System

Abstract

A system is provided.SOLUTION: A system, comprising: means for obtaining transaction information of a financial account of an elderly person; means for analyzing the obtained transaction information and detecting an anomaly; means for notifying a user of the detected anomaly and seeking permission to notify family members; and means for sending a notification to the family members based on the user's permission.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] Elderly people often have difficulty managing their money. For example, they may not notice abnormal expenditures on fixed expenses such as utility bills or mobile phone bills, and end up paying large amounts over a long period of time. It is also difficult for elderly people to deal with large or suspicious transactions quickly. Furthermore, many elderly people are reluctant to entrust the management of their money to their family members. There is a need to solve these issues and provide an environment where elderly people can live with peace of mind. [Means for solving the problem]

[0005] This invention provides a system that acquires transaction information from an elderly person's financial account and analyzes that information to detect abnormal transactions. Specifically, the system calculates the average value and standard deviation of specific fixed costs from the acquired transaction information, and detects transactions that exceed the predicted range as abnormal. It also detects an abnormality when the transaction amount in the transaction information exceeds a set threshold. The system notifies the user (elderly person) of any detected abnormalities and asks for permission to notify family members. Notifications are then sent to family members based on the user's permission. This allows the elderly person to quickly identify abnormal transactions and take necessary action.

[0006] "Elderly" refers to people aged 65 or older who need support when it becomes difficult for them to manage their finances.

[0007] A "financial account" is an account opened at a bank or other financial institution, and is used primarily for transactions such as deposits, withdrawals, and transfers.

[0008] "Transaction information" refers to detailed data on transactions such as deposits, withdrawals, and transfers related to a financial account, including the date, transaction amount, transaction type, and counterparty.

[0009] "Acquisition" refers to the act of collecting transaction information from financial accounts, and is primarily an action that the system performs through the financial institution's API.

[0010] "Analysis" refers to the act of analyzing acquired transaction information and detecting abnormal patterns or transactions.

[0011] "Abnormal" refers to transactions that deviate significantly from normal transaction patterns or unexpectedly large transactions, and indicates a situation in which fraud or suspicious transactions are suspected.

[0012] "Notification" refers to the act of sending information to warn or prompt the user or their family members when an abnormality is detected.

[0013] "User" refers to the entity that uses this system, particularly the elderly person themselves, and is the recipient of notifications from the system.

[0014] "Family" refers to the user's close relatives, who are registered in advance to support the user with financial troubles.

[0015] "Authorization" refers to the user's act of confirming and agreeing to sending notifications to family members.

[0016] "Sending a notification" refers to the act of sending information about an abnormal transaction to family members when an abnormality is detected and the user's permission is obtained.

[0017] "System" refers to a collection of hardware and software for performing a series of processes such as acquiring transaction information on elderly people's financial accounts, analyzing it, detecting anomalies, and sending notifications. [Brief explanation of the drawings]

[0018] [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

[0019] 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.

[0020] First, the terms used in the following description will be explained.

[0021] 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).

[0022] 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.

[0023] 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.

[0024] 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.

[0025] 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."

[0026] [First embodiment]

[0027] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0028] 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.

[0029] 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).

[0030] 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.

[0031] 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.

[0032] 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.

[0033] 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.

[0034] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0035] 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.

[0036] 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.

[0037] 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.

[0038] 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."

[0039] To implement this invention, a system is constructed that monitors transaction information in elderly people's financial accounts, detects abnormal transactions, and notifies them. The system is primarily composed of a server, terminals (such as smartphones and PCs), and users.

[0040] Server Processing

[0041] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database.

[0042] The server then analyzes the transaction information and detects abnormal transactions. Transaction anomalies are defined as follows:

[0043] 1. Fixed cost anomalies: Based on past transaction data, the average value and standard deviation of fixed costs such as utility bills and mobile phone bills are calculated. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ± 3 × standard deviation), it is detected as an anomaly.

[0044] 2. Large Transactions: If a large transaction occurs that exceeds a set threshold, it is detected as an anomaly.

[0045] If an anomaly is detected, the server sends an alert message to the user's device, containing details of the anomalous transaction and a message asking the user to confirm whether to notify family members.

[0046] Terminal handling

[0047] The terminal receives the alert message sent from the server and notifies the user. The user can then check the alert message and confirm the transaction details themselves. They can also choose whether to send notifications to their family members. If the user allows notifications to be sent to their family members, the information is sent from the terminal to the server.

[0048] User Action

[0049] The user checks the alert message displayed on the device and considers how to respond to any unusual transactions. If the transaction is suspicious, they can choose not to notify their family, but if not, it is recommended that they allow notifications. This choice is made on the device, and the result is sent to the server.

[0050] Additional Server Processing

[0051] If the user allows notifications to family members, the server will send an alert message to the family members' devices registered in advance, allowing the family members to receive detailed information about abnormal transactions and take appropriate measures.

[0052] Specific examples

[0053] Specifically, the following scenarios are possible:

[0054] 1. Normal transaction:

[0055] The server retrieves daily utility bills (10,000 yen) from the financial institution API.

[0056] This transaction is within the expected range and is not detected as an anomaly.

[0057] 2. Unusual utility bill transactions:

[0058] The server retrieves a utility bill transaction for 50,000 yen from the financial institution API.

[0059] This transaction is detected as an anomaly because it is far outside the expected range (average 10,000 yen ± 30,000 yen).

[0060] The server generates an alert and sends it to the user's terminal.

[0061] The user checks the alert on the device and allows notification to family members.

[0062] The server sends a notification to the family's device, providing them with detailed information.

[0063] 3. Large transfers:

[0064] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[0065] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[0066] The server generates an alert and notifies the user.

[0067] If the user is not aware of the incident, they can allow their family members to be notified.

[0068] The server sends a notification to the family's devices and action is taken to address the issue.

[0069] This system allows seniors to ensure their financial security while managing any irregularities in their transactions themselves, and allows family members to intervene quickly if necessary, helping to ensure the safety of seniors' daily lives.

[0070] The processing flow will be explained below.

[0071] Program processing

[0072] Server Processing

[0073] Step 1:

[0074] The server accesses the financial institution API at regular intervals (e.g., 9:00 AM every day).

[0075] The server sends a request for transaction data using authentication information for the financial institution API.

[0076] If authentication is successful, the financial institution API returns the user's transaction information.

[0077] The server stores the acquired transaction information in a database.

[0078] Step 2:

[0079] The server retrieves the latest transaction information from the database.

[0080] Filter newly acquired transaction information to identify transactions of interest.

[0081] The transactions covered include the date, transaction amount, transaction type, and counterparty.

[0082] Step 3:

[0083] The server categorizes the transaction information into fixed cost categories.

[0084] The server analyzes past transaction data to identify fixed expense categories such as utility bills and cell phone bills.

[0085] Check if the new transaction falls into the fixed cost category.

[0086] Step 4:

[0087] The server detects anomalies in fixed costs.

[0088] The server calculates the mean and standard deviation for the fixed cost categories.

[0089] If the new transaction amount exceeds the expected range (mean ± 3 x standard deviation), it is marked as an anomaly.

[0090] Step 5:

[0091] The server detects anomalies in large transactions.

[0092] Set a large amount threshold (e.g., 1 million yen) for all transactions.

[0093] If the transaction amount exceeds this threshold, it is marked as an anomaly.

[0094] Step 6:

[0095] If an anomaly is detected, the server generates an alert message.

[0096] The alert message will include details of the unusual transaction, the type of abnormality, and confirmation of permission to notify family members.

[0097] The server sends this message to the user's terminal.

[0098] Terminal handling

[0099] Step 1:

[0100] The terminal receives the alert message from the server.

[0101] Notify the user that new notifications are available.

[0102] The device will display the notification content in a pop-up or notification bar.

[0103] Step 2:

[0104] The terminal waits for the user's action.

[0105] The user opens the notification and sees the details of the alert message.

[0106] The alert message will include transaction details, a description of the anomaly, and an option to notify family members.

[0107] Step 3:

[0108] The user presses the button to allow notifications to family members.

[0109] The terminal transmits the user's authorization information to the server.

[0110] Additional Server Processing

[0111] Step 1:

[0112] The server receives the user's authorization information.

[0113] Retrieve contact information for registered family members from the database.

[0114] Step 2:

[0115] The server sends an alert message to the family.

[0116] A message is sent to the family member's device containing details of the abnormal transaction, the user's selection, and suggestions on how to proceed.

[0117] If the message is sent successfully, a record of it is saved in the database.

[0118] User Action

[0119] Step 1:

[0120] The user checks the alert message on the terminal.

[0121] Check the details of any unusual transactions to see if they apply to you.

[0122] Step 2:

[0123] The user selects whether to notify family members.

[0124] If you are not aware of the incident, allow your family to be notified.

[0125] If permission is granted, the information is sent from the terminal to the server.

[0126] Based on these specific processing steps, abnormal transactions in the elderly person's financial accounts can be monitored and necessary measures can be taken promptly.

[0127] Example 1

[0128] 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."

[0129] The increase in fraudulent and abnormal transactions in elderly people's financial accounts has created a need for detection and rapid notification. Current systems take time to acquire and analyze transaction information, and lack the means to quickly notify users and their families when an abnormality is detected. This increases the likelihood that elderly people will be exposed to financial risk, which is a major source of anxiety.

[0130] 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.

[0131] In this invention, the server includes means for periodically acquiring transaction information from the elderly person's financial account via the financial institution's API, means for storing the acquired transaction information in a database, means for analyzing the stored transaction information and detecting abnormal transactions, means for sending an alert message including details of the detected abnormal transaction to the user's terminal, means for providing an interface that allows the user to check the alert and select whether to notify family members, and means for sending a notification to the family member's terminal based on the user's permission. This makes it possible to quickly detect abnormalities in the elderly person's financial transactions and notify the user and family members in a timely manner.

[0132] "Financial institution API" stands for Application Program Interface provided by financial institutions, which is an interface used by external systems to obtain financial transaction data.

[0133] An "elderly person's financial account" refers to an account opened by an elderly person at a bank or other financial institution, through which financial transactions such as deposits, withdrawals, and transfers are conducted.

[0134] "Transaction information" is data showing the transaction history of a financial account within a specific period, and includes transaction dates, transaction amounts, transaction types, and transaction partner information.

[0135] "Database" means a computer system for systematically storing and managing acquired transaction information, allowing for easy access and analysis.

[0136] An "abnormal transaction" is a transaction that is determined to be out of the ordinary based on past transaction patterns or established thresholds, including transactions that may be fraudulent or involve abnormal amounts.

[0137] An "alert message" is a notification sent to a user when an abnormal transaction is detected, and contains details of the transaction and information urging the user to take action.

[0138] "Users" are those who use this system to receive notifications of abnormal transactions, and primarily refer to elderly people and their families.

[0139] "Terminal" refers to a device used by a user, such as a smartphone, PC, or tablet, that receives and displays alert messages from the server.

[0140] "Family devices" refer to devices such as smartphones, PCs, and tablets owned by the user's family, and alert messages are sent to these devices based on the user's permission.

[0141] To implement this invention, it is necessary to build a system that monitors transaction information in elderly people's financial accounts, detects abnormal transactions, and notifies users. This system is primarily composed of a server, terminals (such as smartphones and PCs), and users.

[0142] Server Processing

[0143] The server periodically retrieves transaction information from the elderly person's financial account via the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The retrieved information is stored in a database. Specifically, the server sends an API request at a set time each day to retrieve daily transaction data and record it in the database.

[0144] The server then analyzes the stored transaction information and detects abnormal transactions. Transaction abnormalities are defined in two patterns: fixed cost abnormalities and large transaction amounts. For fixed cost abnormalities, the server calculates the average value and standard deviation of fixed costs such as utility bills and mobile phone bills based on past transaction data. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ±3 x standard deviation), it is detected as an abnormality. On the other hand, for large transaction amounts, an abnormality is detected if a transaction that exceeds the set threshold occurs. If an abnormality is detected, the server sends an alert message to the user's device.

[0145] Terminal handling

[0146] The terminal receives the alert message sent from the server and is responsible for notifying the user. The user can then review the alert message and investigate the transaction details themselves. The interface displayed on the terminal includes an option to select whether to send notifications to family members. If the user allows notifications to be sent to family members, the terminal sends the selection to the server.

[0147] User Action

[0148] The user checks the alert message displayed on the device. If the transaction is recognizable, they can choose not to notify their family. If they are not recognizable, they can choose to allow notification to their family. This selection is sent to the server via the device.

[0149] Additional Server Processing

[0150] If the user allows notification to family members, the server will send an alert message to the pre-registered family members' devices, allowing the family members to receive detailed information about the abnormal transaction and take appropriate measures.

[0151] Specific examples

[0152] Specifically, the following scenarios are possible:

[0153] 1. Normal transaction:

[0154] The server retrieves daily utility bills (10,000 yen) from the financial institution's API.

[0155] This transaction is within the expected range and is not detected as an anomaly.

[0156] 2. Unusual utility bill transactions:

[0157] The server retrieves a utility bill transaction for 50,000 yen from the financial institution's API.

[0158] This transaction is detected as an anomaly because it is far outside the predicted range (average 10,000 yen ± 3 x 5,000 yen).

[0159] The server generates an alert message and sends it to the user's terminal.

[0160] The user confirms the alert message and allows notification to family members.

[0161] The server sends a notification to the family's device, providing them with detailed information.

[0162] 3. Large transfers:

[0163] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[0164] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[0165] The server generates an alert message and notifies the user.

[0166] If the user is not aware of the incident, they can allow their family members to be notified.

[0167] The server sends a notification to the family's devices and steps are taken to take action.

[0168] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0169] Step 1:

[0170] The server retrieves the transaction information

[0171] Input: Request to financial institution API

[0172] Specific operation: The server sends a request to the financial institution's API at a specific time interval (e.g., every day at midnight), using the authentication token to access the API and requesting transaction information for the target senior citizen's financial account.

[0173] Output: Transaction information (JSON format)

[0174] Specific data processing: Transaction information is returned from financial institutions in JSON format, including the transaction date, transaction amount, transaction type (deposit, withdrawal, transfer, etc.), and counterparty.

[0175] Step 2:

[0176] The server saves the transaction information it has acquired into a database.

[0177] Input: Retrieved transaction information (JSON format)

[0178] Specific operation: The server analyzes the acquired transaction information and extracts necessary fields, such as transaction date, transaction amount, and transaction type.

[0179] Output: Stored transaction information (specific table in database)

[0180] Specific data processing: The extracted data is properly saved in the transactions table of the database.

[0181] Step 3:

[0182] The server analyzes the transaction information and detects abnormal transactions.

[0183] Input: Historical transaction data in the database, retrieved new transaction data

[0184] Specific operation: The server calculates the average and standard deviation of fixed costs based on past transaction data. For example, using utility bill data from the past three months, the server calculates the average (e.g., 10,000 yen) and standard deviation (e.g., 5,000 yen).

[0185] Output: Abnormal transactions (list format)

[0186] Specific data calculation: If a new transaction exceeds the average value ±3 × standard deviation, the transaction is added to the list as an anomaly. Large transactions exceeding a set threshold (e.g., 1 million yen) are also added to the list as an anomaly.

[0187] Step 4:

[0188] The server generates an alert for abnormal transactions and sends it to the terminal.

[0189] Input: Detected abnormal transactions (list format)

[0190] What happens: The server creates an alert message with details of the anomalous transaction. For example, it generates a message like, "A utility bill transaction of 50,000 yen has been detected as anomalous. Does this transaction seem familiar?"

[0191] Output: Alert message (push notification to device)

[0192] Specific data processing: The generated alert message is sent to the user's device as a push notification.

[0193] Step 5:

[0194] The device receives the alert and notifies the user.

[0195] Input: The alert message sent from the server

[0196] Specific behavior: The device will display the received alert message as a pop-up notification on the screen. When the user taps the pop-up, the app will open and display more information.

[0197] Output: Notification to the user (display on screen)

[0198] Specific data display: Detailed information about the alert is displayed on the screen, and the user is given the option to "notify family members."

[0199] Step 6:

[0200] The user responds to the alert and sends their choice from the device to the server

[0201] Input: Alert message confirmation result (whether to notify family members)

[0202] Specific operation: The user checks the details of the alert and selects "Notify family members" or "Don't notify family members." The selection is sent from the device to the server.

[0203] Output: Selection (whether to notify family members)

[0204] Specific data transmission: User selections are transmitted to the server.

[0205] Step 7:

[0206] The server sends a notification to the family

[0207] Input: User's selection result (allow notification to family members)

[0208] Specific operation: The server checks the pre-registered contact information of family members, creates an alert message, and sends it to the family members' devices. For example, it generates a message such as, "An abnormal transaction has been detected in your parent's account. Please check the details."

[0209] Output: Alert message to family members (push notification to device or email)

[0210] Specific data transmission: The generated alert message will be sent to the family's device so that the family can check the detailed information.

[0211] (Application example 1)

[0212] 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."

[0213] Ensuring the security of transactions in elderly people's financial accounts is important for preventing financial damage, but it is difficult for elderly people themselves to constantly monitor their transactions for abnormalities. There is also a need for a method to quickly and reliably notify family members when abnormal transactions occur. Conventional systems lack the ability to detect abnormalities in real time and provide immediate notification, preventing family members from responding quickly. A new system is needed to resolve these issues and provide peace of mind to elderly people and their families.

[0214] 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.

[0215] In this invention, the server includes: [means for acquiring transaction information on the elderly person's financial account]; [means for analyzing the acquired transaction information and detecting abnormalities]; [means for notifying the user of detected abnormalities and requesting permission for notifying family members]; [means for sending notifications to family members based on the user's permission]; [means for sending notifications in real time when an abnormal transaction is detected]; and [means for sending similar alerts to family members]. This enables real-time detection of abnormal transactions and prompt notification, enabling the elderly person and their family members to prevent financial damage. It also provides a system that reduces the burden on the elderly person themselves and allows family members to respond quickly.

[0216] The term "elderly" generally refers to people aged 60 or over, especially those in need of social and medical support.

[0217] "Financial Account" means a deposit account or other asset management account opened by a customer at a financial institution for the purpose of managing assets.

[0218] "Transaction information" refers to detailed data regarding various financial transactions, such as deposits, withdrawals, transfers, and payments, in a financial account.

[0219] "Real-time" means that data is collected, analyzed, and communicated almost immediately, with very little time delay.

[0220] "Abnormal transactions" refer to transactions that deviate significantly from normal trading patterns, including large transactions and transactions that exceed expected ranges.

[0221] A "server" refers to a computer system that provides processing power and data to multiple clients over a network.

