal financial products and services from complex financial products and realizes efficient asset management.

[1884] "Annual income" is the total amount of income an individual receives in a year.

[1885] "Family composition" refers to the composition of members in the household to which an individual belongs, and includes, for example, information on whether or not the individual has a spouse and children.

[1886] "Insurance" generally refers to financial products that individuals and companies contract to reduce risk, and includes life insurance, medical insurance, and automobile insurance.

[1887] A "loan" is a financial product that individuals or companies borrow from financial institutions and repay with a certain interest rate.

[1888] "External sources" refers to information sources provided by third parties such as financial institutions and insurance companies.

[1889] A "database" is a software system for efficiently storing, managing, and retrieving information.

[1890] "Advice" refers to specific advice or suggestions given to an individual.

[1891] "Spending patterns" are information that shows the trends and distribution of how individuals spend their money.

[1892] "Income information" refers to data about an individual's source and amount of income.

[1893] "Real-time" refers to the temporal characteristics of processing data and providing results almost immediately.

[1894] An "AI model" is an algorithm or program that uses machine learning or artificial intelligence to perform a specific task.

[1895] A "prompt" is an instruction or question that is input to a generative AI model and influences the output that the model generates.

[1896] MODE FOR CARRYING OUT THE INVENTION

[1897] This invention is a system that functions as a personal financial planner (FP) for each user, analyzing the user's income and expenditure information in real time and providing optimal financial advice based on that information. This system is implemented using the following hardware and software.

[1898] Collecting and updating information

[1899] Collecting basic information

[1900] Users use a dedicated application to enter information such as annual income, family composition, insurance and loan enrollment. This information is received by a device such as a smartphone and sent to a server, which stores the information in a database and checks for consistency and format.

[1901] Retrieving and updating external information

[1902] The server periodically collects information from external sources such as financial institutions and insurance companies, such as the latest insurance rates, loan interest rates, and economic news, which is then stored in a database and matched with the user's information.

[1903] Generating Advice

[1904] User information analysis

[1905] The server uses an algorithm to analyze the user information stored in the database and external information it has acquired, and generates advice on the best financial products for the user's current situation and tax-saving strategies.

[1906] Providing advice

[1907] The device provides advice and notifies the user. For example, if mortgage interest rates fall, the server generates a new loan offer and the device sends a notification to the user. When the user confirms the notification, the specific advice details are displayed.

[1908] Continuous learning and improvement

[1909] Collecting and learning from conversation logs

[1910] All interactions between users and the system are recorded and stored on a server. These interaction logs are used as training data for machine learning algorithms. The server periodically analyzes this data and learns new models to improve the accuracy of advice.

[1911] Deploying and applying the model

[1912] Once a new machine learning model is learned, the server deploys it and applies it to future advice generation, improving the accuracy of advice to users and providing more appropriate financial support.

[1913] Specific examples

[1914] Scenario 1: New life insurance proposal

[1915] The user reports the birth of a new child to the system. The device sends this information to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and presents the proposal.

[1916] Scenario 2: Loan Restructuring

[1917] The server retrieves the latest loan interest rate information from an external source and compares it with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan refinancing plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[1918] Prompt Sentence Examples

[1919] We collected information such as the user's income of 5 million yen, monthly expenses of 150,000 yen, 50,000 yen, and 30,000 yen, as well as family composition. The latest loan interest rate was obtained from an API as 4.0%. Analyze the user's income and expenditure balance and generate optimal financial advice based on an interest rate lower than the current loan interest rate of 4.5%.

[1920] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1921] Step 1:

[1922] The user uses a dedicated application to input basic information (annual income, family composition, insurance and loan information). This is the input information. The device sends this information to the server, which stores the received information in a database and performs formal checks. Through this process, the user's basic information is stored in the database.

[1923] Step 2:

[1924] The server periodically retrieves information such as the latest insurance rates, loan interest rates, and economic news from external sources (financial institutions and insurance companies). This information becomes the input of external financial data. The server stores the retrieved information in a database and compares it with the user's existing data. This updates the database with the latest financial information.

[1925] Step 3:

[1926] The server performs an algorithmic analysis based on user information and external financial information. At this time, the user's income and expenditure balance, insurance, and loan status are calculated. The input is the user's basic information and external financial information, and the output is the generated advice. Specifically, the system compares the user's income and expenses and creates a proposal for reviewing the optimal financial product based on fluctuations in loan interest rates.

[1927] Step 4:

[1928] The terminal receives advice from the server and provides it to the user in the form of a notification. At this time, the user is shown specific details of the advice. For example, the device may notify the user that mortgage interest rates have dropped and propose a new loan plan. It may also propose a life insurance plan that reflects changes in the user's family structure. This allows the user to receive the latest financial advice in real time.

[1929] Step 5:

[1930] All interactions between the user and the system are recorded and stored on the server. This adds the interaction log to a database. This interaction log serves as training data for future advice generation processes. The server uses machine learning algorithms to analyze this data and improve the accuracy of the advice.

[1931] Step 6:

[1932] Once a new machine learning model is trained, the server deploys it in real time and applies it to the next and subsequent advice generation. This process enables the system to always use the latest information and learning models to provide users with optimal financial advice.

[1933] Step 7:

[1934] For example, when a user reports the birth of a new child to the system, the device sends this information to the server. The server updates the database with the change in family structure and generates a new life insurance plan. The plan is then notified to the user via the device, and the proposed plan is presented in detail.

[1935] Prompt Sentence Examples

[1936] We collected information such as the user's income of 5 million yen, monthly expenses of 150,000 yen, 50,000 yen, and 30,000 yen, as well as family composition. The latest loan interest rate was obtained from an API as 4.0%. Analyze the user's income and expenditure balance and generate optimal financial advice based on an interest rate lower than the current loan interest rate of 4.5%.

[1937] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1938] This invention provides more accurate advice and services by combining a system that functions as a personal financial planner (FP) with an emotion engine that recognizes the user's emotions. This system collects information on an individual's annual income, family composition, and insurance and loan enrollment, and not only provides optimal advice based on financial information that changes daily, but also recognizes the user's emotions and takes an approach based on that emotional data. The program processing of this system is described in detail below in natural language.

[1939] Collecting and updating information

[1940] Collecting basic information

[1941] A user accesses a dedicated application or web interface and enters information about their annual income, family composition, insurance and loans. This information is received by the device and sent to a server, which stores it in a database and checks for consistency and format.

[1942] Retrieving and updating external information

[1943] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news from financial institutions and insurance companies through their APIs. This information is stored in a database and matched with user information.

[1944] Collecting emotional information

[1945] When the device interacts with the user, it activates the emotion engine and collects dialogue logs, voice data, and facial expression data, which are then sent to a server for emotion analysis.

[1946] Generating Advice

[1947] User information analysis

[1948] The server inputs the user information stored in the database and external information into an analytical algorithm, which generates advice on the optimal insurance, loan review, and tax saving measures for the user's current situation.

[1949] Emotional information analysis

[1950] The server analyzes the collected emotional data to recognize the user's current emotional state, and adjusts the content and presentation of advice based on the user's emotional state.

[1951] Generating and saving advice

[1952] The server saves the generated advice in a database as an individual plan, converts the generated advice into JSON format, etc., and sends it to the device.

[1953] Providing advice

[1954] The device sends an advice notification to the user informing them that new advice is available, and the user acknowledges the notification and opens the application or web interface to view the detailed advice.

[1955] Continuous learning and improvement

[1956] Collection and storage of conversation logs

[1957] All interactions between users and the system are recorded and stored on a server, and these interaction logs are used as training data for machine learning algorithms.

[1958] Model training

[1959] The server periodically analyzes the dialogue logs and uses machine learning algorithms to train new models, including emotional data, to improve advice accuracy and user satisfaction.

[1960] Deploying and applying the model

[1961] Once a new machine learning model is learned, the server deploys it and applies it to future advice generation, improving the accuracy of advice to users and providing more appropriate financial support.

[1962] Specific examples

[1963] Example 1: Proposing a new life insurance policy

[1964] scenario

[1965] The user reports the birth of a new child to the system. The device sends this information to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and asks them to confirm the proposal.

[1966] Use of emotion engine

[1967] When a user reviews a new suggestion, the device activates an emotion engine to analyze the user's reaction. For example, if the user is feeling stressed, the server will provide customized advice to reduce stress.