[0222] "Database" means a system for systematically storing and managing information in digital form.

[0223] "Notification" means a message or alert sent to inform a user or interested party of specific information.

[0224] An "alert" refers to a notification method that immediately warns the user when an abnormality or a condition requiring attention occurs.

[0225] "Family" refers to a person related by blood or marriage to an older person and who is responsible for assisting and supporting the older person in managing their life, health, and property.

[0226] To implement this invention, it is necessary to build a system consisting of a server, terminals (such as smartphones or PCs), and users. The main function of the system is to provide a mechanism for monitoring transaction information in elderly people's financial accounts and detecting and notifying abnormal transactions.

[0227] Server Processing

[0228] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and counterparty. The obtained information is stored in a database. The server uses the NumPy library to calculate the mean and standard deviation from past transaction data. It defines the mean value and prediction range (e.g., mean value ± 3 × standard deviation) of fixed costs, and detects newly obtained transactions that exceed the prediction range as anomalous transactions. Large transactions whose transaction amount exceeds a set threshold are also detected as anomalous. When an anomaly is detected, the server uses Pushbullet to send an alert message via push notification to the user's device.

[0229] Terminal handling

[0230] The device receives the alert message sent from the server and notifies the user. The user can check the alert message and check the transaction details themselves. The alert message includes details of the abnormal transaction and an option to notify family members. If the user allows notification to family members, the device sends the information to the server.

[0231] User Action

[0232] The user checks the alert message displayed on the device and decides whether or not the transaction is recognizable. If not, the user is encouraged to allow notifications to family members. This selection is made on the device, and the result is sent to the server. Based on the user's permission, the server can send alert messages to the devices of family members who have been registered in advance.

[0233] Specific examples

[0234] For example, if the average utility bill for the past six months was 10,000 yen with a standard deviation of 5,000 yen, and a new transaction of 20,000 yen occurs, the server will detect this as an anomaly. In this case, the server will send an alert to the user's smartphone stating, "Anomalous transaction detected: A utility bill transaction of 20,000 yen occurred on 2023-10-01." If the user allows notifications, similar alerts will also be sent to the smartphones of family members.

[0235] Example prompts for generative AI models

[0236] "We will design a system that monitors transaction information for elderly people's financial accounts, detects abnormal transactions, and notifies users. An abnormal transaction in the transaction data we received (for example, the average utility bill for the past six months is 10,000 yen, with a standard deviation of 5,000 yen) is one that exceeds the average value ±3 x standard deviation. We will set it up so that a push notification is sent stating, 'Abnormal transaction detected: A utility bill transaction of 50,000 yen occurred on 2023-10-01.' Please create a specific program based on this content."

[0237] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0238] Step 1:

[0239] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. The information obtained includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The input for this process is the API call, and the output is transaction information data in JSON format. The server saves the obtained data in a database.

[0240] Step 2:

[0241] The server extracts historical transaction data from the stored transaction information and calculates the mean and standard deviation of a particular fixed cost (e.g., utility bills). It uses the historical transaction data (in the database) as input and gets the mean and standard deviation as output. It uses the NumPy library to perform statistical operations.

[0242] Step 3:

[0243] The server analyzes newly acquired transaction data and determines whether the transaction falls within the expected range based on the calculated mean and standard deviation. The input is the latest transaction data, and the output is a flag indicating whether the transaction is "anomalous" or "normal." For example, if the amount of a new transaction exceeds the mean value ± 3 x standard deviation, the server detects the transaction as "anomalous."

[0244] Step 4:

[0245] The server uses Pushbullet's API to send an alert message to the user's device for transactions detected as abnormal. The alert includes details of the abnormal transaction (transaction date, amount, and counterparty) and an option to notify family members. The input is the abnormal transaction data, and the output is the alert message.

[0246] Step 5:

[0247] The device receives the alert message sent from the server and notifies the user. The input is the alert message, and the output is a push notification to the user. The user checks the content of the alert message on the device.

[0248] Step 6:

[0249] The user checks the alert message displayed on the terminal and selects how to respond to the abnormal transaction. Specifically, the user selects whether to notify family members. The input is the notified alert information, and the output is the user's selection.

[0250] Step 7:

[0251] When the user allows notifications to family members, the device sends the selection information to the server. The input is the user's selection, and the output is the notification permission data for the server.

[0252] Step 8:

[0253] Based on the user's permission, the server sends an alert message to pre-registered family members' devices. This allows family members to receive information about abnormal transactions and take prompt action. The input is notification permission data from the user, and the output is an alert message for the family members.

[0254] By following the above steps, financial transactions of elderly people can be monitored in real time, and if any abnormalities are detected, the user and their family can be notified promptly, thereby preventing financial damage.

[0255] 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.

[0256] To implement this invention, we will build a system that monitors transaction information on elderly people's financial accounts, detects and notifies abnormal transactions, and combines it with an emotion engine that recognizes user emotions. The system is primarily composed of a server, a terminal (such as a smartphone or PC), the emotion engine, and the user.

[0257] Server Processing

[0258] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database.

[0259] The server then analyzes the transaction information and detects abnormal transactions. Transaction anomalies are defined as follows:

[0260] 1. Fixed cost anomalies: Based on past transaction data, the average value and standard deviation of fixed costs such as utility bills and mobile phone bills are calculated. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ± 3 × standard deviation), it is detected as an anomaly.

[0261] 2. Large Transactions: If a large transaction occurs that exceeds a set threshold, it is detected as an anomaly.

[0262] When an anomaly is detected, the server generates an alert message, which includes details of the anomalous transaction, the type of anomaly, and a prompt to notify family members. Furthermore, an emotion engine recognizes the user's emotions and tailors the notification accordingly.

[0263] Emotion engine processing

[0264] The emotion engine recognizes emotions from the user's facial expressions, voice, text input, etc. The emotion engine performs the following tasks:

[0265] 1. Emotion recognition: Recognizes user emotions in real time through the device's camera, microphone, and keyboard input.

[0266] 2. Sentiment analysis: Send feedback to the server based on the recognized emotion (surprise, anxiety, confusion, etc.).

[0267] 3. Notification adjustment: If the user feels anxious or confused, we recommend notifying their family members. We may also change the notification content to be more gentle.

[0268] Terminal handling

[0269] The device receives the alert message sent by the server and notifies the user. The notification is coordinated as follows:

[0270] 1. Normal notification: If an abnormal transaction is detected, the user will be notified with detailed information.

[0271] 2. Adjusted Notification: Based on the analysis results of the emotion engine, notifications are sent in a gentler tone. For example, a message like, "I'm a little worried, are you okay?"

[0272] 3. Recommend notifying family: If the user is feeling anxious or confused, the option to recommend notifying family will be automatically selected.

[0273] The user checks the alert message and selects whether to notify their family. If the user allows notification to their family, the information is sent from the device to the server.

[0274] Additional Server Processing

[0275] After receiving the user's authorization, the server retrieves the family member's contact information from a database and sends an alert message to the family member, including details of the abnormal transaction, the user's selection, and suggestions for how to proceed.

[0276] Specific examples

[0277] Specifically, the following scenarios are possible:

[0278] 1. Normal transaction:

[0279] The server retrieves daily utility bills (10,000 yen) from the financial institution API.

[0280] This transaction is within the expected range and is not detected as an anomaly.

[0281] 2. Unusual utility bill transactions:

[0282] The server retrieves a utility bill transaction for 50,000 yen from the financial institution API.

[0283] This transaction is detected as an anomaly because it is far outside the expected range (average 10,000 yen ± 30,000 yen).

[0284] The emotion engine detects the user's surprise or anxiety and generates a gentle alert recommending notifying family members.

[0285] The server generates the adjusted alert and sends it to the user's device.

[0286] The user confirms the notification and allows notification to family members.

[0287] The server sends tailored alerts to family devices, allowing them to respond quickly.

[0288] 3. Large transfers:

[0289] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[0290] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[0291] The emotion engine analyzes the user's emotions and generates appropriate notification content.

[0292] The server generates an alert containing details of the abnormal transaction and notifies the user.

[0293] If the user is not aware of the incident, they can allow their family members to be notified.

[0294] The server sends a notification to the family's devices and action is taken to address the issue.

[0295] This system allows seniors to manage transaction anomalies themselves while ensuring their financial security. The introduction of an emotion engine also reduces the psychological burden on users, allowing family members to intervene quickly if necessary, further enhancing the security of seniors' daily lives.

[0296] The processing flow will be explained below.

[0297] Server Processing

[0298] Step 1:

[0299] The server accesses the financial institution API at regular intervals (e.g., 9:00 AM every day).

[0300] The server sends a request for transaction data using authentication information for the financial institution API.

[0301] If authentication is successful, the financial institution API returns the user's transaction information.

[0302] The server stores the acquired transaction information in a database.

[0303] Step 2:

[0304] The server retrieves the latest transaction information from the database.

[0305] The server filters the newly acquired transaction information to identify eligible transactions.

[0306] The transactions covered include the date, transaction amount, transaction type, and counterparty.

[0307] Step 3:

[0308] The server categorizes the transaction information into fixed cost categories.

[0309] The server analyzes past transaction data to identify fixed expense categories such as utility bills and cell phone bills.

[0310] Check if the new transaction falls into the fixed cost category.

[0311] Step 4:

[0312] The server detects anomalies in fixed costs.

[0313] The server calculates the mean and standard deviation for the fixed cost categories.

[0314] If the new transaction amount exceeds the expected range (mean ± 3 x standard deviation), it is marked as an anomaly.

[0315] Step 5:

[0316] The server detects anomalies in large transactions.

[0317] The server sets a large amount threshold (e.g., 1 million yen) for all transactions.

[0318] If the transaction amount exceeds this threshold, it is marked as an anomaly.

[0319] Step 6:

[0320] If an anomaly is detected, the server generates an alert message.

[0321] The alert message will include details of the abnormal transaction, the type of abnormality, and permission to notify family members.

[0322] Emotion engine processing

[0323] Step 1:

[0324] The emotion engine captures the user's facial expressions and voice in real time.

[0325] Data is collected through cameras and microphones.

[0326] Step 2:

[0327] The emotion engine analyzes the acquired data and recognizes the user's emotions.

[0328] The emotion engine uses facial recognition algorithms and voice analysis to identify the user's emotions.

[0329] If the user's emotion corresponds to anxiety, surprise, confusion, etc., the information is sent to the server.

[0330] Step 3:

[0331] The emotion engine sends feedback to the server to adjust the notification content.

[0332] If the user is feeling anxious or confused, change the notification message to a more gentle one.

[0333] An option to recommend notifying family members will also be added.

[0334] Terminal handling

[0335] Step 1:

[0336] The terminal receives the alert message from the server.

[0337] Notify the user that new notifications are available.

[0338] The device will display the notification content in a pop-up or notification bar.

[0339] Step 2:

[0340] The terminal waits for the user's action.

[0341] The user opens the notification to view the details of the alert message.

[0342] The message will include transaction details, a description of the anomaly, and an option to notify family members.

[0343] Step 3:

[0344] When the user presses the button to allow notifications to family members,

[0345] The terminal transmits the user's authorization information to the server.

[0346] Additional Server Processing

[0347] Step 1:

[0348] The server receives the user's authorization information.

[0349] The server retrieves contact information for registered family members from the database.

[0350] Step 2:

[0351] The server sends an alert message to the family.

[0352] A message is sent to the family member's device containing details of the abnormal transaction, the user's selection, and suggestions on how to proceed.

[0353] If the message is sent successfully, a record of it is saved in the database.

[0354] Specific processing examples

[0355] Specifically, the following scenarios are possible:

[0356] 1. Normal transaction:

[0357] The server retrieves daily utility bills (10,000 yen) from the financial institution API.

[0358] This transaction is within the expected range and is not detected as an anomaly.

[0359] 2. Unusual utility bill transactions:

[0360] The server retrieves a utility bill transaction for 50,000 yen from the financial institution API.

[0361] This transaction is detected as an anomaly because it is far outside the expected range (average 10,000 yen ± 30,000 yen).

[0362] The emotion engine detects the user's surprise or anxiety and generates a gentle alert recommending notifying family members.

[0363] The server generates the adjusted alert and sends it to the user's device.

[0364] The user confirms the notification and allows notification to family members.

[0365] The server sends tailored alerts to family devices, allowing them to respond quickly.

[0366] 3. Large transfers:

[0367] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[0368] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[0369] The emotion engine analyzes the user's emotions and generates appropriate notification content.

[0370] The server generates an alert containing details of the abnormal transaction and notifies the user.

[0371] If the user is not aware of the incident, they can allow their family members to be notified.

[0372] The server sends notifications to family members' devices so that countermeasures can be taken quickly.

[0373] Example 2

[0374] 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."

[0375] There is a need for a method and system that can respond quickly and appropriately when abnormal financial transactions occur among elderly people. Furthermore, it is necessary to consider the user's feelings when notifying them of abnormal transactions. Conventional systems have had problems such as causing unnecessary anxiety to users and delaying notification to family members. The purpose of this invention is to solve these problems and simultaneously ensure the financial security and psychological peace of mind of elderly people.

[0376] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for acquiring transaction information of the elderly person's financial account, a means for analyzing the acquired transaction information and detecting abnormalities, and a means for recognizing the user's emotions and adjusting the notification content. This makes it possible to notify the user of abnormal transactions in a manner that takes into consideration the user's emotions, and also to notify family members at an appropriate time.

[0377] "Elderly" refers to people who are older, especially those aged 65 or older, and who require special consideration in financial transactions and daily life.

[0378] "Financial account" is a general term for deposit accounts, savings accounts, fixed-term deposit accounts, etc. opened at banks, credit unions, and other financial institutions, and is a basic means for recording and managing financial transactions.

[0379] "Transaction information" refers to detailed data relating to financial transactions such as deposits, withdrawals, transfers, and debits related to a financial account, including the date, amount, type, and counterparty of the transaction.

[0380] "Abnormal transactions" refer to transactions that significantly deviate from normal trading patterns, including, for example, transactions that exceed the mean and standard deviation of a particular fixed cost or large transactions that exceed a set threshold.

[0381] "Emotion engine" is a general term for software or hardware that recognizes and analyzes emotions in real time from a user's facial expressions, voice, text input, etc.

[0382] "Adjusting the notification content" refers to a process of changing the wording and content of the notification message depending on the user's emotional state, so as not to make the user feel excessively anxious.

[0383] "Family" refers to a person who has a certain kinship relationship with the user and who has the role of supporting the user in emergencies or abnormal situations.

[0384] "Server" means the main computer system that manages the entire system and acquires, analyzes, and stores transaction information, and generates and transmits notifications.

[0385] A "terminal" refers to a device such as a smartphone or PC that is directly operated by a user and that receives notifications sent from a server.

[0386] "Notification" refers to a message sent by the system to the user or family member informing them of an abnormal transaction, and may be sent in the form of voice, text, push notification, etc.

[0387] To implement this invention, it is necessary to build a system that monitors transaction information on elderly people's financial accounts, detects abnormal transactions, and notifies them, as well as an emotion engine that recognizes user emotions. This system is primarily composed of a server, a terminal (such as a smartphone or PC), an emotion engine, and a user.

[0388] Server Processing

[0389] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, transaction amount, transaction type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database on the server. The server analyzes this data and runs an algorithm to detect abnormal transactions.

[0390] Specifically, the server detects abnormal transactions through the following steps:

[0391] 1. Fixed cost anomaly detection: Based on past transaction data, the average value and standard deviation of fixed costs such as utility bills and mobile phone bills are calculated. If the newly acquired transaction amount exceeds the predicted range of the average value ±3 x standard deviation, it is detected as an anomaly.

[0392] 2. Large transaction detection: The server detects an abnormality when a large transaction that exceeds a set threshold occurs.

[0393] When an anomaly is detected, the server sends feedback to the emotion engine, and also generates alert messages and adjusts the notification content as needed.

[0394] Emotion engine processing

[0395] The emotion engine recognizes emotions in real time from the user's facial expressions, voice, text input, etc. For this purpose, the device requires input devices such as a camera and microphone. The emotion engine provides the following functions:

[0396] 1. Emotion recognition: Recognizes the user's emotions through the device's camera, microphone, and keyboard input.

[0397] 2. Sentiment analysis: Send feedback to the server based on the recognized emotion (surprise, anxiety, confusion, etc.).

[0398] 3. Notification adjustment: If the user feels anxious or confused, we recommend notifying their family members. We may also change the notification content to be more gentle.

[0399] Terminal handling

[0400] The device receives the alert message sent by the server and notifies the user. The notification is arranged in the following pattern:

[0401] 1. Normal notification: If an abnormal transaction is detected, the user will be notified of the details.

[0402] 2. Adjusted Notification: Based on the analysis results of the emotion engine, notifications are sent in a gentler tone. For example, a message like, "I'm a little worried, are you okay?"

[0403] 3. Recommend notifying family: If the user is feeling anxious or confused, the option to recommend notifying family will be automatically selected.

[0404] The user checks the alert message and selects whether to notify their family. If the user allows notification to their family, the information is sent from the device to the server.

[0405] Additional Server Processing

[0406] After receiving the user's authorization, the server retrieves the family member's contact information from a database and sends an alert message to the family member, including details of the abnormal transaction, the user's selection, and suggestions for how to proceed.

[0407] Specific examples

[0408] The following example further illustrates the operation of the system:

[0409] Regular transactions

[0410] The server retrieves a daily utility bill (10,000 yen) from a financial institution API. This transaction is within the expected range and is not detected as an anomaly.

[0411] Unusual utility bill transactions

[0412] The server receives a utility bill transaction of 50,000 yen. This transaction is detected as an anomaly because it is far outside the predicted range (average 10,000 yen ± 30,000 yen). The emotion engine detects the user's surprise or anxiety and generates an alert in a gentle manner recommending notifying family members. The server generates an adjusted alert and sends it to the user's device. The user confirms the notification and allows the notification to family members. The server then sends the adjusted alert to the family members' devices, allowing them to respond quickly.

[0413] Large remittances

[0414] The server receives a remittance transaction of 2 million yen. Because this transaction exceeds the threshold (1 million yen), it is automatically detected as an anomaly. The emotion engine analyzes the user's emotions and generates appropriate notification content. The server generates an alert with details of the anomalous transaction and sends it to the device. If the user does not recognize the transaction, they can allow notification to family members. The server sends a notification to the family member's device, and action is taken to take measures.

[0415] Prompt Sentence Examples

[0416] By inputting the following prompts into the generative AI model, we can build specific system functions:

[0417] "Create a system that collects transaction information from senior citizens' financial accounts, detects abnormal transactions, and notifies them. Add a function that recognizes the user's emotions and adjusts the content of notifications accordingly."