[1968] Example 2: Loan review

[1969] scenario

[1970] The server retrieves the latest loan interest rates from external sources and compares them with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[1971] Use of emotion engine

[1972] When a user receives a notification, the emotion engine analyzes the user's facial expression and tone of voice. If the user indicates anxiety, the server provides additional information to provide further explanation or reassurance.

[1973] In this way, the system continuously collects and updates user information and provides optimal advice, optimizing the user's financial situation and providing personalized support that takes into account their emotional state.

[1974] The processing flow will be explained below.

[1975] Collecting and updating information

[1976] Step 1: Initial input of user information

[1977] The user logs into a dedicated application or web interface and enters information about their annual income, family composition, and insurance and loan enrollment.

[1978] The terminal receives the input information and transmits it to the server.

[1979] Step 2: Save user information

[1980] The server stores the received user information in a database.

[1981] The server checks the information for consistency and format validity.

[1982] Step 3: Obtaining external information

[1983] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies.

[1984] The server stores this information in a database and matches it with the user's information.

[1985] Generating Advice

[1986] Step 4: Analyzing User Information

[1987] The server inputs user information stored in a database and external information into an analysis algorithm.

[1988] The server generates advice on the optimal insurance, loan review, and tax saving measures for the user's current situation.

[1989] Step 5: Collecting emotional information

[1990] When the device interacts with the user, it activates an emotion engine and collects dialogue logs, voice data, and facial expression data.

[1991] The device transmits the collected emotion data to a server.

[1992] Step 6: Emotional Analysis

[1993] The server analyzes the collected emotional data to recognize the user's current emotional state.

[1994] The server adjusts the content and presentation of advice based on the emotional state.

[1995] Step 7: Generate and save advice

[1996] The server stores the generated advice in a database as a separate plan.

[1997] The server converts the generated advice into JSON format or similar and sends it to the terminal.

[1998] Step 8: User notification and confirmation

[1999] The terminal sends an advice notification to the user, informing them that new advice is available.

[2000] The user acknowledges the notification and opens the application or web interface to view detailed advice.

[2001] Continuous learning and improvement

[2002] Step 9: Collect and store conversation logs

[2003] The server records all user interaction logs and stores them in a database.

[2004] Step 10: Train the model

[2005] The server periodically analyzes the dialogue logs and uses machine learning algorithms to train new models.

[2006] The server also learns from emotional data, improving the accuracy of advice and user satisfaction.

[2007] Step 11: Deploy and apply the model

[2008] The server deploys the newly trained model and applies it to future advice generation.

[2009] The server will provide better financial support based on the new model.

[2010] Specific examples

[2011] Example 1: Proposing a new life insurance policy

[2012] Step 1: Enter changes to family composition

[2013] A user reports the birth of a new child to the system.

[2014] The terminal receives this information and sends it to the server.

[2015] Step 2: Update the database

[2016] The server stores the new family structure information in the database.

[2017] The server compares the insurance coverage with existing insurance coverage and evaluates the need for new insurance.

[2018] Step 3: Collecting emotion data

[2019] The device activates the emotion engine when a user reports something and collects dialogue logs and voice data.

[2020] The device transmits the emotion data to the server.

[2021] Step 4: Analyze the sentiment data

[2022] The server analyzes the collected emotional data to understand how the user feels about the birth of the new child.

[2023] The server sets the content of advice in an approach that matches the user's emotions.

[2024] Step 5: Generate and notify advice

[2025] The server generates the optimal life insurance plan and sends it to the terminal.

[2026] The device will send a notification to the user prompting them to confirm the suggestion.

[2027] Example 2: Loan review

[2028] Step 1: Obtaining external information

[2029] The server retrieves the latest loan interest rate information from an external source.

[2030] The server stores this information in a database.

[2031] Step 2: Reevaluate the loan

[2032] The server compares the new interest rate information with the user's current loan terms.

[2033] If the server finds favorable terms, it generates a new loan plan.

[2034] Step 3: Collecting emotion data

[2035] The device activates an emotion engine when the user interacts with it, and collects voice data and facial expression data.

[2036] The device transmits the emotion data to the server.

[2037] Step 4: Analyze the sentiment data

[2038] The server analyzes the emotional data and determines the emotional state of the user when they receive the notification.

[2039] The server provides additional information to allay the user's concerns and doubts.