[0418] This system allows seniors to manage transaction anomalies themselves while ensuring their financial security. The introduction of an emotion engine also reduces the psychological burden on users, allowing family members to intervene quickly if necessary, further enhancing the security of seniors' daily lives.

[0419] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0420] Step 1:

[0421] The server obtains transaction information for the elderly person's financial account through the financial institution's API. The input is the request information to the API, and the output is the obtained transaction details. Specifically, the server sends a request to the endpoint "https: / / api.bank.com / transactions" and receives the transaction data in JSON format.

[0422] Step 2:

[0423] The server saves the acquired transaction information in a database. The input is the transaction details data acquired in step 1, and the output is the transaction record saved in the database. Specifically, the server uses the INSERT statement to insert the transaction data it received into the "transactions" table in the database called "transaction_db".

[0424] Step 3:

[0425] The server analyzes the transaction information and detects abnormal transactions. The input to this step is the transaction data stored in the database. The output is the abnormal transaction detection results. The server processes the data in the following steps:

[0426] 1. Fixed cost anomaly detection: The server calculates the average and standard deviation of fixed costs such as utility bills and mobile phone bills based on past transaction data. Specifically, it executes the SQL query "SELECT AVG(amount), STDDEV(amount) FROM transactions WHERE type='utility'" to obtain the results. If the newly acquired transaction amount exceeds the predicted range of the average value ±3 x standard deviation, it is detected as an anomaly.

[0427] 2. Detection of large transactions: If a large transaction that exceeds a set threshold (for example, 1 million yen) occurs, the server detects it as an anomaly.

[0428] Step 4:

[0429] If an anomaly is detected, the server generates an alert message. The input to this step is the abnormal transaction detection result obtained in step 3. The output is the generated alert message. Specifically, the server generates the message "Your electricity bill this month is unusually high. Please review it."

[0430] Step 5:

[0431] The emotion engine recognizes emotions in real time from the user's facial expressions, voice, text input, etc. The inputs for this step are the user's facial images, voice data, and text input obtained from the device. The output is recognized emotion data. Specifically, the device captures the user's facial expressions using the camera and analyzes them using an emotion recognition model (e.g., Google Cloud Vision API). It also recognizes audio from the microphone and analyzes the audio data using an emotion analysis API (e.g., IBM Watson Tone Analyzer).

[0432] Step 6:

[0433] The emotion engine adjusts the alert message based on the recognized emotion data. The input to this step is the emotion data obtained in step 5 and the alert message generated in step 4. The output is the adjusted alert message. Specifically, the server changes the message to a gentler one, such as "Your electricity bill seems a little high. Please check it," based on feedback from the emotion engine.

[0434] Step 7:

[0435] The device notifies the user of the adjusted alert message. The input to this step is the alert message adjusted in step 6. The output is a notification to the user. Specifically, the device uses a push notification to display the message "I'm a little worried, are you okay?"

[0436] Step 8:

[0437] The user checks the alert message and selects whether to notify family members. The input to this step is the alert message displayed in step 7. The output is the user's selection information. Specifically, the user clicks the "Notify family members" button displayed on the terminal.

[0438] Step 9:

[0439] The server receives the user's permission information and sends an alert message to the family. The input to this step is the user's selection information obtained in step 8. The output is an alert message to the family. Specifically, the server executes the SQL query "SELECT contact_info FROM family_contacts WHERE user_id='12345'" to obtain the family's contact information and sends a message via email or SMS saying, "Your parent has an unusual transaction. Please check on them."

[0440] (Application example 2)

[0441] 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."

[0442] In modern society, there has been an increase in fraudulent access to elderly people's financial accounts and fraudulent transactions. This not only threatens the financial security of elderly people, but also increases psychological burdens such as anxiety and confusion. Current systems are capable of detecting and notifying abnormal transactions, but they are unable to respond appropriately based on the user's emotions, which can cause excessive stress to the user. In addition, notification to family members is not incorporated into the process, resulting in a lack of cooperation between elderly people and their families.

[0443] 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.

[0444] In this invention, the server includes: [means for acquiring transaction information for the elderly person's financial account;] [means for analyzing the acquired transaction information and detecting abnormalities;] [means for notifying the user of the detected abnormality, adjusting the content of the notification using emotion analysis means that recognizes the user's emotion, and requesting permission for notifying family members; and [means for sending notifications to family members based on the user's permission.] This ensures the safety of the elderly person's financial account while realizing appropriate notifications that take emotion into consideration, and enables the elderly person and their family members to intervene quickly if necessary.

[0445] "Elderly" refers to people over a certain age, primarily those considered at high risk of financial fraud and unauthorized access.

[0446] "Financial account" refers to an account opened with a bank or financial institution for the purpose of depositing money or conducting transactions.

[0447] "Transaction Information" refers to data regarding transaction dates, amounts, types, and counterparties obtained through a financial institution's API.

[0448] "Analysis" refers to the process of evaluating acquired transaction information and detecting anomalies based on patterns and regularities.

[0449] An "abnormal transaction" refers to a transaction that is deemed to be unnatural and different from the norm, and requires safety confirmation.

[0450] "User" refers to the elderly person who uses this system or someone acting on their behalf.

[0451] "Emotion analysis means" refers to technology for recognizing a user's psychological state from facial expressions, voice, text input, etc.

[0452] "Adjusting notification content" refers to softening or emphasizing alert or notification content based on the user's current emotional state.

[0453] "Permission" refers to the act of a user deciding whether or not to send notifications to family members.

[0454] "Family" refers to the user's children, siblings, and other close relatives who are responsible for providing support as needed.

[0455] A "server" refers to a computer system that primarily acquires, analyzes, stores, and notifies data.

[0456] To implement this invention, it is necessary to build a system that monitors transaction information in elderly people's financial accounts, detects abnormal transactions, and notifies users. The system mainly consists of a server, terminals, a sentiment analysis engine, and users.

[0457] Server Processing

[0458] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database.

[0459] The server then analyzes the transaction information and detects abnormal transactions. Transaction anomalies are defined as follows:

[0460] 1. Fixed cost anomalies: Based on past transaction data, the average value and standard deviation of fixed costs such as utility and communication fees are calculated. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ± 3 × standard deviation), it is detected as an anomaly.

[0461] 2. Large Transactions: If a large transaction occurs that exceeds a set threshold, it is detected as an anomaly.

[0462] When an anomaly is detected, the server generates an alert message, which includes details of the anomalous transaction, the type of anomaly, and a prompt to notify family members. Furthermore, a sentiment analysis engine recognizes the user's emotions and tailors the notification accordingly.

[0463] Sentiment analysis engine processing

[0464] The sentiment analysis engine recognizes emotions from the user's facial expressions, voice, text input, etc. The sentiment analysis engine performs the following tasks:

[0465] 1. Emotion recognition: Recognize the user's emotions in real time through the device's camera, microphone, and input devices.

[0466] 2. Sentiment analysis: Send feedback to the server based on the recognized emotion (surprise, anxiety, confusion, etc.).

[0467] 3. Notification adjustment: If the user feels anxious or confused, we recommend notifying their family members. We may also change the notification content to be more gentle.

[0468] Terminal handling

[0469] The device receives the alert message sent by the server and notifies the user. The notification is coordinated as follows:

[0470] 1. Normal notification: If an abnormal transaction is detected, the user will be notified with detailed information.

[0471] 2. Tailored notifications: Based on the results of the sentiment analysis engine, notifications are sent in a gentler tone. For example, a message like, "I'm a little worried, are you okay?"

[0472] 3. Recommend notifying family: If the user is feeling anxious or confused, the option to recommend notifying family will be automatically selected.

[0473] The user checks the alert message and selects whether to notify their family. If the user allows notification to their family, the information is sent from the device to the server.

[0474] Additional Server Processing

[0475] After receiving the user's authorization, the server retrieves the family member's contact information from a database and sends an alert message to the family member, including details of the abnormal transaction, the user's selection, and suggestions for how to proceed.

[0476] Specific examples

[0477] For example, if a user's transaction history shows a utility bill of 100,000 yen, it is considered to be within the normal expected range and is not detected as an anomaly. However, if a utility bill transaction of 500,000 yen occurs, it is detected as an anomaly and the user is notified. If the user is concerned, their family members will also be notified.

[0478] Prompt sentence for generative AI model

[0479] Create a program for a system that monitors elderly people's financial transactions and detects and notifies them of any abnormal transactions. The system will include a server that collects and analyzes transaction information and detects abnormalities, adjusts notifications based on emotion recognition, and notifies family members. Please also specify the specific names of the APIs and libraries to be used. For example, transaction information could be obtained from a financial institution's API, and notifications could be sent using a communication API.

[0480] This system allows seniors to manage transaction anomalies themselves while ensuring their financial security. The adoption of a sentiment analysis engine also reduces the user's psychological burden, allowing family members to quickly intervene when necessary, further enhancing the security of seniors' daily lives.

[0481] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0482] Step 1:

[0483] The server obtains transaction information for the elderly person's financial account through the financial institution's API.

[0484] (Input) Transaction information (transaction date, amount, type, and counterparty) from the financial institution's API

[0485] (Output) Save the acquired transaction information in the database

[0486] (Specific operation) The server periodically sends API requests to obtain and store transaction information.

[0487] Step 2:

[0488] The server analyzes the transaction information it acquires and detects any abnormal transactions.

[0489] (Input) Transaction information stored in the database

[0490] (Output) A list of transactions that were determined to be anomalous

[0491] (Specific operation) The server calculates the average value and standard deviation of fixed costs, and detects an anomaly when new transaction information exceeds the predicted range or when a large transaction exceeding a set threshold occurs.

[0492] Step 3:

[0493] If the server detects an abnormality, it generates an alert message.

[0494] (Input) A list of transactions that are determined to be abnormal

[0495] (Output) Alert message

[0496] (Specific Operation) The server generates an alert message containing details of the abnormal transaction and the type of abnormality.

[0497] Step 4:

[0498] The server uses a sentiment analysis engine to recognize the user's sentiment.

[0499] (Input) User facial expressions, voice, and text input obtained from the device's camera, microphone, and input device

[0500] (Output) User emotion data (surprise, anxiety, confusion, etc.)

[0501] (Specific operation) The emotion analysis engine analyzes the user's emotions in real time and feeds the results back to the server.

[0502] Step 5:

[0503] The server adjusts the notification content based on the user's emotions.

[0504] (Input) User emotion data, alert message

[0505] (Output) Adjusted alert message

[0506] (Specific operation) If the user feels anxious or confused, the server changes the notification content to softer language and generates an alert message recommending that the user notify family members.

[0507] Step 6:

[0508] The terminal receives the alert message sent from the server and notifies the user.

[0509] (Input) Adjusted alert message

[0510] (Output) User notification

[0511] (Specific operation) The device notifies the user in the form of a pop-up or a voice message.

[0512] Step 7:

[0513] The user can choose whether to notify their family.

[0514] (Input) Alert message content

[0515] (Output) Allow or deny notification to family members

[0516] (Specific operation) The user checks the alert message and selects "Notify family" or "Do not notify" via the device.

[0517] Step 8:

[0518] The server sends notifications to family members based on the user's permission.

[0519] (Input) User notification permission information

[0520] (Output) Alert message sent to family members

[0521] (Specific Operation) The server retrieves the family member's contact information from the database and sends an alert message containing details of the abnormal transaction to the family member's terminal.

[0522] This way, seniors' financial accounts can be kept safe while still providing timely and sensitive notifications, allowing seniors and their families to intervene quickly if needed.

[0523] 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.

[0524] 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.

[0525] 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.

[0526] [Second embodiment]

[0527] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0528] 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.

[0529] 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).

[0530] 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.

[0531] 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.

[0532] 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).

[0533] 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.

[0534] 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.

[0535] 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.

[0536] 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.

[0537] 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.

[0538] 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."

[0539] To implement this invention, a system is constructed that monitors transaction information in elderly people's financial accounts, detects abnormal transactions, and notifies them. The system is primarily composed of a server, terminals (such as smartphones and PCs), and users.

[0540] Server Processing

[0541] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database.

[0542] The server then analyzes the transaction information and detects abnormal transactions. Transaction anomalies are defined as follows:

[0543] 1. Fixed cost anomalies: Based on past transaction data, the average value and standard deviation of fixed costs such as utility bills and mobile phone bills are calculated. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ± 3 × standard deviation), it is detected as an anomaly.

[0544] 2. Large Transactions: If a large transaction occurs that exceeds a set threshold, it is detected as an anomaly.

[0545] If an anomaly is detected, the server sends an alert message to the user's device, containing details of the anomalous transaction and a message asking the user to confirm whether to notify family members.

[0546] Terminal handling

[0547] The terminal receives the alert message sent from the server and notifies the user. The user can then check the alert message and confirm the transaction details themselves. They can also choose whether to send notifications to their family members. If the user allows notifications to be sent to their family members, the information is sent from the terminal to the server.

[0548] User Action

[0549] The user checks the alert message displayed on the device and considers how to respond to any unusual transactions. If the transaction is suspicious, they can choose not to notify their family, but if not, it is recommended that they allow notifications. This choice is made on the device, and the result is sent to the server.

[0550] Additional Server Processing

[0551] If the user allows notifications to family members, the server will send an alert message to the family members' devices registered in advance, allowing the family members to receive detailed information about abnormal transactions and take appropriate measures.

[0552] Specific examples

[0553] Specifically, the following scenarios are possible:

[0554] 1. Normal transaction:

[0555] The server retrieves daily utility bills (10,000 yen) from the financial institution API.

[0556] This transaction is within the expected range and is not detected as an anomaly.

[0557] 2. Unusual utility bill transactions:

[0558] The server retrieves a utility bill transaction for 50,000 yen from the financial institution API.

[0559] This transaction is detected as an anomaly because it is far outside the expected range (average 10,000 yen ± 30,000 yen).

[0560] The server generates an alert and sends it to the user's terminal.

[0561] The user checks the alert on the device and allows notification to family members.

[0562] The server sends a notification to the family's device, providing them with detailed information.

[0563] 3. Large transfers:

[0564] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[0565] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[0566] The server generates an alert and notifies the user.

[0567] If the user is not aware of the incident, they can allow their family members to be notified.

[0568] The server sends a notification to the family's devices and action is taken to address the issue.

[0569] This system allows seniors to ensure their financial security while managing any irregularities in their transactions themselves, and allows family members to intervene quickly if necessary, helping to ensure the safety of seniors' daily lives.

[0570] The processing flow will be explained below.

[0571] Program processing

[0572] Server Processing

[0573] Step 1:

[0574] The server accesses the financial institution API at regular intervals (e.g., 9:00 AM every day).

[0575] The server sends a request for transaction data using authentication information for the financial institution API.

[0576] If authentication is successful, the financial institution API returns the user's transaction information.

[0577] The server stores the acquired transaction information in a database.

[0578] Step 2:

[0579] The server retrieves the latest transaction information from the database.

[0580] Filter newly acquired transaction information to identify transactions of interest.

[0581] The transactions covered include the date, transaction amount, transaction type, and counterparty.

[0582] Step 3:

[0583] The server categorizes the transaction information into fixed cost categories.

[0584] The server analyzes past transaction data to identify fixed expense categories such as utility bills and cell phone bills.

[0585] Check if the new transaction falls into the fixed cost category.

[0586] Step 4:

[0587] The server detects anomalies in fixed costs.

[0588] The server calculates the mean and standard deviation for the fixed cost categories.

[0589] If the new transaction amount exceeds the expected range (mean ± 3 x standard deviation), it is marked as an anomaly.

[0590] Step 5:

[0591] The server detects anomalies in large transactions.

[0592] Set a large amount threshold (e.g., 1 million yen) for all transactions.

[0593] If the transaction amount exceeds this threshold, it is marked as an anomaly.

[0594] Step 6:

[0595] If an anomaly is detected, the server generates an alert message.

[0596] The alert message will include details of the unusual transaction, the type of abnormality, and confirmation of permission to notify family members.

[0597] The server sends this message to the user's terminal.

[0598] Terminal handling

[0599] Step 1:

[0600] The terminal receives the alert message from the server.

[0601] Notify the user that new notifications are available.

[0602] The device will display the notification content in a pop-up or notification bar.

[0603] Step 2:

[0604] The terminal waits for the user's action.

[0605] The user opens the notification and sees the details of the alert message.

[0606] The alert message will include transaction details, a description of the anomaly, and an option to notify family members.

[0607] Step 3:

[0608] The user presses the button to allow notifications to family members.

[0609] The terminal transmits the user's authorization information to the server.

[0610] Additional Server Processing

[0611] Step 1:

[0612] The server receives the user's authorization information.

[0613] Retrieve contact information for registered family members from the database.

[0614] Step 2:

[0615] The server sends an alert message to the family.

[0616] A message is sent to the family member's device containing details of the abnormal transaction, the user's selection, and suggestions on how to proceed.

[0617] If the message is sent successfully, a record of it is saved in the database.

[0618] User Action

[0619] Step 1:

[0620] The user checks the alert message on the terminal.

[0621] Check the details of any unusual transactions to see if they apply to you.

[0622] Step 2:

[0623] The user selects whether to notify family members.

[0624] If you are not aware of the incident, allow your family to be notified.

[0625] If permission is granted, the information is sent from the terminal to the server.

[0626] Based on these specific processing steps, abnormal transactions in the elderly person's financial accounts can be monitored and necessary measures can be taken promptly.

[0627] Example 1

[0628] 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."

[0629] The increase in fraudulent and abnormal transactions in elderly people's financial accounts has created a need for detection and rapid notification. Current systems take time to acquire and analyze transaction information, and lack the means to quickly notify users and their families when an abnormality is detected. This increases the likelihood that elderly people will be exposed to financial risk, which is a major source of anxiety.

[0630] 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.

[0631] In this invention, the server includes means for periodically acquiring transaction information from the elderly person's financial account via the financial institution's API, means for storing the acquired transaction information in a database, means for analyzing the stored transaction information and detecting abnormal transactions, means for sending an alert message including details of the detected abnormal transaction to the user's terminal, means for providing an interface that allows the user to check the alert and select whether to notify family members, and means for sending a notification to the family member's terminal based on the user's permission. This makes it possible to quickly detect abnormalities in the elderly person's financial transactions and notify the user and family members in a timely manner.

[0632] "Financial institution API" stands for Application Program Interface provided by financial institutions, which is an interface used by external systems to obtain financial transaction data.

[0633] An "elderly person's financial account" refers to an account opened by an elderly person at a bank or other financial institution, through which financial transactions such as deposits, withdrawals, and transfers are conducted.