[2040] Step 5: Generate and notify advice

[2041] The server transmits the generated loan plan to the terminal.

[2042] The device will then send a notification to the user prompting them to confirm the new loan offer.

[2043] This allows users to always receive the latest and most appropriate financial advice, and also enjoy personalized support that takes into account their emotional state.

[2044] Example 2

[2045] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2046] In modern society, personal financial information is diverse, making it difficult to obtain accurate advice tailored to daily changing market information and individual circumstances. Furthermore, there is a lack of personalized support that takes into account users' emotions and stress levels, resulting in reduced user satisfaction and convenience. Furthermore, mechanisms for continuous learning and model improvement to improve the accuracy of financial advice have not been effectively implemented. There is a need for a system that can resolve these issues and provide more accurate financial advice.

[2047] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[2048] In this invention, the server includes: means for inputting information on an individual's annual income, family composition, and insurance and loan enrollments; means for transmitting and storing the information input by the user to a database; means for acquiring information on various insurance and loan interest rates from external sources and periodically updating the database; means for collecting emotional information such as a dialogue log, voice data, and facial expression data using an emotion engine; means for generating optimal insurance and loan review proposals and tax-saving advice based on the collected user information and emotional data; means for storing the generated advice in a database and transmitting it to a terminal; means for adjusting the content and expression of the advice based on the emotional data and providing personalized information to the user; means for sending new advice and financial product plan change notifications to the user; means for recording a dialogue log with the user, periodically training a machine learning algorithm to improve the accuracy of the advice; and means for deploying a new machine learning model and applying it thereafter. This enables the provision of highly accurate personalized advice tailored to an individual's financial situation and emotional state.

[2049] "Individual annual income" is the total income earned by a user in a year.

[2050] "Family structure" is information that indicates the number of members in the user's household and their relationships.

[2051] "Insurance" refers to insurance contracts such as life insurance, medical insurance, and non-life insurance that the user has subscribed to.

[2052] A "loan" is a mortgage, car loan, personal loan, or other debt agreement taken out by a user.

[2053] A "database" is a part of a computer system where collected information is stored and managed.

[2054] "External Source" is an institution or data source that provides information from outside the system, such as a financial institution or insurance company.

[2055] An "emotion engine" is software or a system that analyzes dialogue logs, voice data, and facial expression data to identify a user's emotional state.

[2056] A "dialogue log" is data that records the content of conversations between a user and a system.

[2057] A "machine learning algorithm" is an algorithm that analyzes large amounts of data, recognizes patterns, and makes future predictions and classifications.

[2058] An "AI model" is a mathematical model built on machine learning algorithms for making predictions and classifications.

[2059] "Personalized information" is information or advice that is customized based on a user's individual situation or emotional state.

[2060] A "notification" is a message that informs the user that new information or advice is available.

[2061] A "Plan Change Notification" is a message informing a user of a proposed change to their financial product or insurance plan.

[2062] "Means" refers to the functions or processes used to achieve a particular purpose.

[2063] Collecting basic information

[2064] The user accesses a dedicated application or web interface and enters information such as annual income, family composition, and insurance and loan enrollment. The device receives this information and temporarily stores it. The device then sends the saved user information to a server, which stores it in a database and checks its consistency and format. This ensures that the user information is accurately registered in the database.

[2065] Retrieving and updating external information

[2066] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies. The server stores the information obtained from external sources in a database and compares it with user information. This ensures that the database always contains up-to-date information, forming the basis for generating accurate advice.

[2067] Collecting and analyzing emotional information

[2068] When a user begins interacting with the system, the device activates the emotion engine. The device collects emotional information, such as dialogue logs, voice data, and facial expression data, in real time and sends it to the server. The server then analyzes the received emotional data using a machine learning algorithm to identify the user's current emotional state. This allows the system to provide optimal advice based on the user's emotional state.

[2069] Generating and Serving Advice

[2070] The server uses analytical algorithms to generate optimal financial advice based on user information in the database and external information. The generated advice is stored in the database, converted to JSON format, etc., and sent to the device. The device sends an advice notification to the user, informing them that new advice is available. The user confirms the notification and views the detailed advice in the application or web interface.