[0634] "Transaction information" is data showing the transaction history of a financial account within a specific period, and includes transaction dates, transaction amounts, transaction types, and transaction partner information.

[0635] "Database" means a computer system for systematically storing and managing acquired transaction information, allowing for easy access and analysis.

[0636] An "abnormal transaction" is a transaction that is determined to be out of the ordinary based on past transaction patterns or established thresholds, including transactions that may be fraudulent or involve abnormal amounts.

[0637] An "alert message" is a notification sent to a user when an abnormal transaction is detected, and contains details of the transaction and information urging the user to take action.

[0638] "Users" are those who use this system to receive notifications of abnormal transactions, and primarily refer to elderly people and their families.

[0639] "Terminal" refers to a device used by a user, such as a smartphone, PC, or tablet, that receives and displays alert messages from the server.

[0640] "Family devices" refer to devices such as smartphones, PCs, and tablets owned by the user's family, and alert messages are sent to these devices based on the user's permission.

[0641] To implement this invention, it is necessary to build a system that monitors transaction information in elderly people's financial accounts, detects abnormal transactions, and notifies users. This system is primarily composed of a server, terminals (such as smartphones and PCs), and users.

[0642] Server Processing

[0643] The server periodically retrieves transaction information from the elderly person's financial account via the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The retrieved information is stored in a database. Specifically, the server sends an API request at a set time each day to retrieve daily transaction data and record it in the database.

[0644] The server then analyzes the stored transaction information and detects abnormal transactions. Transaction abnormalities are defined in two patterns: fixed cost abnormalities and large transaction amounts. For fixed cost abnormalities, the server calculates the average value and standard deviation of fixed costs such as utility bills and mobile phone bills based on past transaction data. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ±3 x standard deviation), it is detected as an abnormality. On the other hand, for large transaction amounts, an abnormality is detected if a transaction that exceeds the set threshold occurs. If an abnormality is detected, the server sends an alert message to the user's device.

[0645] Terminal handling

[0646] The terminal receives the alert message sent from the server and is responsible for notifying the user. The user can then review the alert message and investigate the transaction details themselves. The interface displayed on the terminal includes an option to select whether to send notifications to family members. If the user allows notifications to be sent to family members, the terminal sends the selection to the server.

[0647] User Action

[0648] The user checks the alert message displayed on the device. If the transaction is recognizable, they can choose not to notify their family. If they are not recognizable, they can choose to allow notification to their family. This selection is sent to the server via the device.

[0649] Additional Server Processing

[0650] If the user allows notification to family members, the server will send an alert message to the pre-registered family members' devices, allowing the family members to receive detailed information about the abnormal transaction and take appropriate measures.

[0651] Specific examples

[0652] Specifically, the following scenarios are possible:

[0653] 1. Normal transaction:

[0654] The server retrieves daily utility bills (10,000 yen) from the financial institution's API.

[0655] This transaction is within the expected range and is not detected as an anomaly.

[0656] 2. Unusual utility bill transactions:

[0657] The server retrieves a utility bill transaction for 50,000 yen from the financial institution's API.

[0658] This transaction is detected as an anomaly because it is far outside the predicted range (average 10,000 yen ± 3 x 5,000 yen).

[0659] The server generates an alert message and sends it to the user's terminal.

[0660] The user confirms the alert message and allows notification to family members.

[0661] The server sends a notification to the family's device, providing them with detailed information.

[0662] 3. Large transfers:

[0663] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[0664] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[0665] The server generates an alert message and notifies the user.

[0666] If the user is not aware of the incident, they can allow their family members to be notified.

[0667] The server sends a notification to the family's devices and steps are taken to take action.

[0668] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0669] Step 1:

[0670] The server retrieves the transaction information

[0671] Input: Request to financial institution API

[0672] Specific operation: The server sends a request to the financial institution's API at a specific time interval (e.g., every day at midnight), using the authentication token to access the API and requesting transaction information for the target senior citizen's financial account.

[0673] Output: Transaction information (JSON format)

[0674] Specific data processing: Transaction information is returned from financial institutions in JSON format, including the transaction date, transaction amount, transaction type (deposit, withdrawal, transfer, etc.), and counterparty.

[0675] Step 2:

[0676] The server saves the transaction information it has acquired into a database.

[0677] Input: Retrieved transaction information (JSON format)

[0678] Specific operation: The server analyzes the acquired transaction information and extracts necessary fields, such as transaction date, transaction amount, and transaction type.

[0679] Output: Stored transaction information (specific table in database)

[0680] Specific data processing: The extracted data is properly saved in the transactions table of the database.

[0681] Step 3:

[0682] The server analyzes the transaction information and detects abnormal transactions.

[0683] Input: Historical transaction data in the database, retrieved new transaction data

[0684] Specific operation: The server calculates the average and standard deviation of fixed costs based on past transaction data. For example, using utility bill data from the past three months, the server calculates the average (e.g., 10,000 yen) and standard deviation (e.g., 5,000 yen).

[0685] Output: Abnormal transactions (list format)

[0686] Specific data calculation: If a new transaction exceeds the average value ±3 × standard deviation, the transaction is added to the list as an anomaly. Large transactions exceeding a set threshold (e.g., 1 million yen) are also added to the list as an anomaly.

[0687] Step 4:

[0688] The server generates an alert for abnormal transactions and sends it to the terminal.

[0689] Input: Detected abnormal transactions (list format)

[0690] What happens: The server creates an alert message with details of the anomalous transaction. For example, it generates a message like, "A utility bill transaction of 50,000 yen has been detected as anomalous. Does this transaction seem familiar?"

[0691] Output: Alert message (push notification to device)

[0692] Specific data processing: The generated alert message is sent to the user's device as a push notification.

[0693] Step 5:

[0694] The device receives the alert and notifies the user.

[0695] Input: The alert message sent from the server

[0696] Specific behavior: The device will display the received alert message as a pop-up notification on the screen. When the user taps the pop-up, the app will open and display more information.

[0697] Output: Notification to the user (display on screen)

[0698] Specific data display: Detailed information about the alert is displayed on the screen, and the user is given the option to "notify family members."

[0699] Step 6:

[0700] The user responds to the alert and sends their choice from the device to the server

[0701] Input: Alert message confirmation result (whether to notify family members)

[0702] Specific operation: The user checks the details of the alert and selects "Notify family members" or "Don't notify family members." The selection is sent from the device to the server.

[0703] Output: Selection (whether to notify family members)

[0704] Specific data transmission: User selections are transmitted to the server.

[0705] Step 7:

[0706] The server sends a notification to the family

[0707] Input: User's selection result (allow notification to family members)

[0708] Specific operation: The server checks the pre-registered contact information of family members, creates an alert message, and sends it to the family members' devices. For example, it generates a message such as, "An abnormal transaction has been detected in your parent's account. Please check the details."

[0709] Output: Alert message to family members (push notification to device or email)

[0710] Specific data transmission: The generated alert message will be sent to the family's device so that the family can check the detailed information.

[0711] (Application example 1)

[0712] 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."

[0713] Ensuring the security of transactions in elderly people's financial accounts is important for preventing financial damage, but it is difficult for elderly people themselves to constantly monitor their transactions for abnormalities. There is also a need for a method to quickly and reliably notify family members when abnormal transactions occur. Conventional systems lack the ability to detect abnormalities in real time and provide immediate notification, preventing family members from responding quickly. A new system is needed to resolve these issues and provide peace of mind to elderly people and their families.

[0714] 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.

[0715] In this invention, the server includes: [means for acquiring transaction information on the elderly person's financial account]; [means for analyzing the acquired transaction information and detecting abnormalities]; [means for notifying the user of detected abnormalities and requesting permission for notifying family members]; [means for sending notifications to family members based on the user's permission]; [means for sending notifications in real time when an abnormal transaction is detected]; and [means for sending similar alerts to family members]. This enables real-time detection of abnormal transactions and prompt notification, enabling the elderly person and their family members to prevent financial damage. It also provides a system that reduces the burden on the elderly person themselves and allows family members to respond quickly.

[0716] The term "elderly" generally refers to people aged 60 or over, especially those in need of social and medical support.

[0717] "Financial Account" means a deposit account or other asset management account opened by a customer at a financial institution for the purpose of managing assets.

[0718] "Transaction information" refers to detailed data regarding various financial transactions, such as deposits, withdrawals, transfers, and payments, in a financial account.

[0719] "Real-time" means that data is collected, analyzed, and communicated almost immediately, with very little time delay.

[0720] "Abnormal transactions" refer to transactions that deviate significantly from normal trading patterns, including large transactions and transactions that exceed expected ranges.

[0721] A "server" refers to a computer system that provides processing power and data to multiple clients over a network.

[0722] "Database" means a system for systematically storing and managing information in digital form.

[0723] "Notification" means a message or alert sent to inform a user or interested party of specific information.

[0724] An "alert" refers to a notification method that immediately warns the user when an abnormality or a condition requiring attention occurs.

[0725] "Family" refers to a person related by blood or marriage to an older person and who is responsible for assisting and supporting the older person in managing their life, health, and property.

[0726] To implement this invention, it is necessary to build a system consisting of a server, terminals (such as smartphones or PCs), and users. The main function of the system is to provide a mechanism for monitoring transaction information in elderly people's financial accounts and detecting and notifying abnormal transactions.

[0727] Server Processing

[0728] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and counterparty. The obtained information is stored in a database. The server uses the NumPy library to calculate the mean and standard deviation from past transaction data. It defines the mean value and prediction range (e.g., mean value ± 3 × standard deviation) of fixed costs, and detects newly obtained transactions that exceed the prediction range as anomalous transactions. Large transactions whose transaction amount exceeds a set threshold are also detected as anomalous. When an anomaly is detected, the server uses Pushbullet to send an alert message via push notification to the user's device.

[0729] Terminal handling

[0730] The device receives the alert message sent from the server and notifies the user. The user can check the alert message and check the transaction details themselves. The alert message includes details of the abnormal transaction and an option to notify family members. If the user allows notification to family members, the device sends the information to the server.

[0731] User Action

[0732] The user checks the alert message displayed on the device and decides whether or not the transaction is recognizable. If not, the user is encouraged to allow notifications to family members. This selection is made on the device, and the result is sent to the server. Based on the user's permission, the server can send alert messages to the devices of family members who have been registered in advance.

[0733] Specific examples

[0734] For example, if the average utility bill for the past six months was 10,000 yen with a standard deviation of 5,000 yen, and a new transaction of 20,000 yen occurs, the server will detect this as an anomaly. In this case, the server will send an alert to the user's smartphone stating, "Anomalous transaction detected: A utility bill transaction of 20,000 yen occurred on 2023-10-01." If the user allows notifications, similar alerts will also be sent to the smartphones of family members.

[0735] Example prompts for generative AI models

[0736] "We will design a system that monitors transaction information for elderly people's financial accounts, detects abnormal transactions, and notifies users. An abnormal transaction in the transaction data we received (for example, the average utility bill for the past six months is 10,000 yen, with a standard deviation of 5,000 yen) is one that exceeds the average value ±3 x standard deviation. We will set it up so that a push notification is sent stating, 'Abnormal transaction detected: A utility bill transaction of 50,000 yen occurred on 2023-10-01.' Please create a specific program based on this content."

[0737] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0738] Step 1:

[0739] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. The information obtained includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The input for this process is the API call, and the output is transaction information data in JSON format. The server saves the obtained data in a database.

[0740] Step 2:

[0741] The server extracts historical transaction data from the stored transaction information and calculates the mean and standard deviation of a particular fixed cost (e.g., utility bills). It uses the historical transaction data (in the database) as input and gets the mean and standard deviation as output. It uses the NumPy library to perform statistical operations.

[0742] Step 3:

[0743] The server analyzes newly acquired transaction data and determines whether the transaction falls within the expected range based on the calculated mean and standard deviation. The input is the latest transaction data, and the output is a flag indicating whether the transaction is "anomalous" or "normal." For example, if the amount of a new transaction exceeds the mean value ± 3 x standard deviation, the server detects the transaction as "anomalous."

[0744] Step 4:

[0745] The server uses Pushbullet's API to send an alert message to the user's device for transactions detected as abnormal. The alert includes details of the abnormal transaction (transaction date, amount, and counterparty) and an option to notify family members. The input is the abnormal transaction data, and the output is the alert message.

[0746] Step 5:

[0747] The device receives the alert message sent from the server and notifies the user. The input is the alert message, and the output is a push notification to the user. The user checks the content of the alert message on the device.

[0748] Step 6:

[0749] The user checks the alert message displayed on the terminal and selects how to respond to the abnormal transaction. Specifically, the user selects whether to notify family members. The input is the notified alert information, and the output is the user's selection.

[0750] Step 7:

[0751] When the user allows notifications to family members, the device sends the selection information to the server. The input is the user's selection, and the output is the notification permission data for the server.

[0752] Step 8:

[0753] Based on the user's permission, the server sends an alert message to pre-registered family members' devices. This allows family members to receive information about abnormal transactions and take prompt action. The input is notification permission data from the user, and the output is an alert message for the family members.

[0754] By following the above steps, financial transactions of elderly people can be monitored in real time, and if any abnormalities are detected, the user and their family can be notified promptly, thereby preventing financial damage.

[0755] 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.

[0756] To implement this invention, we will build a system that monitors transaction information on elderly people's financial accounts, detects and notifies abnormal transactions, and combines it with an emotion engine that recognizes user emotions. The system is primarily composed of a server, a terminal (such as a smartphone or PC), the emotion engine, and the user.

[0757] Server Processing

[0758] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database.

[0759] The server then analyzes the transaction information and detects abnormal transactions. Transaction anomalies are defined as follows:

[0760] 1. Fixed cost anomalies: Based on past transaction data, the average value and standard deviation of fixed costs such as utility bills and mobile phone bills are calculated. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ± 3 × standard deviation), it is detected as an anomaly.

[0761] 2. Large Transactions: If a large transaction occurs that exceeds a set threshold, it is detected as an anomaly.

[0762] When an anomaly is detected, the server generates an alert message, which includes details of the anomalous transaction, the type of anomaly, and a prompt to notify family members. Furthermore, an emotion engine recognizes the user's emotions and tailors the notification accordingly.

[0763] Emotion engine processing

[0764] The emotion engine recognizes emotions from the user's facial expressions, voice, text input, etc. The emotion engine performs the following tasks:

[0765] 1. Emotion recognition: Recognizes user emotions in real time through the device's camera, microphone, and keyboard input.

[0766] 2. Sentiment analysis: Send feedback to the server based on the recognized emotion (surprise, anxiety, confusion, etc.).

[0767] 3. Notification adjustment: If the user feels anxious or confused, we recommend notifying their family members. We may also change the notification content to be more gentle.

[0768] Terminal handling

[0769] The device receives the alert message sent by the server and notifies the user. The notification is coordinated as follows:

[0770] 1. Normal notification: If an abnormal transaction is detected, the user will be notified with detailed information.

[0771] 2. Adjusted Notification: Based on the analysis results of the emotion engine, notifications are sent in a gentler tone. For example, a message like, "I'm a little worried, are you okay?"

[0772] 3. Recommend notifying family: If the user is feeling anxious or confused, the option to recommend notifying family will be automatically selected.

[0773] The user checks the alert message and selects whether to notify their family. If the user allows notification to their family, the information is sent from the device to the server.

[0774] Additional Server Processing

[0775] After receiving the user's authorization, the server retrieves the family member's contact information from a database and sends an alert message to the family member, including details of the abnormal transaction, the user's selection, and suggestions for how to proceed.

[0776] Specific examples

[0777] Specifically, the following scenarios are possible:

[0778] 1. Normal transaction:

[0779] The server retrieves daily utility bills (10,000 yen) from the financial institution API.

[0780] This transaction is within the expected range and is not detected as an anomaly.

[0781] 2. Unusual utility bill transactions:

[0782] The server retrieves a utility bill transaction for 50,000 yen from the financial institution API.

[0783] This transaction is detected as an anomaly because it is far outside the expected range (average 10,000 yen ± 30,000 yen).

[0784] The emotion engine detects the user's surprise or anxiety and generates a gentle alert recommending notifying family members.

[0785] The server generates the adjusted alert and sends it to the user's device.

[0786] The user confirms the notification and allows notification to family members.

[0787] The server sends tailored alerts to family devices, allowing them to respond quickly.

[0788] 3. Large transfers:

[0789] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[0790] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[0791] The emotion engine analyzes the user's emotions and generates appropriate notification content.

[0792] The server generates an alert containing details of the abnormal transaction and notifies the user.

[0793] If the user is not aware of the incident, they can allow their family members to be notified.

[0794] The server sends a notification to the family's devices and action is taken to address the issue.

[0795] This system allows seniors to manage transaction anomalies themselves while ensuring their financial security. The introduction of an emotion engine also reduces the psychological burden on users, allowing family members to intervene quickly if necessary, further enhancing the security of seniors' daily lives.

[0796] The processing flow will be explained below.

[0797] Server Processing

[0798] Step 1:

[0799] The server accesses the financial institution API at regular intervals (e.g., 9:00 AM every day).

[0800] The server sends a request for transaction data using authentication information for the financial institution API.

[0801] If authentication is successful, the financial institution API returns the user's transaction information.

[0802] The server stores the acquired transaction information in a database.

[0803] Step 2:

[0804] The server retrieves the latest transaction information from the database.

[0805] The server filters the newly acquired transaction information to identify eligible transactions.

[0806] The transactions covered include the date, transaction amount, transaction type, and counterparty.

[0807] Step 3:

[0808] The server categorizes the transaction information into fixed cost categories.

[0809] The server analyzes past transaction data to identify fixed expense categories such as utility bills and cell phone bills.

[0810] Check if the new transaction falls into the fixed cost category.

[0811] Step 4:

[0812] The server detects anomalies in fixed costs.

[0813] The server calculates the mean and standard deviation for the fixed cost categories.

[0814] If the new transaction amount exceeds the expected range (mean ± 3 x standard deviation), it is marked as an anomaly.

[0815] Step 5:

[0816] The server detects anomalies in large transactions.

[0817] The server sets a large amount threshold (e.g., 1 million yen) for all transactions.

[0818] If the transaction amount exceeds this threshold, it is marked as an anomaly.

[0819] Step 6:

[0820] If an anomaly is detected, the server generates an alert message.

[0821] The alert message will include details of the abnormal transaction, the type of abnormality, and permission to notify family members.

[0822] Emotion engine processing

[0823] Step 1:

[0824] The emotion engine captures the user's facial expressions and voice in real time.

[0825] Data is collected through cameras and microphones.

[0826] Step 2:

[0827] The emotion engine analyzes the acquired data and recognizes the user's emotions.

[0828] The emotion engine uses facial recognition algorithms and voice analysis to identify the user's emotions.

[0829] If the user's emotion corresponds to anxiety, surprise, confusion, etc., the information is sent to the server.

[0830] Step 3:

[0831] The emotion engine sends feedback to the server to adjust the notification content.

[0832] If the user is feeling anxious or confused, change the notification message to a more gentle one.

[0833] An option to recommend notifying family members will also be added.

[0834] Terminal handling

[0835] Step 1:

[0836] The terminal receives the alert message from the server.

[0837] Notify the user that new notifications are available.

[0838] The device will display the notification content in a pop-up or notification bar.

[0839] Step 2:

[0840] The terminal waits for the user's action.

[0841] The user opens the notification to view the details of the alert message.

[0842] The message will include transaction details, a description of the anomaly, and an option to notify family members.

[0843] Step 3:

[0844] When the user presses the button to allow notifications to family members,

[0845] The terminal transmits the user's authorization information to the server.

[0846] Additional Server Processing

[0847] Step 1:

[0848] The server receives the user's authorization information.

[0849] The server retrieves contact information for registered family members from the database.

[0850] Step 2:

[0851] The server sends an alert message to the family.

[0852] A message is sent to the family member's device containing details of the abnormal transaction, the user's selection, and suggestions on how to proceed.

[0853] If the message is sent successfully, a record of it is saved in the database.

[0854] Specific processing examples

[0855] Specifically, the following scenarios are possible:

[0856] 1. Normal transaction:

[0857] The server retrieves daily utility bills (10,000 yen) from the financial institution API.

[0858] This transaction is within the expected range and is not detected as an anomaly.

[0859] 2. Unusual utility bill transactions:

[0860] The server retrieves a utility bill transaction for 50,000 yen from the financial institution API.

[0861] This transaction is detected as an anomaly because it is far outside the expected range (average 10,000 yen ± 30,000 yen).

[0862] The emotion engine detects the user's surprise or anxiety and generates a gentle alert recommending notifying family members.

[0863] The server generates the adjusted alert and sends it to the user's device.

[0864] The user confirms the notification and allows notification to family members.

[0865] The server sends tailored alerts to family devices, allowing them to respond quickly.

[0866] 3. Large transfers:

[0867] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[0868] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[0869] The emotion engine analyzes the user's emotions and generates appropriate notification content.

[0870] The server generates an alert containing details of the abnormal transaction and notifies the user.

[0871] If the user is not aware of the incident, they can allow their family members to be notified.

[0872] The server sends notifications to family members' devices so that countermeasures can be taken quickly.

[0873] Example 2

[0874] 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."

[0875] There is a need for a method and system that can respond quickly and appropriately when abnormal financial transactions occur among elderly people. Furthermore, it is necessary to consider the user's feelings when notifying them of abnormal transactions. Conventional systems have had problems such as causing unnecessary anxiety to users and delaying notification to family members. The purpose of this invention is to solve these problems and simultaneously ensure the financial security and psychological peace of mind of elderly people.

[0876] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for acquiring transaction information of the elderly person's financial account, a means for analyzing the acquired transaction information and detecting abnormalities, and a means for recognizing the user's emotions and adjusting the notification content. This makes it possible to notify the user of abnormal transactions in a manner that takes into consideration the user's emotions, and also to notify family members at an appropriate time.

[0877] "Elderly" refers to people who are older, especially those aged 65 or older, and who require special consideration in financial transactions and daily life.

[0878] "Financial account" is a general term for deposit accounts, savings accounts, fixed-term deposit accounts, etc. opened at banks, credit unions, and other financial institutions, and is a basic means for recording and managing financial transactions.

[0879] "Transaction information" refers to detailed data relating to financial transactions such as deposits, withdrawals, transfers, and debits related to a financial account, including the date, amount, type, and counterparty of the transaction.

[0880] "Abnormal transactions" refer to transactions that significantly deviate from normal trading patterns, including, for example, transactions that exceed the mean and standard deviation of a particular fixed cost or large transactions that exceed a set threshold.

[0881] "Emotion engine" is a general term for software or hardware that recognizes and analyzes emotions in real time from a user's facial expressions, voice, text input, etc.

[0882] "Adjusting the notification content" refers to a process of changing the wording and content of the notification message depending on the user's emotional state, so as not to make the user feel excessively anxious.

[0883] "Family" refers to a person who has a certain kinship relationship with the user and who has the role of supporting the user in emergencies or abnormal situations.

[0884] "Server" means the main computer system that manages the entire system and acquires, analyzes, and stores transaction information, and generates and transmits notifications.

[0885] A "terminal" refers to a device such as a smartphone or PC that is directly operated by a user and that receives notifications sent from a server.

[0886] "Notification" refers to a message sent by the system to the user or family member informing them of an abnormal transaction, and may be sent in the form of voice, text, push notification, etc.

[0887] To implement this invention, it is necessary to build a system that monitors transaction information on elderly people's financial accounts, detects abnormal transactions, and notifies them, as well as an emotion engine that recognizes user emotions. This system is primarily composed of a server, a terminal (such as a smartphone or PC), an emotion engine, and a user.

[0888] Server Processing

[0889] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, transaction amount, transaction type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database on the server. The server analyzes this data and runs an algorithm to detect abnormal transactions.

[0890] Specifically, the server detects abnormal transactions through the following steps:

[0891] 1. Fixed cost anomaly detection: Based on past transaction data, the average value and standard deviation of fixed costs such as utility bills and mobile phone bills are calculated. If the newly acquired transaction amount exceeds the predicted range of the average value ±3 x standard deviation, it is detected as an anomaly.

[0892] 2. Large transaction detection: The server detects an abnormality when a large transaction that exceeds a set threshold occurs.

[0893] When an anomaly is detected, the server sends feedback to the emotion engine, and also generates alert messages and adjusts the notification content as needed.

[0894] Emotion engine processing

[0895] The emotion engine recognizes emotions in real time from the user's facial expressions, voice, text input, etc. For this purpose, the device requires input devices such as a camera and microphone. The emotion engine provides the following functions:

[0896] 1. Emotion recognition: Recognizes the user's emotions through the device's camera, microphone, and keyboard input.

[0897] 2. Sentiment analysis: Send feedback to the server based on the recognized emotion (surprise, anxiety, confusion, etc.).

[0898] 3. Notification adjustment: If the user feels anxious or confused, we recommend notifying their family members. We may also change the notification content to be more gentle.

[0899] Terminal handling

[0900] The device receives the alert message sent by the server and notifies the user. The notification is arranged in the following pattern:

[0901] 1. Normal notification: If an abnormal transaction is detected, the user will be notified of the details.

[0902] 2. Adjusted Notification: Based on the analysis results of the emotion engine, notifications are sent in a gentler tone. For example, a message like, "I'm a little worried, are you okay?"

[0903] 3. Recommend notifying family: If the user is feeling anxious or confused, the option to recommend notifying family will be automatically selected.

[0904] The user checks the alert message and selects whether to notify their family. If the user allows notification to their family, the information is sent from the device to the server.

[0905] Additional Server Processing

[0906] After receiving the user's authorization, the server retrieves the family member's contact information from a database and sends an alert message to the family member, including details of the abnormal transaction, the user's selection, and suggestions for how to proceed.

[0907] Specific examples

[0908] The following example further illustrates the operation of the system:

[0909] Regular transactions

[0910] The server retrieves a daily utility bill (10,000 yen) from a financial institution API. This transaction is within the expected range and is not detected as an anomaly.

[0911] Unusual utility bill transactions

[0912] The server receives a utility bill transaction of 50,000 yen. This transaction is detected as an anomaly because it is far outside the predicted range (average 10,000 yen ± 30,000 yen). The emotion engine detects the user's surprise or anxiety and generates an alert in a gentle manner recommending notifying family members. The server generates an adjusted alert and sends it to the user's device. The user confirms the notification and allows the notification to family members. The server then sends the adjusted alert to the family members' devices, allowing them to respond quickly.

[0913] Large remittances

[0914] The server receives a remittance transaction of 2 million yen. Because this transaction exceeds the threshold (1 million yen), it is automatically detected as an anomaly. The emotion engine analyzes the user's emotions and generates appropriate notification content. The server generates an alert with details of the anomalous transaction and sends it to the device. If the user does not recognize the transaction, they can allow notification to family members. The server sends a notification to the family member's device, and action is taken to take measures.

[0915] Prompt Sentence Examples

[0916] By inputting the following prompts into the generative AI model, we can build specific system functions:

[0917] "Create a system that collects transaction information from senior citizens' financial accounts, detects abnormal transactions, and notifies them. Add a function that recognizes the user's emotions and adjusts the content of notifications accordingly."

[0918] This system allows seniors to manage transaction anomalies themselves while ensuring their financial security. The introduction of an emotion engine also reduces the psychological burden on users, allowing family members to intervene quickly if necessary, further enhancing the security of seniors' daily lives.

[0919] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0920] Step 1:

[0921] The server obtains transaction information for the elderly person's financial account through the financial institution's API. The input is the request information to the API, and the output is the obtained transaction details. Specifically, the server sends a request to the endpoint "https: / / api.bank.com / transactions" and receives the transaction data in JSON format.

[0922] Step 2:

[0923] The server saves the acquired transaction information in a database. The input is the transaction details data acquired in step 1, and the output is the transaction record saved in the database. Specifically, the server uses the INSERT statement to insert the transaction data it received into the "transactions" table in the database called "transaction_db".

[0924] Step 3:

[0925] The server analyzes the transaction information and detects abnormal transactions. The input to this step is the transaction data stored in the database. The output is the abnormal transaction detection results. The server processes the data in the following steps:

[0926] 1. Fixed cost anomaly detection: The server calculates the average and standard deviation of fixed costs such as utility bills and mobile phone bills based on past transaction data. Specifically, it executes the SQL query "SELECT AVG(amount), STDDEV(amount) FROM transactions WHERE type='utility'" to obtain the results. If the newly acquired transaction amount exceeds the predicted range of the average value ±3 x standard deviation, it is detected as an anomaly.

[0927] 2. Detection of large transactions: If a large transaction that exceeds a set threshold (for example, 1 million yen) occurs, the server detects it as an anomaly.

[0928] Step 4:

[0929] If an anomaly is detected, the server generates an alert message. The input to this step is the abnormal transaction detection result obtained in step 3. The output is the generated alert message. Specifically, the server generates the message "Your electricity bill this month is unusually high. Please review it."

[0930] Step 5:

[0931] The emotion engine recognizes emotions in real time from the user's facial expressions, voice, text input, etc. The inputs for this step are the user's facial images, voice data, and text input obtained from the device. The output is recognized emotion data. Specifically, the device captures the user's facial expressions using the camera and analyzes them using an emotion recognition model (e.g., Google Cloud Vision API). It also recognizes audio from the microphone and analyzes the audio data using an emotion analysis API (e.g., IBM Watson Tone Analyzer).

[0932] Step 6:

[0933] The emotion engine adjusts the alert message based on the recognized emotion data. The input to this step is the emotion data obtained in step 5 and the alert message generated in step 4. The output is the adjusted alert message. Specifically, the server changes the message to a gentler one, such as "Your electricity bill seems a little high. Please check it," based on feedback from the emotion engine.

[0934] Step 7:

[0935] The device notifies the user of the adjusted alert message. The input to this step is the alert message adjusted in step 6. The output is a notification to the user. Specifically, the device uses a push notification to display the message "I'm a little worried, are you okay?"

[0936] Step 8:

[0937] The user checks the alert message and selects whether to notify family members. The input to this step is the alert message displayed in step 7. The output is the user's selection information. Specifically, the user clicks the "Notify family members" button displayed on the terminal.

[0938] Step 9:

[0939] The server receives the user's permission information and sends an alert message to the family. The input to this step is the user's selection information obtained in step 8. The output is an alert message to the family. Specifically, the server executes the SQL query "SELECT contact_info FROM family_contacts WHERE user_id='12345'" to obtain the family's contact information and sends a message via email or SMS saying, "Your parent has an unusual transaction. Please check on them."

[0940] (Application example 2)

[0941] 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."

[0942] In modern society, there has been an increase in fraudulent access to elderly people's financial accounts and fraudulent transactions. This not only threatens the financial security of elderly people, but also increases psychological burdens such as anxiety and confusion. Current systems are capable of detecting and notifying abnormal transactions, but they are unable to respond appropriately based on the user's emotions, which can cause excessive stress to the user. In addition, notification to family members is not incorporated into the process, resulting in a lack of cooperation between elderly people and their families.

[0943] 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.

[0944] In this invention, the server includes: [means for acquiring transaction information for the elderly person's financial account;] [means for analyzing the acquired transaction information and detecting abnormalities;] [means for notifying the user of the detected abnormality, adjusting the content of the notification using emotion analysis means that recognizes the user's emotion, and requesting permission for notifying family members; and [means for sending notifications to family members based on the user's permission.] This ensures the safety of the elderly person's financial account while realizing appropriate notifications that take emotion into consideration, and enables the elderly person and their family members to intervene quickly if necessary.

[0945] "Elderly" refers to people over a certain age, primarily those considered at high risk of financial fraud and unauthorized access.

[0946] "Financial account" refers to an account opened with a bank or financial institution for the purpose of depositing money or conducting transactions.

[0947] "Transaction Information" refers to data regarding transaction dates, amounts, types, and counterparties obtained through a financial institution's API.

[0948] "Analysis" refers to the process of evaluating acquired transaction information and detecting anomalies based on patterns and regularities.

[0949] An "abnormal transaction" refers to a transaction that is deemed to be unnatural and different from the norm, and requires safety confirmation.

[0950] "User" refers to the elderly person who uses this system or someone acting on their behalf.

[0951] "Emotion analysis means" refers to technology for recognizing a user's psychological state from facial expressions, voice, text input, etc.

[0952] "Adjusting notification content" refers to softening or emphasizing alert or notification content based on the user's current emotional state.

[0953] "Permission" refers to the act of a user deciding whether or not to send notifications to family members.

[0954] "Family" refers to the user's children, siblings, and other close relatives who are responsible for providing support as needed.

[0955] A "server" refers to a computer system that primarily acquires, analyzes, stores, and notifies data.

[0956] To implement this invention, it is necessary to build a system that monitors transaction information in elderly people's financial accounts, detects abnormal transactions, and notifies users. The system mainly consists of a server, terminals, a sentiment analysis engine, and users.

[0957] Server Processing

[0958] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database.

[0959] The server then analyzes the transaction information and detects abnormal transactions. Transaction anomalies are defined as follows:

[0960] 1. Fixed cost anomalies: Based on past transaction data, the average value and standard deviation of fixed costs such as utility and communication fees are calculated. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ± 3 × standard deviation), it is detected as an anomaly.

[0961] 2. Large Transactions: If a large transaction occurs that exceeds a set threshold, it is detected as an anomaly.

[0962] When an anomaly is detected, the server generates an alert message, which includes details of the anomalous transaction, the type of anomaly, and a prompt to notify family members. Furthermore, a sentiment analysis engine recognizes the user's emotions and tailors the notification accordingly.

[0963] Sentiment analysis engine processing

[0964] The sentiment analysis engine recognizes emotions from the user's facial expressions, voice, text input, etc. The sentiment analysis engine performs the following tasks:

[0965] 1. Emotion recognition: Recognize the user's emotions in real time through the device's camera, microphone, and input devices.

[0966] 2. Sentiment analysis: Send feedback to the server based on the recognized emotion (surprise, anxiety, confusion, etc.).

[0967] 3. Notification adjustment: If the user feels anxious or confused, we recommend notifying their family members. We may also change the notification content to be more gentle.

[0968] Terminal handling

[0969] The device receives the alert message sent by the server and notifies the user. The notification is coordinated as follows:

[0970] 1. Normal notification: If an abnormal transaction is detected, the user will be notified with detailed information.

[0971] 2. Tailored notifications: Based on the results of the sentiment analysis engine, notifications are sent in a gentler tone. For example, a message like, "I'm a little worried, are you okay?"

[0972] 3. Recommend notifying family: If the user is feeling anxious or confused, the option to recommend notifying family will be automatically selected.

[0973] The user checks the alert message and selects whether to notify their family. If the user allows notification to their family, the information is sent from the device to the server.

[0974] Additional Server Processing

[0975] After receiving the user's authorization, the server retrieves the family member's contact information from a database and sends an alert message to the family member, including details of the abnormal transaction, the user's selection, and suggestions for how to proceed.

[0976] Specific examples

[0977] For example, if a user's transaction history shows a utility bill of 100,000 yen, it is considered to be within the normal expected range and is not detected as an anomaly. However, if a utility bill transaction of 500,000 yen occurs, it is detected as an anomaly and the user is notified. If the user is concerned, their family members will also be notified.

[0978] Prompt sentence for generative AI model

[0979] Create a program for a system that monitors elderly people's financial transactions and detects and notifies them of any abnormal transactions. The system will include a server that collects and analyzes transaction information and detects abnormalities, adjusts notifications based on emotion recognition, and notifies family members. Please also specify the specific names of the APIs and libraries to be used. For example, transaction information could be obtained from a financial institution's API, and notifications could be sent using a communication API.

[0980] This system allows seniors to manage transaction anomalies themselves while ensuring their financial security. The adoption of a sentiment analysis engine also reduces the user's psychological burden, allowing family members to quickly intervene when necessary, further enhancing the security of seniors' daily lives.

[0981] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0982] Step 1:

[0983] The server obtains transaction information for the elderly person's financial account through the financial institution's API.

[0984] (Input) Transaction information (transaction date, amount, type, and counterparty) from the financial institution's API

[0985] (Output) Save the acquired transaction information in the database

[0986] (Specific operation) The server periodically sends API requests to obtain and store transaction information.

[0987] Step 2:

[0988] The server analyzes the transaction information it acquires and detects any abnormal transactions.

[0989] (Input) Transaction information stored in the database

[0990] (Output) A list of transactions that were determined to be anomalous

[0991] (Specific operation) The server calculates the average value and standard deviation of fixed costs, and detects an anomaly when new transaction information exceeds the predicted range or when a large transaction exceeding a set threshold occurs.

[0992] Step 3:

[0993] If the server detects an abnormality, it generates an alert message.

[0994] (Input) A list of transactions that are determined to be abnormal

[0995] (Output) Alert message

[0996] (Specific Operation) The server generates an alert message containing details of the abnormal transaction and the type of abnormality.

[0997] Step 4:

[0998] The server uses a sentiment analysis engine to recognize the user's sentiment.