[2071] Continuous training and model deployment

[2072] The server records dialogue logs with the user and stores them in a database. The dialogue logs and emotion data are periodically analyzed, and a new model is trained using a machine learning algorithm. The new trained model is deployed and applied to future advice generation. This improves advice accuracy and user satisfaction.

[2073] Specific examples

[2074] Example 1: New life insurance proposal

[2075] The user reports the birth of a new child to the system via a web interface. The device sends this information to the server, which updates the database with the change in family structure. The server generates a new life insurance plan taking the change in family structure into account. The device notifies the user of the new insurance plan proposal and asks them to confirm the proposal. As the user confirms the proposal, the device activates an emotion engine to analyze the user's reaction in real time. The server provides customized advice to reduce the user's stress.

[2076] Example 2: Loan Restructuring

[2077] The server retrieves the latest loan interest rate information from an external source and compares it with the user's current loan interest rate. If there is a favorable change in interest rate, the server generates a new loan plan and the device notifies the user. The user receives the notification, reviews the proposal, and decides whether to proceed. The device activates an emotion engine to analyze the user's facial expressions and tone of voice. If the user shows signs of anxiety, the server provides additional information to provide further explanation or reassurance.

[2078] Example prompt sentence:

[2079] "Please let us know about the birth of your new child and provide appropriate life insurance recommendations."

[2080] The flow of the identification process in the second embodiment will be described with reference to FIG.

[2081] Step 1: Gather basic information

[2082] User initiated

[2083] Users access a dedicated application or web interface and enter information about their annual income, family composition, and insurance and loan coverage.

[2084] Input: Annual income, family composition, insurance, and loan information entered by the user.

[2085] Output: User information received by the device.

[2086] Terminal Processing

[2087] The device receives the information entered by the user and temporarily stores it.

[2088] Input: Information entered by the user.

[2089] Output: Data in a format that can be sent to the server.

[2090] Server Processing

[2091] The device sends the stored user information to the server, which stores this information in a database and checks its consistency and format.

[2092] Input: User information received from the device.

[2093] Output: User information stored in a database, data checked for consistency and format.

[2094] Step 2: Retrieving and updating external information

[2095] Server activation

[2096] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies.

[2097] Input: API request.

[2098] Output: Financial information obtained from external sources.

[2099] Server Processing

[2100] The server stores the information obtained from external sources in a database and matches it with user information.

[2101] Input: Financial information from external sources.

[2102] Output: The external information updated in the database.

[2103] Step 3: Collecting and analyzing emotional information

[2104] User initiated

[2105] When the user starts interacting with the system, the device activates the emotion engine.

[2106] Input: User interaction begins.

[2107] Output: Launch of emotion engine.

[2108] Terminal Processing

[2109] The device collects emotional information such as dialogue logs, voice data, and facial expression data in real time and sends it to the server.

[2110] Input: Dialogue logs, voice data, facial expression data.

[2111] Output: Emotion data sent to the server.

[2112] Server Processing

[2113] The server analyzes the received emotional data using machine learning algorithms to identify the user's current emotional state.

[2114] Input: Emotion data received from the device.

[2115] Output: Parsed emotional state data.

[2116] Step 4: Generating and serving advice

[2117] Server activation

[2118] The server uses analytical algorithms to generate optimal financial advice based on user information in the database and external information.

[2119] Input: User information in the database and external information.

[2120] Output: The generated financial advice.

[2121] Server Processing

[2122] The generated advice is stored in a database, converted to JSON format, etc., and sent to the terminal.

[2123] Input: Generated financial advice.

[2124] Output: Advice data in a format that can be sent to a terminal.

[2125] Terminal activation

[2126] The device sends an advice notification to the user informing them that new advice is available, and the user can view the notification and the detailed advice in the application or web interface.

[2127] Input: Advice data from the server.

[2128] Output: Notify the user and provide further advice.

[2129] Step 5: Continue training and deploy the model

[2130] Server activation

[2131] The server records the user interaction log and stores it in a database.

[2132] Input: User interaction log.

[2133] Output: Interaction logs stored in a database.

[2134] Server Processing

[2135] Dialogue logs and sentiment data are analyzed periodically, and new models are trained using machine learning algorithms.

[2136] Input: Dialogue logs and emotion data.

[2137] Output: A new trained machine learning model.