[0999] (Input) User facial expressions, voice, and text input obtained from the device's camera, microphone, and input device

[1000] (Output) User emotion data (surprise, anxiety, confusion, etc.)

[1001] (Specific operation) The emotion analysis engine analyzes the user's emotions in real time and feeds the results back to the server.

[1002] Step 5:

[1003] The server adjusts the notification content based on the user's emotions.

[1004] (Input) User emotion data, alert message

[1005] (Output) Adjusted alert message

[1006] (Specific operation) If the user feels anxious or confused, the server changes the notification content to softer language and generates an alert message recommending that the user notify family members.

[1007] Step 6:

[1008] The terminal receives the alert message sent from the server and notifies the user.

[1009] (Input) Adjusted alert message

[1010] (Output) User notification

[1011] (Specific operation) The device notifies the user in the form of a pop-up or a voice message.

[1012] Step 7:

[1013] The user can choose whether to notify their family.

[1014] (Input) Alert message content

[1015] (Output) Allow or deny notification to family members

[1016] (Specific operation) The user checks the alert message and selects "Notify family" or "Do not notify" via the device.

[1017] Step 8:

[1018] The server sends notifications to family members based on the user's permission.

[1019] (Input) User notification permission information

[1020] (Output) Alert message sent to family members

[1021] (Specific Operation) The server retrieves the family member's contact information from the database and sends an alert message containing details of the abnormal transaction to the family member's terminal.

[1022] This way, seniors' financial accounts can be kept safe while still providing timely and sensitive notifications, allowing seniors and their families to intervene quickly if needed.

[1023] 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.

[1024] 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.

[1025] 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.

[1026] [Third embodiment]

[1027] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1028] 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.

[1029] 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).

[1030] 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.

[1031] 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.

[1032] 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).

[1033] 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.

[1034] 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.

[1035] 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.

[1036] 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.

[1037] 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.

[1038] 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."

[1039] To implement this invention, a system is constructed that monitors transaction information in elderly people's financial accounts, detects abnormal transactions, and notifies them. The system is primarily composed of a server, terminals (such as smartphones and PCs), and users.

[1040] Server Processing

[1041] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database.

[1042] The server then analyzes the transaction information and detects abnormal transactions. Transaction anomalies are defined as follows:

[1043] 1. Fixed cost anomalies: Based on past transaction data, the average value and standard deviation of fixed costs such as utility bills and mobile phone bills are calculated. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ± 3 × standard deviation), it is detected as an anomaly.

[1044] 2. Large Transactions: If a large transaction occurs that exceeds a set threshold, it is detected as an anomaly.

[1045] If an anomaly is detected, the server sends an alert message to the user's device, containing details of the anomalous transaction and a message asking the user to confirm whether to notify family members.

[1046] Terminal handling

[1047] The terminal receives the alert message sent from the server and notifies the user. The user can then check the alert message and confirm the transaction details themselves. They can also choose whether to send notifications to their family members. If the user allows notifications to be sent to their family members, the information is sent from the terminal to the server.

[1048] User Action

[1049] The user checks the alert message displayed on the device and considers how to respond to any unusual transactions. If the transaction is suspicious, they can choose not to notify their family, but if not, it is recommended that they allow notifications. This choice is made on the device, and the result is sent to the server.

[1050] Additional Server Processing

[1051] If the user allows notifications to family members, the server will send an alert message to the family members' devices registered in advance, allowing the family members to receive detailed information about abnormal transactions and take appropriate measures.

[1052] Specific examples

[1053] Specifically, the following scenarios are possible:

[1054] 1. Normal transaction:

[1055] The server retrieves daily utility bills (10,000 yen) from the financial institution API.

[1056] This transaction is within the expected range and is not detected as an anomaly.

[1057] 2. Unusual utility bill transactions:

[1058] The server retrieves a utility bill transaction for 50,000 yen from the financial institution API.

[1059] This transaction is detected as an anomaly because it is far outside the expected range (average 10,000 yen ± 30,000 yen).

[1060] The server generates an alert and sends it to the user's terminal.

[1061] The user checks the alert on the device and allows notification to family members.

[1062] The server sends a notification to the family's device, providing them with detailed information.

[1063] 3. Large transfers:

[1064] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[1065] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[1066] The server generates an alert and notifies the user.

[1067] If the user is not aware of the incident, they can allow their family members to be notified.

[1068] The server sends a notification to the family's devices and action is taken to address the issue.

[1069] This system allows seniors to ensure their financial security while managing any irregularities in their transactions themselves, and allows family members to intervene quickly if necessary, helping to ensure the safety of seniors' daily lives.

[1070] The processing flow will be explained below.

[1071] Program processing

[1072] Server Processing

[1073] Step 1:

[1074] The server accesses the financial institution API at regular intervals (e.g., 9:00 AM every day).

[1075] The server sends a request for transaction data using authentication information for the financial institution API.

[1076] If authentication is successful, the financial institution API returns the user's transaction information.

[1077] The server stores the acquired transaction information in a database.

[1078] Step 2:

[1079] The server retrieves the latest transaction information from the database.

[1080] Filter newly acquired transaction information to identify transactions of interest.

[1081] The transactions covered include the date, transaction amount, transaction type, and counterparty.

[1082] Step 3:

[1083] The server categorizes the transaction information into fixed cost categories.

[1084] The server analyzes past transaction data to identify fixed expense categories such as utility bills and cell phone bills.

[1085] Check if the new transaction falls into the fixed cost category.

[1086] Step 4:

[1087] The server detects anomalies in fixed costs.

[1088] The server calculates the mean and standard deviation for the fixed cost categories.

[1089] If the new transaction amount exceeds the expected range (mean ± 3 x standard deviation), it is marked as an anomaly.

[1090] Step 5:

[1091] The server detects anomalies in large transactions.

[1092] Set a large amount threshold (e.g., 1 million yen) for all transactions.

[1093] If the transaction amount exceeds this threshold, it is marked as an anomaly.

[1094] Step 6:

[1095] If an anomaly is detected, the server generates an alert message.

[1096] The alert message will include details of the unusual transaction, the type of abnormality, and confirmation of permission to notify family members.

[1097] The server sends this message to the user's terminal.

[1098] Terminal handling

[1099] Step 1:

[1100] The terminal receives the alert message from the server.

[1101] Notify the user that new notifications are available.

[1102] The device will display the notification content in a pop-up or notification bar.

[1103] Step 2:

[1104] The terminal waits for the user's action.

[1105] The user opens the notification and sees the details of the alert message.

[1106] The alert message will include transaction details, a description of the anomaly, and an option to notify family members.

[1107] Step 3:

[1108] The user presses the button to allow notifications to family members.

[1109] The terminal transmits the user's authorization information to the server.

[1110] Additional Server Processing

[1111] Step 1:

[1112] The server receives the user's authorization information.

[1113] Retrieve contact information for registered family members from the database.

[1114] Step 2:

[1115] The server sends an alert message to the family.

[1116] A message is sent to the family member's device containing details of the abnormal transaction, the user's selection, and suggestions on how to proceed.

[1117] If the message is sent successfully, a record of it is saved in the database.

[1118] User Action

[1119] Step 1:

[1120] The user checks the alert message on the terminal.

[1121] Check the details of any unusual transactions to see if they apply to you.

[1122] Step 2:

[1123] The user selects whether to notify family members.

[1124] If you are not aware of the incident, allow your family to be notified.

[1125] If permission is granted, the information is sent from the terminal to the server.

[1126] Based on these specific processing steps, abnormal transactions in the elderly person's financial accounts can be monitored and necessary measures can be taken promptly.

[1127] Example 1

[1128] 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."

[1129] The increase in fraudulent and abnormal transactions in elderly people's financial accounts has created a need for detection and rapid notification. Current systems take time to acquire and analyze transaction information, and lack the means to quickly notify users and their families when an abnormality is detected. This increases the likelihood that elderly people will be exposed to financial risk, which is a major source of anxiety.

[1130] 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.

[1131] In this invention, the server includes means for periodically acquiring transaction information from the elderly person's financial account via the financial institution's API, means for storing the acquired transaction information in a database, means for analyzing the stored transaction information and detecting abnormal transactions, means for sending an alert message including details of the detected abnormal transaction to the user's terminal, means for providing an interface that allows the user to check the alert and select whether to notify family members, and means for sending a notification to the family member's terminal based on the user's permission. This makes it possible to quickly detect abnormalities in the elderly person's financial transactions and notify the user and family members in a timely manner.

[1132] "Financial institution API" stands for Application Program Interface provided by financial institutions, which is an interface used by external systems to obtain financial transaction data.

[1133] An "elderly person's financial account" refers to an account opened by an elderly person at a bank or other financial institution, through which financial transactions such as deposits, withdrawals, and transfers are conducted.

[1134] "Transaction information" is data showing the transaction history of a financial account within a specific period, and includes transaction dates, transaction amounts, transaction types, and transaction partner information.

[1135] "Database" means a computer system for systematically storing and managing acquired transaction information, allowing for easy access and analysis.

[1136] An "abnormal transaction" is a transaction that is determined to be out of the ordinary based on past transaction patterns or established thresholds, including transactions that may be fraudulent or involve abnormal amounts.

[1137] An "alert message" is a notification sent to a user when an abnormal transaction is detected, and contains details of the transaction and information urging the user to take action.

[1138] "Users" are those who use this system to receive notifications of abnormal transactions, and primarily refer to elderly people and their families.

[1139] "Terminal" refers to a device used by a user, such as a smartphone, PC, or tablet, that receives and displays alert messages from the server.

[1140] "Family devices" refer to devices such as smartphones, PCs, and tablets owned by the user's family, and alert messages are sent to these devices based on the user's permission.

[1141] To implement this invention, it is necessary to build a system that monitors transaction information in elderly people's financial accounts, detects abnormal transactions, and notifies users. This system is primarily composed of a server, terminals (such as smartphones and PCs), and users.

[1142] Server Processing

[1143] The server periodically retrieves transaction information from the elderly person's financial account via the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The retrieved information is stored in a database. Specifically, the server sends an API request at a set time each day to retrieve daily transaction data and record it in the database.

[1144] The server then analyzes the stored transaction information and detects abnormal transactions. Transaction abnormalities are defined in two patterns: fixed cost abnormalities and large transaction amounts. For fixed cost abnormalities, the server calculates the average value and standard deviation of fixed costs such as utility bills and mobile phone bills based on past transaction data. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ±3 x standard deviation), it is detected as an abnormality. On the other hand, for large transaction amounts, an abnormality is detected if a transaction that exceeds the set threshold occurs. If an abnormality is detected, the server sends an alert message to the user's device.

[1145] Terminal handling

[1146] The terminal receives the alert message sent from the server and is responsible for notifying the user. The user can then review the alert message and investigate the transaction details themselves. The interface displayed on the terminal includes an option to select whether to send notifications to family members. If the user allows notifications to be sent to family members, the terminal sends the selection to the server.

[1147] User Action

[1148] The user checks the alert message displayed on the device. If the transaction is recognizable, they can choose not to notify their family. If they are not recognizable, they can choose to allow notification to their family. This selection is sent to the server via the device.

[1149] Additional Server Processing

[1150] If the user allows notification to family members, the server will send an alert message to the pre-registered family members' devices, allowing the family members to receive detailed information about the abnormal transaction and take appropriate measures.

[1151] Specific examples

[1152] Specifically, the following scenarios are possible:

[1153] 1. Normal transaction:

[1154] The server retrieves daily utility bills (10,000 yen) from the financial institution's API.

[1155] This transaction is within the expected range and is not detected as an anomaly.

[1156] 2. Unusual utility bill transactions:

[1157] The server retrieves a utility bill transaction for 50,000 yen from the financial institution's API.

[1158] This transaction is detected as an anomaly because it is far outside the predicted range (average 10,000 yen ± 3 x 5,000 yen).

[1159] The server generates an alert message and sends it to the user's terminal.

[1160] The user confirms the alert message and allows notification to family members.

[1161] The server sends a notification to the family's device, providing them with detailed information.

[1162] 3. Large transfers:

[1163] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[1164] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[1165] The server generates an alert message and notifies the user.

[1166] If the user is not aware of the incident, they can allow their family members to be notified.

[1167] The server sends a notification to the family's devices and steps are taken to take action.

[1168] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1169] Step 1:

[1170] The server retrieves the transaction information

[1171] Input: Request to financial institution API

[1172] Specific operation: The server sends a request to the financial institution's API at a specific time interval (e.g., every day at midnight), using the authentication token to access the API and requesting transaction information for the target senior citizen's financial account.

[1173] Output: Transaction information (JSON format)

[1174] Specific data processing: Transaction information is returned from financial institutions in JSON format, including the transaction date, transaction amount, transaction type (deposit, withdrawal, transfer, etc.), and counterparty.

[1175] Step 2:

[1176] The server saves the transaction information it has acquired into a database.

[1177] Input: Retrieved transaction information (JSON format)

[1178] Specific operation: The server analyzes the acquired transaction information and extracts necessary fields, such as transaction date, transaction amount, and transaction type.

[1179] Output: Stored transaction information (specific table in database)

[1180] Specific data processing: The extracted data is properly saved in the transactions table of the database.

[1181] Step 3:

[1182] The server analyzes the transaction information and detects abnormal transactions.

[1183] Input: Historical transaction data in the database, retrieved new transaction data

[1184] Specific operation: The server calculates the average and standard deviation of fixed costs based on past transaction data. For example, using utility bill data from the past three months, the server calculates the average (e.g., 10,000 yen) and standard deviation (e.g., 5,000 yen).

[1185] Output: Abnormal transactions (list format)

[1186] Specific data calculation: If a new transaction exceeds the average value ±3 × standard deviation, the transaction is added to the list as an anomaly. Large transactions exceeding a set threshold (e.g., 1 million yen) are also added to the list as an anomaly.

[1187] Step 4:

[1188] The server generates an alert for abnormal transactions and sends it to the terminal.

[1189] Input: Detected abnormal transactions (list format)

[1190] What happens: The server creates an alert message with details of the anomalous transaction. For example, it generates a message like, "A utility bill transaction of 50,000 yen has been detected as anomalous. Does this transaction seem familiar?"

[1191] Output: Alert message (push notification to device)

[1192] Specific data processing: The generated alert message is sent to the user's device as a push notification.

[1193] Step 5:

[1194] The device receives the alert and notifies the user.

[1195] Input: The alert message sent from the server

[1196] Specific behavior: The device will display the received alert message as a pop-up notification on the screen. When the user taps the pop-up, the app will open and display more information.

[1197] Output: Notification to the user (display on screen)

[1198] Specific data display: Detailed information about the alert is displayed on the screen, and the user is given the option to "notify family members."

[1199] Step 6:

[1200] The user responds to the alert and sends their choice from the device to the server

[1201] Input: Alert message confirmation result (whether to notify family members)

[1202] Specific operation: The user checks the details of the alert and selects "Notify family members" or "Don't notify family members." The selection is sent from the device to the server.

[1203] Output: Selection (whether to notify family members)

[1204] Specific data transmission: User selections are transmitted to the server.

[1205] Step 7:

[1206] The server sends a notification to the family

[1207] Input: User's selection result (allow notification to family members)

[1208] Specific operation: The server checks the pre-registered contact information of family members, creates an alert message, and sends it to the family members' devices. For example, it generates a message such as, "An abnormal transaction has been detected in your parent's account. Please check the details."

[1209] Output: Alert message to family members (push notification to device or email)

[1210] Specific data transmission: The generated alert message will be sent to the family's device so that the family can check the detailed information.

[1211] (Application example 1)

[1212] 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."

[1213] Ensuring the security of transactions in elderly people's financial accounts is important for preventing financial damage, but it is difficult for elderly people themselves to constantly monitor their transactions for abnormalities. There is also a need for a method to quickly and reliably notify family members when abnormal transactions occur. Conventional systems lack the ability to detect abnormalities in real time and provide immediate notification, preventing family members from responding quickly. A new system is needed to resolve these issues and provide peace of mind to elderly people and their families.

[1214] 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.

[1215] In this invention, the server includes: [means for acquiring transaction information on the elderly person's financial account]; [means for analyzing the acquired transaction information and detecting abnormalities]; [means for notifying the user of detected abnormalities and requesting permission for notifying family members]; [means for sending notifications to family members based on the user's permission]; [means for sending notifications in real time when an abnormal transaction is detected]; and [means for sending similar alerts to family members]. This enables real-time detection of abnormal transactions and prompt notification, enabling the elderly person and their family members to prevent financial damage. It also provides a system that reduces the burden on the elderly person themselves and allows family members to respond quickly.

[1216] The term "elderly" generally refers to people aged 60 or over, especially those in need of social and medical support.

[1217] "Financial Account" means a deposit account or other asset management account opened by a customer at a financial institution for the purpose of managing assets.

[1218] "Transaction information" refers to detailed data regarding various financial transactions, such as deposits, withdrawals, transfers, and payments, in a financial account.

[1219] "Real-time" means that data is collected, analyzed, and communicated almost immediately, with very little time delay.

[1220] "Abnormal transactions" refer to transactions that deviate significantly from normal trading patterns, including large transactions and transactions that exceed expected ranges.

[1221] A "server" refers to a computer system that provides processing power and data to multiple clients over a network.

[1222] "Database" means a system for systematically storing and managing information in digital form.

[1223] "Notification" means a message or alert sent to inform a user or interested party of specific information.

[1224] An "alert" refers to a notification method that immediately warns the user when an abnormality or a condition requiring attention occurs.

[1225] "Family" refers to a person related by blood or marriage to an older person and who is responsible for assisting and supporting the older person in managing their life, health, and property.

[1226] To implement this invention, it is necessary to build a system consisting of a server, terminals (such as smartphones or PCs), and users. The main function of the system is to provide a mechanism for monitoring transaction information in elderly people's financial accounts and detecting and notifying abnormal transactions.

[1227] Server Processing

[1228] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and counterparty. The obtained information is stored in a database. The server uses the NumPy library to calculate the mean and standard deviation from past transaction data. It defines the mean value and prediction range (e.g., mean value ± 3 × standard deviation) of fixed costs, and detects newly obtained transactions that exceed the prediction range as anomalous transactions. Large transactions whose transaction amount exceeds a set threshold are also detected as anomalous. When an anomaly is detected, the server uses Pushbullet to send an alert message via push notification to the user's device.

[1229] Terminal handling

[1230] The device receives the alert message sent from the server and notifies the user. The user can check the alert message and check the transaction details themselves. The alert message includes details of the abnormal transaction and an option to notify family members. If the user allows notification to family members, the device sends the information to the server.

[1231] User Action

[1232] The user checks the alert message displayed on the device and decides whether or not the transaction is recognizable. If not, the user is encouraged to allow notifications to family members. This selection is made on the device, and the result is sent to the server. Based on the user's permission, the server can send alert messages to the devices of family members who have been registered in advance.

[1233] Specific examples

[1234] For example, if the average utility bill for the past six months was 10,000 yen with a standard deviation of 5,000 yen, and a new transaction of 20,000 yen occurs, the server will detect this as an anomaly. In this case, the server will send an alert to the user's smartphone stating, "Anomalous transaction detected: A utility bill transaction of 20,000 yen occurred on 2023-10-01." If the user allows notifications, similar alerts will also be sent to the smartphones of family members.

[1235] Example prompts for generative AI models

[1236] "We will design a system that monitors transaction information for elderly people's financial accounts, detects abnormal transactions, and notifies users. An abnormal transaction in the transaction data we received (for example, the average utility bill for the past six months is 10,000 yen, with a standard deviation of 5,000 yen) is one that exceeds the average value ±3 x standard deviation. We will set it up so that a push notification is sent stating, 'Abnormal transaction detected: A utility bill transaction of 50,000 yen occurred on 2023-10-01.' Please create a specific program based on this content."

[1237] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1238] Step 1:

[1239] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. The information obtained includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The input for this process is the API call, and the output is transaction information data in JSON format. The server saves the obtained data in a database.

[1240] Step 2:

[1241] The server extracts historical transaction data from the stored transaction information and calculates the mean and standard deviation of a particular fixed cost (e.g., utility bills). It uses the historical transaction data (in the database) as input and gets the mean and standard deviation as output. It uses the NumPy library to perform statistical operations.

[1242] Step 3:

[1243] The server analyzes newly acquired transaction data and determines whether the transaction falls within the expected range based on the calculated mean and standard deviation. The input is the latest transaction data, and the output is a flag indicating whether the transaction is "anomalous" or "normal." For example, if the amount of a new transaction exceeds the mean value ± 3 x standard deviation, the server detects the transaction as "anomalous."

[1244] Step 4:

[1245] The server uses Pushbullet's API to send an alert message to the user's device for transactions detected as abnormal. The alert includes details of the abnormal transaction (transaction date, amount, and counterparty) and an option to notify family members. The input is the abnormal transaction data, and the output is the alert message.

[1246] Step 5:

[1247] The device receives the alert message sent from the server and notifies the user. The input is the alert message, and the output is a push notification to the user. The user checks the content of the alert message on the device.

[1248] Step 6:

[1249] The user checks the alert message displayed on the terminal and selects how to respond to the abnormal transaction. Specifically, the user selects whether to notify family members. The input is the notified alert information, and the output is the user's selection.

[1250] Step 7:

[1251] When the user allows notifications to family members, the device sends the selection information to the server. The input is the user's selection, and the output is the notification permission data for the server.

[1252] Step 8:

[1253] Based on the user's permission, the server sends an alert message to pre-registered family members' devices. This allows family members to receive information about abnormal transactions and take prompt action. The input is notification permission data from the user, and the output is an alert message for the family members.

[1254] By following the above steps, financial transactions of elderly people can be monitored in real time, and if any abnormalities are detected, the user and their family can be notified promptly, thereby preventing financial damage.

[1255] 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.

[1256] To implement this invention, we will build a system that monitors transaction information on elderly people's financial accounts, detects and notifies abnormal transactions, and combines it with an emotion engine that recognizes user emotions. The system is primarily composed of a server, a terminal (such as a smartphone or PC), the emotion engine, and the user.

[1257] Server Processing

[1258] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database.

[1259] The server then analyzes the transaction information and detects abnormal transactions. Transaction anomalies are defined as follows:

[1260] 1. Fixed cost anomalies: Based on past transaction data, the average value and standard deviation of fixed costs such as utility bills and mobile phone bills are calculated. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ± 3 × standard deviation), it is detected as an anomaly.

[1261] 2. Large Transactions: If a large transaction occurs that exceeds a set threshold, it is detected as an anomaly.

[1262] When an anomaly is detected, the server generates an alert message, which includes details of the anomalous transaction, the type of anomaly, and a prompt to notify family members. Furthermore, an emotion engine recognizes the user's emotions and tailors the notification accordingly.

[1263] Emotion engine processing

[1264] The emotion engine recognizes emotions from the user's facial expressions, voice, text input, etc. The emotion engine performs the following tasks:

[1265] 1. Emotion recognition: Recognizes user emotions in real time through the device's camera, microphone, and keyboard input.

[1266] 2. Sentiment analysis: Send feedback to the server based on the recognized emotion (surprise, anxiety, confusion, etc.).

[1267] 3. Notification adjustment: If the user feels anxious or confused, we recommend notifying their family members. We may also change the notification content to be more gentle.

[1268] Terminal handling

[1269] The device receives the alert message sent by the server and notifies the user. The notification is coordinated as follows:

[1270] 1. Normal notification: If an abnormal transaction is detected, the user will be notified with detailed information.

[1271] 2. Adjusted Notification: Based on the analysis results of the emotion engine, notifications are sent in a gentler tone. For example, a message like, "I'm a little worried, are you okay?"

[1272] 3. Recommend notifying family: If the user is feeling anxious or confused, the option to recommend notifying family will be automatically selected.

[1273] The user checks the alert message and selects whether to notify their family. If the user allows notification to their family, the information is sent from the device to the server.

[1274] Additional Server Processing

[1275] After receiving the user's authorization, the server retrieves the family member's contact information from a database and sends an alert message to the family member, including details of the abnormal transaction, the user's selection, and suggestions for how to proceed.

[1276] Specific examples

[1277] Specifically, the following scenarios are possible:

[1278] 1. Normal transaction:

[1279] The server retrieves daily utility bills (10,000 yen) from the financial institution API.

[1280] This transaction is within the expected range and is not detected as an anomaly.

[1281] 2. Unusual utility bill transactions:

[1282] The server retrieves a utility bill transaction for 50,000 yen from the financial institution API.

[1283] This transaction is detected as an anomaly because it is far outside the expected range (average 10,000 yen ± 30,000 yen).

[1284] The emotion engine detects the user's surprise or anxiety and generates a gentle alert recommending notifying family members.

[1285] The server generates the adjusted alert and sends it to the user's device.

[1286] The user confirms the notification and allows notification to family members.

[1287] The server sends tailored alerts to family devices, allowing them to respond quickly.

[1288] 3. Large transfers:

[1289] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[1290] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[1291] The emotion engine analyzes the user's emotions and generates appropriate notification content.

[1292] The server generates an alert containing details of the abnormal transaction and notifies the user.

[1293] If the user is not aware of the incident, they can allow their family members to be notified.

[1294] The server sends a notification to the family's devices and action is taken to address the issue.

[1295] This system allows seniors to manage transaction anomalies themselves while ensuring their financial security. The introduction of an emotion engine also reduces the psychological burden on users, allowing family members to intervene quickly if necessary, further enhancing the security of seniors' daily lives.

[1296] The processing flow will be explained below.

[1297] Server Processing

[1298] Step 1:

[1299] The server accesses the financial institution API at regular intervals (e.g., 9:00 AM every day).

[1300] The server sends a request for transaction data using authentication information for the financial institution API.

[1301] If authentication is successful, the financial institution API returns the user's transaction information.

[1302] The server stores the acquired transaction information in a database.

[1303] Step 2:

[1304] The server retrieves the latest transaction information from the database.

[1305] The server filters the newly acquired transaction information to identify eligible transactions.

[1306] The transactions covered include the date, transaction amount, transaction type, and counterparty.

[1307] Step 3:

[1308] The server categorizes the transaction information into fixed cost categories.

[1309] The server analyzes past transaction data to identify fixed expense categories such as utility bills and cell phone bills.

[1310] Check if the new transaction falls into the fixed cost category.

[1311] Step 4:

[1312] The server detects anomalies in fixed costs.

[1313] The server calculates the mean and standard deviation for the fixed cost categories.

[1314] If the new transaction amount exceeds the expected range (mean ± 3 x standard deviation), it is marked as an anomaly.

[1315] Step 5:

[1316] The server detects anomalies in large transactions.

[1317] The server sets a large amount threshold (e.g., 1 million yen) for all transactions.

[1318] If the transaction amount exceeds this threshold, it is marked as an anomaly.

[1319] Step 6:

[1320] If an anomaly is detected, the server generates an alert message.

[1321] The alert message will include details of the abnormal transaction, the type of abnormality, and permission to notify family members.

[1322] Emotion engine processing

[1323] Step 1:

[1324] The emotion engine captures the user's facial expressions and voice in real time.

[1325] Data is collected through cameras and microphones.

[1326] Step 2:

[1327] The emotion engine analyzes the acquired data and recognizes the user's emotions.

[1328] The emotion engine uses facial recognition algorithms and voice analysis to identify the user's emotions.

[1329] If the user's emotion corresponds to anxiety, surprise, confusion, etc., the information is sent to the server.

[1330] Step 3:

[1331] The emotion engine sends feedback to the server to adjust the notification content.

[1332] If the user is feeling anxious or confused, change the notification message to a more gentle one.

[1333] An option to recommend notifying family members will also be added.

[1334] Terminal handling

[1335] Step 1:

[1336] The terminal receives the alert message from the server.

[1337] Notify the user that new notifications are available.

[1338] The device will display the notification content in a pop-up or notification bar.

[1339] Step 2:

[1340] The terminal waits for the user's action.

[1341] The user opens the notification to view the details of the alert message.

[1342] The message will include transaction details, a description of the anomaly, and an option to notify family members.

[1343] Step 3:

[1344] When the user presses the button to allow notifications to family members,

[1345] The terminal transmits the user's authorization information to the server.

[1346] Additional Server Processing

[1347] Step 1:

[1348] The server receives the user's authorization information.

[1349] The server retrieves contact information for registered family members from the database.

[1350] Step 2:

[1351] The server sends an alert message to the family.

[1352] A message is sent to the family member's device containing details of the abnormal transaction, the user's selection, and suggestions on how to proceed.

[1353] If the message is sent successfully, a record of it is saved in the database.

[1354] Specific processing examples

[1355] Specifically, the following scenarios are possible:

[1356] 1. Normal transaction:

[1357] The server retrieves daily utility bills (10,000 yen) from the financial institution API.

[1358] This transaction is within the expected range and is not detected as an anomaly.

[1359] 2. Unusual utility bill transactions:

[1360] The server retrieves a utility bill transaction for 50,000 yen from the financial institution API.

[1361] This transaction is detected as an anomaly because it is far outside the expected range (average 10,000 yen ± 30,000 yen).

[1362] The emotion engine detects the user's surprise or anxiety and generates a gentle alert recommending notifying family members.

[1363] The server generates the adjusted alert and sends it to the user's device.

[1364] The user confirms the notification and allows notification to family members.

[1365] The server sends tailored alerts to family devices, allowing them to respond quickly.

[1366] 3. Large transfers:

[1367] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[1368] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[1369] The emotion engine analyzes the user's emotions and generates appropriate notification content.

[1370] The server generates an alert containing details of the abnormal transaction and notifies the user.

[1371] If the user is not aware of the incident, they can allow their family members to be notified.

[1372] The server sends notifications to family members' devices so that countermeasures can be taken quickly.

[1373] Example 2

[1374] 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."

[1375] There is a need for a method and system that can respond quickly and appropriately when abnormal financial transactions occur among elderly people. Furthermore, it is necessary to consider the user's feelings when notifying them of abnormal transactions. Conventional systems have had problems such as causing unnecessary anxiety to users and delaying notification to family members. The purpose of this invention is to solve these problems and simultaneously ensure the financial security and psychological peace of mind of elderly people.

[1376] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for acquiring transaction information of the elderly person's financial account, a means for analyzing the acquired transaction information and detecting abnormalities, and a means for recognizing the user's emotions and adjusting the notification content. This makes it possible to notify the user of abnormal transactions in a manner that takes into consideration the user's emotions, and also to notify family members at an appropriate time.

[1377] "Elderly" refers to people who are older, especially those aged 65 or older, and who require special consideration in financial transactions and daily life.

[1378] "Financial account" is a general term for deposit accounts, savings accounts, fixed-term deposit accounts, etc. opened at banks, credit unions, and other financial institutions, and is a basic means for recording and managing financial transactions.

[1379] "Transaction information" refers to detailed data relating to financial transactions such as deposits, withdrawals, transfers, and debits related to a financial account, including the date, amount, type, and counterparty of the transaction.

[1380] "Abnormal transactions" refer to transactions that significantly deviate from normal trading patterns, including, for example, transactions that exceed the mean and standard deviation of a particular fixed cost or large transactions that exceed a set threshold.

[1381] "Emotion engine" is a general term for software or hardware that recognizes and analyzes emotions in real time from a user's facial expressions, voice, text input, etc.

[1382] "Adjusting the notification content" refers to a process of changing the wording and content of the notification message depending on the user's emotional state, so as not to make the user feel excessively anxious.

[1383] "Family" refers to a person who has a certain kinship relationship with the user and who has the role of supporting the user in emergencies or abnormal situations.

[1384] "Server" means the main computer system that manages the entire system and acquires, analyzes, and stores transaction information, and generates and transmits notifications.

[1385] A "terminal" refers to a device such as a smartphone or PC that is directly operated by a user and that receives notifications sent from a server.

[1386] "Notification" refers to a message sent by the system to the user or family member informing them of an abnormal transaction, and may be sent in the form of voice, text, push notification, etc.

[1387] To implement this invention, it is necessary to build a system that monitors transaction information on elderly people's financial accounts, detects abnormal transactions, and notifies them, as well as an emotion engine that recognizes user emotions. This system is primarily composed of a server, a terminal (such as a smartphone or PC), an emotion engine, and a user.

[1388] Server Processing

[1389] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, transaction amount, transaction type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database on the server. The server analyzes this data and runs an algorithm to detect abnormal transactions.

[1390] Specifically, the server detects abnormal transactions through the following steps:

[1391] 1. Fixed cost anomaly detection: Based on past transaction data, the average value and standard deviation of fixed costs such as utility bills and mobile phone bills are calculated. If the newly acquired transaction amount exceeds the predicted range of the average value ±3 x standard deviation, it is detected as an anomaly.

[1392] 2. Large transaction detection: The server detects an abnormality when a large transaction that exceeds a set threshold occurs.

[1393] When an anomaly is detected, the server sends feedback to the emotion engine, and also generates alert messages and adjusts the notification content as needed.

[1394] Emotion engine processing

[1395] The emotion engine recognizes emotions in real time from the user's facial expressions, voice, text input, etc. For this purpose, the device requires input devices such as a camera and microphone. The emotion engine provides the following functions:

[1396] 1. Emotion recognition: Recognizes the user's emotions through the device's camera, microphone, and keyboard input.

[1397] 2. Sentiment analysis: Send feedback to the server based on the recognized emotion (surprise, anxiety, confusion, etc.).

[1398] 3. Notification adjustment: If the user feels anxious or confused, we recommend notifying their family members. We may also change the notification content to be more gentle.

[1399] Terminal handling

[1400] The device receives the alert message sent by the server and notifies the user. The notification is arranged in the following pattern:

[1401] 1. Normal notification: If an abnormal transaction is detected, the user will be notified of the details.

[1402] 2. Adjusted Notification: Based on the analysis results of the emotion engine, notifications are sent in a gentler tone. For example, a message like, "I'm a little worried, are you okay?"

[1403] 3. Recommend notifying family: If the user is feeling anxious or confused, the option to recommend notifying family will be automatically selected.

[1404] The user checks the alert message and selects whether to notify their family. If the user allows notification to their family, the information is sent from the device to the server.

[1405] Additional Server Processing

[1406] After receiving the user's authorization, the server retrieves the family member's contact information from a database and sends an alert message to the family member, including details of the abnormal transaction, the user's selection, and suggestions for how to proceed.

[1407] Specific examples

[1408] The following example further illustrates the operation of the system:

[1409] Regular transactions

[1410] The server retrieves a daily utility bill (10,000 yen) from a financial institution API. This transaction is within the expected range and is not detected as an anomaly.

[1411] Unusual utility bill transactions

[1412] The server receives a utility bill transaction of 50,000 yen. This transaction is detected as an anomaly because it is far outside the predicted range (average 10,000 yen ± 30,000 yen). The emotion engine detects the user's surprise or anxiety and generates an alert in a gentle manner recommending notifying family members. The server generates an adjusted alert and sends it to the user's device. The user confirms the notification and allows the notification to family members. The server then sends the adjusted alert to the family members' devices, allowing them to respond quickly.

[1413] Large remittances

[1414] The server receives a remittance transaction of 2 million yen. Because this transaction exceeds the threshold (1 million yen), it is automatically detected as an anomaly. The emotion engine analyzes the user's emotions and generates appropriate notification content. The server generates an alert with details of the anomalous transaction and sends it to the device. If the user does not recognize the transaction, they can allow notification to family members. The server sends a notification to the family member's device, and action is taken to take measures.

[1415] Prompt Sentence Examples

[1416] By inputting the following prompts into the generative AI model, we can build specific system functions:

[1417] "Create a system that collects transaction information from senior citizens' financial accounts, detects abnormal transactions, and notifies them. Add a function that recognizes the user's emotions and adjusts the content of notifications accordingly."

[1418] This system allows seniors to manage transaction anomalies themselves while ensuring their financial security. The introduction of an emotion engine also reduces the psychological burden on users, allowing family members to intervene quickly if necessary, further enhancing the security of seniors' daily lives.

[1419] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1420] Step 1:

[1421] The server obtains transaction information for the elderly person's financial account through the financial institution's API. The input is the request information to the API, and the output is the obtained transaction details. Specifically, the server sends a request to the endpoint "https: / / api.bank.com / transactions" and receives the transaction data in JSON format.

[1422] Step 2:

[1423] The server saves the acquired transaction information in a database. The input is the transaction details data acquired in step 1, and the output is the transaction record saved in the database. Specifically, the server uses the INSERT statement to insert the transaction data it received into the "transactions" table in the database called "transaction_db".

[1424] Step 3:

[1425] The server analyzes the transaction information and detects abnormal transactions. The input to this step is the transaction data stored in the database. The output is the abnormal transaction detection results. The server processes the data in the following steps:

[1426] 1. Fixed cost anomaly detection: The server calculates the average and standard deviation of fixed costs such as utility bills and mobile phone bills based on past transaction data. Specifically, it executes the SQL query "SELECT AVG(amount), STDDEV(amount) FROM transactions WHERE type='utility'" to obtain the results. If the newly acquired transaction amount exceeds the predicted range of the average value ±3 x standard deviation, it is detected as an anomaly.