[2138] Server Processing

[2139] The new trained model is deployed and applied to future advice generation.

[2140] Input: A trained machine learning model.

[2141] Output: The new deployed machine learning model.

[2142] (Application example 2)

[2143] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[2144] Conventional financial planner (FP) systems provide optimal advice based on a user's annual income, family structure, insurance, and loan information. However, because they do not take the user's emotional state into account, the receptivity and degree of personalization of the advice is insufficient. This poses a challenge in that they are unable to respond appropriately in situations where the user feels anxious or stressed. Furthermore, there is a need for a system that can quickly respond to changes in the user's daily consumption behavior and emotions.

[2145] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[2146] In this invention, the server includes means for learning information on an individual's annual income, family composition, and insurance and loan enrollments, means for acquiring daily changing economic information from external sources and updating the database, means for generating and providing optimal insurance and loan review proposals and tax saving advice to the user, means for analyzing the user's facial expression data and voice data and recognizing the emotional state, means for adjusting the content and presentation of the advice based on the user's emotional data, and means for recording a dialogue log with the user and continuously training the AI ​​model to improve the accuracy of the advice. This enables personalized, highly accurate advice that takes into account both the user's economic situation and emotional state.

[2147] "Individual annual income" refers to the total income earned by a particular individual over a certain period of time.

[2148] "Family structure" refers to the composition of members and their relationships within a particular household.

[2149] "Insurance" is a contract that provides financial compensation for risks to persons or property.

[2150] A "loan" is a financial contract in which you promise to repay a borrowed amount under certain conditions.

[2151] "Means of learning" refers to the function of collecting, analyzing, and memorizing information.

[2152] "External source" refers to an information source provided from outside the system.

[2153] "Means for updating the database" refers to a function for updating existing information to the latest information.

[2154] "Optimal insurance and loan review proposals" refer to proposals for changes to insurance and loans that are most suitable for the user's current situation.

[2155] "Tax advice" means instructions or suggestions for minimizing your tax burden.

[2156] "Means of providing" refers to the function for delivering information and advice to users.

[2157] "Financial products" are products traded on the market, including stocks, bonds, investment trusts, etc.

[2158] "Means to support procedures" refers to support functions that allow users to smoothly carry out the necessary procedures.

[2159] An "interaction log" is a history of interactions between a user and a system.

[2160] An "AI model" is a data analysis model based on artificial intelligence.

[2161] "Means for recognizing emotional states" refers to a function that analyzes and recognizes emotions from input data such as the user's facial expressions and voice.

[2162] "Facial expression data and voice data" refers to digital information related to the user's facial expressions and voice.

[2163] "Means for adjusting the content and presentation of advice based on emotional data" refers to a function for changing the content and presentation of advice provided according to the user's emotional state.

[2164] This invention is a system that combines a personal financial planner (FP) system with an emotion recognition engine to provide highly accurate advice to users. Specifically, the system is configured using the following hardware and software:

[2165] System Components

[2166] 1. User data collection

[2167] The device (smartphone, tablet, smart glasses, etc.) provides an interface to collect information about the user's annual income, family composition, insurance, and loans. This information is sent to a server and stored in a database.

[2168] 2. Obtaining information from external sources

[2169] The server periodically obtains information such as the latest insurance rates, loan interest rates, and economic news through the APIs of financial institutions and insurance companies. This information is also stored in a database and compared with user information.

[2170] 3. Collecting emotional information

[2171] The device's built-in camera and microphone are used to collect the user's facial expression and voice data, which is then analyzed by an emotion recognition engine to recognize the user's emotional state.

[2172] 4. Generating Advice

[2173] The server uses user information stored in a database, external information, and emotional data to generate optimal advice, including insurance and loan restructuring and tax-saving strategies. It also takes into account the user's emotional state and adjusts the content and presentation of the advice.

[2174] 5. Providing Advice and Notification

[2175] The generated advice is converted to JSON format or similar and sent to the terminal. The terminal then sends an advice notification to the user informing them that new advice is available. The user can then confirm the notification and open the interface to view the detailed advice.

[2176] 6. Continuous learning and improvement

[2177] All user interaction logs are recorded and stored on the server. These interaction logs are used by the machine learning algorithm to learn new models and improve the accuracy of advice. Once a new model is learned, the server deploys it and applies it to future advice generation.

[2178] Specific examples

[2179] Example 1: Proposing a new life insurance policy

[2180] The user reports the birth of a new child to the system. This information is sent to the server, which updates the database. The server takes the changes in family structure into account and generates a new life insurance plan. The device notifies the user and asks them to confirm the proposal.

[2181] Use of Emotion Recognition

[2182] When a user confirms a new suggestion, the device activates an emotion recognition engine to analyze the user's reaction. For example, if the user is feeling stressed, the server will provide customized advice to reduce stress.

[2183] Example 2: Loan review

[2184] The server retrieves the latest loan interest rates from external sources and compares them with the user's current loan interest rate. If there is a favorable change in interest rates, the server generates a new loan plan and the terminal notifies the user. The user reviews the proposal and decides whether to proceed.

[2185] Use of Emotion Recognition

[2186] When a user receives a notification, an emotion recognition engine analyzes the user's facial expressions and tone of voice. If the user indicates anxiety, the server provides additional information to provide further explanation or reassurance.

[2187] Prompt Sentence Examples

[2188] "Give me an example of a user opening a finance app and checking their latest finances and sentiment data."

[2189] "Give me an example of a system where your personal financial planner provides advice based on today's spending behavior and sentiment data."

[2190] In this way, the system enables personalized, highly accurate advice that takes into account both the user's financial situation and emotional state.

[2191] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2192] Step 1:

[2193] Collection of User Information

[2194] Input: User's annual income, family composition, insurance and loan information.

[2195] How it works: The user opens a dedicated application on a device such as a smartphone or tablet and enters the necessary information, which is then sent from the device to the server.

[2196] Output: The user information sent to the server.

[2197] Step 2:

[2198] Obtaining external information

[2199] Input: The API endpoint of the financial institution or insurance company.

[2200] How it works: The server periodically calls the API to retrieve information such as the latest insurance rates, loan interest rates, and economic news.

[2201] Output: A database containing up-to-date financial information.

[2202] Step 3:

[2203] Collecting emotional information

[2204] Input: User's facial expression data and voice data.

[2205] How it works: When a user uses the device, the camera and microphone are activated to collect the necessary emotional data. The emotion recognition engine analyzes this data to recognize the user's emotional state.

[2206] Output: Emotion data sent to the server.

[2207] Step 4:

[2208] Data analysis and advice generation

[2209] Input: User information, external financial information, sentiment data.

[2210] How it works: The server feeds all the information stored in the database into an analytical algorithm to generate advice on the best insurance, loan review, tax savings, etc. The content and presentation of the advice is adjusted based on the emotional data.

[2211] Output: The generated advice plan.

[2212] Step 5:

[2213] Notification and provision of advice

[2214] Input: The generated advice plan.

[2215] How it works: The server converts the advice plan into a format such as JSON and sends it to the device. The device then sends an advice notification to the user informing them that new advice is available. The user confirms the notification and opens the application to view the detailed advice.

[2216] Output: Advice sent to the user's device.

[2217] Step 6:

[2218] Interaction logging and continuous learning

[2219] Input: Log of interactions between the user and the system.

[2220] How it works: All user interactions with the system are logged and stored on a server. These logs are used as training data for machine learning algorithms to learn new AI models. Once a new model is learned, the server deploys it and applies it to future advice generation.

[2221] Output: The updated AI model.

[2222] This series of processing steps allows the system to take into account both the user's financial situation and emotional state and provide highly accurate and personalized advice.

[2223] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquir...

Claims

1. A means to learn information about an individual's annual income, family structure, insurance and loans, A means to obtain information on various insurance and loan interest rates that change daily from external sources and update the database, A means for generating and providing optimal insurance, loan review proposals, and tax saving advice to users; We will propose the most suitable financial products according to the family structure and the number of years since the death, and provide support for the procedures. A system that includes a means to record dialogue logs with users, continuously train the AI ​​model, and improve the accuracy of advice.

2. A means of obtaining economic news and market information and assessing events that may affect the User's financial situation; 10. The system of claim 1, further comprising means for sending notifications to a user based on those events.

3. A means of verifying the consistency and format of user input; The system according to claim 1 , further comprising means for sending a notification to the user to prompt the user to correct any deficiencies.

Images