[1427] 2. Detection of large transactions: If a large transaction that exceeds a set threshold (for example, 1 million yen) occurs, the server detects it as an anomaly.

[1428] Step 4:

[1429] If an anomaly is detected, the server generates an alert message. The input to this step is the abnormal transaction detection result obtained in step 3. The output is the generated alert message. Specifically, the server generates the message "Your electricity bill this month is unusually high. Please review it."

[1430] Step 5:

[1431] The emotion engine recognizes emotions in real time from the user's facial expressions, voice, text input, etc. The inputs for this step are the user's facial images, voice data, and text input obtained from the device. The output is recognized emotion data. Specifically, the device captures the user's facial expressions using the camera and analyzes them using an emotion recognition model (e.g., Google Cloud Vision API). It also recognizes audio from the microphone and analyzes the audio data using an emotion analysis API (e.g., IBM Watson Tone Analyzer).

[1432] Step 6:

[1433] The emotion engine adjusts the alert message based on the recognized emotion data. The input to this step is the emotion data obtained in step 5 and the alert message generated in step 4. The output is the adjusted alert message. Specifically, the server changes the message to a gentler one, such as "Your electricity bill seems a little high. Please check it," based on feedback from the emotion engine.

[1434] Step 7:

[1435] The device notifies the user of the adjusted alert message. The input to this step is the alert message adjusted in step 6. The output is a notification to the user. Specifically, the device uses a push notification to display the message "I'm a little worried, are you okay?"

[1436] Step 8:

[1437] The user checks the alert message and selects whether to notify family members. The input to this step is the alert message displayed in step 7. The output is the user's selection information. Specifically, the user clicks the "Notify family members" button displayed on the terminal.

[1438] Step 9:

[1439] The server receives the user's permission information and sends an alert message to the family. The input to this step is the user's selection information obtained in step 8. The output is an alert message to the family. Specifically, the server executes the SQL query "SELECT contact_info FROM family_contacts WHERE user_id='12345'" to obtain the family's contact information and sends a message via email or SMS saying, "Your parent has an unusual transaction. Please check on them."

[1440] (Application example 2)

[1441] 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."

[1442] In modern society, there has been an increase in fraudulent access to elderly people's financial accounts and fraudulent transactions. This not only threatens the financial security of elderly people, but also increases psychological burdens such as anxiety and confusion. Current systems are capable of detecting and notifying abnormal transactions, but they are unable to respond appropriately based on the user's emotions, which can cause excessive stress to the user. In addition, notification to family members is not incorporated into the process, resulting in a lack of cooperation between elderly people and their families.

[1443] 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.

[1444] In this invention, the server includes: [means for acquiring transaction information for the elderly person's financial account;] [means for analyzing the acquired transaction information and detecting abnormalities;] [means for notifying the user of the detected abnormality, adjusting the content of the notification using emotion analysis means that recognizes the user's emotion, and requesting permission for notifying family members; and [means for sending notifications to family members based on the user's permission.] This ensures the safety of the elderly person's financial account while realizing appropriate notifications that take emotion into consideration, and enables the elderly person and their family members to intervene quickly if necessary.

[1445] "Elderly" refers to people over a certain age, primarily those considered at high risk of financial fraud and unauthorized access.

[1446] "Financial account" refers to an account opened with a bank or financial institution for the purpose of depositing money or conducting transactions.

[1447] "Transaction Information" refers to data regarding transaction dates, amounts, types, and counterparties obtained through a financial institution's API.

[1448] "Analysis" refers to the process of evaluating acquired transaction information and detecting anomalies based on patterns and regularities.

[1449] An "abnormal transaction" refers to a transaction that is deemed to be unnatural and different from the norm, and requires safety confirmation.

[1450] "User" refers to the elderly person who uses this system or someone acting on their behalf.

[1451] "Emotion analysis means" refers to technology for recognizing a user's psychological state from facial expressions, voice, text input, etc.

[1452] "Adjusting notification content" refers to softening or emphasizing alert or notification content based on the user's current emotional state.

[1453] "Permission" refers to the act of a user deciding whether or not to send notifications to family members.

[1454] "Family" refers to the user's children, siblings, and other close relatives who are responsible for providing support as needed.

[1455] A "server" refers to a computer system that primarily acquires, analyzes, stores, and notifies data.

[1456] To implement this invention, it is necessary to build a system that monitors transaction information in elderly people's financial accounts, detects abnormal transactions, and notifies users. The system mainly consists of a server, terminals, a sentiment analysis engine, and users.

[1457] Server Processing

[1458] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database.

[1459] The server then analyzes the transaction information and detects abnormal transactions. Transaction anomalies are defined as follows:

[1460] 1. Fixed cost anomalies: Based on past transaction data, the average value and standard deviation of fixed costs such as utility and communication fees are calculated. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ± 3 × standard deviation), it is detected as an anomaly.

[1461] 2. Large Transactions: If a large transaction occurs that exceeds a set threshold, it is detected as an anomaly.

[1462] When an anomaly is detected, the server generates an alert message, which includes details of the anomalous transaction, the type of anomaly, and a prompt to notify family members. Furthermore, a sentiment analysis engine recognizes the user's emotions and tailors the notification accordingly.

[1463] Sentiment analysis engine processing

[1464] The sentiment analysis engine recognizes emotions from the user's facial expressions, voice, text input, etc. The sentiment analysis engine performs the following tasks:

[1465] 1. Emotion recognition: Recognize the user's emotions in real time through the device's camera, microphone, and input devices.

[1466] 2. Sentiment analysis: Send feedback to the server based on the recognized emotion (surprise, anxiety, confusion, etc.).

[1467] 3. Notification adjustment: If the user feels anxious or confused, we recommend notifying their family members. We may also change the notification content to be more gentle.

[1468] Terminal handling

[1469] The device receives the alert message sent by the server and notifies the user. The notification is coordinated as follows:

[1470] 1. Normal notification: If an abnormal transaction is detected, the user will be notified with detailed information.

[1471] 2. Tailored notifications: Based on the results of the sentiment analysis engine, notifications are sent in a gentler tone. For example, a message like, "I'm a little worried, are you okay?"

[1472] 3. Recommend notifying family: If the user is feeling anxious or confused, the option to recommend notifying family will be automatically selected.

[1473] The user checks the alert message and selects whether to notify their family. If the user allows notification to their family, the information is sent from the device to the server.

[1474] Additional Server Processing

[1475] After receiving the user's authorization, the server retrieves the family member's contact information from a database and sends an alert message to the family member, including details of the abnormal transaction, the user's selection, and suggestions for how to proceed.

[1476] Specific examples

[1477] For example, if a user's transaction history shows a utility bill of 100,000 yen, it is considered to be within the normal expected range and is not detected as an anomaly. However, if a utility bill transaction of 500,000 yen occurs, it is detected as an anomaly and the user is notified. If the user is concerned, their family members will also be notified.

[1478] Prompt sentence for generative AI model

[1479] Create a program for a system that monitors elderly people's financial transactions and detects and notifies them of any abnormal transactions. The system will include a server that collects and analyzes transaction information and detects abnormalities, adjusts notifications based on emotion recognition, and notifies family members. Please also specify the specific names of the APIs and libraries to be used. For example, transaction information could be obtained from a financial institution's API, and notifications could be sent using a communication API.

[1480] This system allows seniors to manage transaction anomalies themselves while ensuring their financial security. The adoption of a sentiment analysis engine also reduces the user's psychological burden, allowing family members to quickly intervene when necessary, further enhancing the security of seniors' daily lives.

[1481] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1482] Step 1:

[1483] The server obtains transaction information for the elderly person's financial account through the financial institution's API.

[1484] (Input) Transaction information (transaction date, amount, type, and counterparty) from the financial institution's API

[1485] (Output) Save the acquired transaction information in the database

[1486] (Specific operation) The server periodically sends API requests to obtain and store transaction information.

[1487] Step 2:

[1488] The server analyzes the transaction information it acquires and detects any abnormal transactions.

[1489] (Input) Transaction information stored in the database

[1490] (Output) A list of transactions that were determined to be anomalous

[1491] (Specific operation) The server calculates the average value and standard deviation of fixed costs, and detects an anomaly when new transaction information exceeds the predicted range or when a large transaction exceeding a set threshold occurs.

[1492] Step 3:

[1493] If the server detects an abnormality, it generates an alert message.

[1494] (Input) A list of transactions that are determined to be abnormal

[1495] (Output) Alert message

[1496] (Specific Operation) The server generates an alert message containing details of the abnormal transaction and the type of abnormality.

[1497] Step 4:

[1498] The server uses a sentiment analysis engine to recognize the user's sentiment.

[1499] (Input) User facial expressions, voice, and text input obtained from the device's camera, microphone, and input device

[1500] (Output) User emotion data (surprise, anxiety, confusion, etc.)

[1501] (Specific operation) The emotion analysis engine analyzes the user's emotions in real time and feeds the results back to the server.

[1502] Step 5:

[1503] The server adjusts the notification content based on the user's emotions.

[1504] (Input) User emotion data, alert message

[1505] (Output) Adjusted alert message

[1506] (Specific operation) If the user feels anxious or confused, the server changes the notification content to softer language and generates an alert message recommending that the user notify family members.

[1507] Step 6:

[1508] The terminal receives the alert message sent from the server and notifies the user.

[1509] (Input) Adjusted alert message

[1510] (Output) User notification

[1511] (Specific operation) The device notifies the user in the form of a pop-up or a voice message.

[1512] Step 7:

[1513] The user can choose whether to notify their family.

[1514] (Input) Alert message content

[1515] (Output) Allow or deny notification to family members

[1516] (Specific operation) The user checks the alert message and selects "Notify family" or "Do not notify" via the device.

[1517] Step 8:

[1518] The server sends notifications to family members based on the user's permission.

[1519] (Input) User notification permission information

[1520] (Output) Alert message sent to family members

[1521] (Specific Operation) The server retrieves the family member's contact information from the database and sends an alert message containing details of the abnormal transaction to the family member's terminal.

[1522] This way, seniors' financial accounts can be kept safe while still providing timely and sensitive notifications, allowing seniors and their families to intervene quickly if needed.

[1523] 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.

[1524] 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.

[1525] 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.

[1526] [Fourth embodiment]

[1527] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1528] 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.

[1529] 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).

[1530] 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.

[1531] 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.

[1532] 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).

[1533] 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.

[1534] 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.

[1535] 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.

[1536] 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.

[1537] 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.

[1538] 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.

[1539] 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."

[1540] To implement this invention, a system is constructed that monitors transaction information in elderly people's financial accounts, detects abnormal transactions, and notifies them. The system is primarily composed of a server, terminals (such as smartphones and PCs), and users.

[1541] Server Processing

[1542] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database.

[1543] The server then analyzes the transaction information and detects abnormal transactions. Transaction anomalies are defined as follows:

[1544] 1. Fixed cost anomalies: Based on past transaction data, the average value and standard deviation of fixed costs such as utility bills and mobile phone bills are calculated. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ± 3 × standard deviation), it is detected as an anomaly.

[1545] 2. Large Transactions: If a large transaction occurs that exceeds a set threshold, it is detected as an anomaly.

[1546] If an anomaly is detected, the server sends an alert message to the user's device, containing details of the anomalous transaction and a message asking the user to confirm whether to notify family members.

[1547] Terminal handling

[1548] The terminal receives the alert message sent from the server and notifies the user. The user can then check the alert message and confirm the transaction details themselves. They can also choose whether to send notifications to their family members. If the user allows notifications to be sent to their family members, the information is sent from the terminal to the server.

[1549] User Action

[1550] The user checks the alert message displayed on the device and considers how to respond to any unusual transactions. If the transaction is suspicious, they can choose not to notify their family, but if not, it is recommended that they allow notifications. This choice is made on the device, and the result is sent to the server.

[1551] Additional Server Processing

[1552] If the user allows notifications to family members, the server will send an alert message to the family members' devices registered in advance, allowing the family members to receive detailed information about abnormal transactions and take appropriate measures.

[1553] Specific examples

[1554] Specifically, the following scenarios are possible:

[1555] 1. Normal transaction:

[1556] The server retrieves daily utility bills (10,000 yen) from the financial institution API.

[1557] This transaction is within the expected range and is not detected as an anomaly.

[1558] 2. Unusual utility bill transactions:

[1559] The server retrieves a utility bill transaction for 50,000 yen from the financial institution API.

[1560] This transaction is detected as an anomaly because it is far outside the expected range (average 10,000 yen ± 30,000 yen).

[1561] The server generates an alert and sends it to the user's terminal.

[1562] The user checks the alert on the device and allows notification to family members.

[1563] The server sends a notification to the family's device, providing them with detailed information.

[1564] 3. Large transfers:

[1565] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[1566] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[1567] The server generates an alert and notifies the user.

[1568] If the user is not aware of the incident, they can allow their family members to be notified.

[1569] The server sends a notification to the family's devices and action is taken to address the issue.

[1570] This system allows seniors to ensure their financial security while managing any irregularities in their transactions themselves, and allows family members to intervene quickly if necessary, helping to ensure the safety of seniors' daily lives.

[1571] The processing flow will be explained below.

[1572] Program processing

[1573] Server Processing

[1574] Step 1:

[1575] The server accesses the financial institution API at regular intervals (e.g., 9:00 AM every day).

[1576] The server sends a request for transaction data using authentication information for the financial institution API.

[1577] If authentication is successful, the financial institution API returns the user's transaction information.

[1578] The server stores the acquired transaction information in a database.

[1579] Step 2:

[1580] The server retrieves the latest transaction information from the database.

[1581] Filter newly acquired transaction information to identify transactions of interest.

[1582] The transactions covered include the date, transaction amount, transaction type, and counterparty.

[1583] Step 3:

[1584] The server categorizes the transaction information into fixed cost categories.

[1585] The server analyzes past transaction data to identify fixed expense categories such as utility bills and cell phone bills.

[1586] Check if the new transaction falls into the fixed cost category.

[1587] Step 4:

[1588] The server detects anomalies in fixed costs.

[1589] The server calculates the mean and standard deviation for the fixed cost categories.

[1590] If the new transaction amount exceeds the expected range (mean ± 3 x standard deviation), it is marked as an anomaly.

[1591] Step 5:

[1592] The server detects anomalies in large transactions.

[1593] Set a large amount threshold (e.g., 1 million yen) for all transactions.

[1594] If the transaction amount exceeds this threshold, it is marked as an anomaly.

[1595] Step 6:

[1596] If an anomaly is detected, the server generates an alert message.

[1597] The alert message will include details of the unusual transaction, the type of abnormality, and confirmation of permission to notify family members.

[1598] The server sends this message to the user's terminal.

[1599] Terminal handling

[1600] Step 1:

[1601] The terminal receives the alert message from the server.

[1602] Notify the user that new notifications are available.

[1603] The device will display the notification content in a pop-up or notification bar.

[1604] Step 2:

[1605] The terminal waits for the user's action.

[1606] The user opens the notification and sees the details of the alert message.

[1607] The alert message will include transaction details, a description of the anomaly, and an option to notify family members.

[1608] Step 3:

[1609] The user presses the button to allow notifications to family members.

[1610] The terminal transmits the user's authorization information to the server.

[1611] Additional Server Processing

[1612] Step 1:

[1613] The server receives the user's authorization information.

[1614] Retrieve contact information for registered family members from the database.

[1615] Step 2:

[1616] The server sends an alert message to the family.

[1617] A message is sent to the family member's device containing details of the abnormal transaction, the user's selection, and suggestions on how to proceed.

[1618] If the message is sent successfully, a record of it is saved in the database.

[1619] User Action

[1620] Step 1:

[1621] The user checks the alert message on the terminal.

[1622] Check the details of any unusual transactions to see if they apply to you.

[1623] Step 2:

[1624] The user selects whether to notify family members.

[1625] If you are not aware of the incident, allow your family to be notified.

[1626] If permission is granted, the information is sent from the terminal to the server.

[1627] Based on these specific processing steps, abnormal transactions in the elderly person's financial accounts can be monitored and necessary measures can be taken promptly.

[1628] Example 1

[1629] 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."

[1630] The increase in fraudulent and abnormal transactions in elderly people's financial accounts has created a need for detection and rapid notification. Current systems take time to acquire and analyze transaction information, and lack the means to quickly notify users and their families when an abnormality is detected. This increases the likelihood that elderly people will be exposed to financial risk, which is a major source of anxiety.

[1631] 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.

[1632] In this invention, the server includes means for periodically acquiring transaction information from the elderly person's financial account via the financial institution's API, means for storing the acquired transaction information in a database, means for analyzing the stored transaction information and detecting abnormal transactions, means for sending an alert message including details of the detected abnormal transaction to the user's terminal, means for providing an interface that allows the user to check the alert and select whether to notify family members, and means for sending a notification to the family member's terminal based on the user's permission. This makes it possible to quickly detect abnormalities in the elderly person's financial transactions and notify the user and family members in a timely manner.

[1633] "Financial institution API" stands for Application Program Interface provided by financial institutions, which is an interface used by external systems to obtain financial transaction data.

[1634] An "elderly person's financial account" refers to an account opened by an elderly person at a bank or other financial institution, through which financial transactions such as deposits, withdrawals, and transfers are conducted.

[1635] "Transaction information" is data showing the transaction history of a financial account within a specific period, and includes transaction dates, transaction amounts, transaction types, and transaction partner information.

[1636] "Database" means a computer system for systematically storing and managing acquired transaction information, allowing for easy access and analysis.

[1637] An "abnormal transaction" is a transaction that is determined to be out of the ordinary based on past transaction patterns or established thresholds, including transactions that may be fraudulent or involve abnormal amounts.

[1638] An "alert message" is a notification sent to a user when an abnormal transaction is detected, and contains details of the transaction and information urging the user to take action.

[1639] "Users" are those who use this system to receive notifications of abnormal transactions, and primarily refer to elderly people and their families.

[1640] "Terminal" refers to a device used by a user, such as a smartphone, PC, or tablet, that receives and displays alert messages from the server.

[1641] "Family devices" refer to devices such as smartphones, PCs, and tablets owned by the user's family, and alert messages are sent to these devices based on the user's permission.

[1642] To implement this invention, it is necessary to build a system that monitors transaction information in elderly people's financial accounts, detects abnormal transactions, and notifies users. This system is primarily composed of a server, terminals (such as smartphones and PCs), and users.

[1643] Server Processing

[1644] The server periodically retrieves transaction information from the elderly person's financial account via the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The retrieved information is stored in a database. Specifically, the server sends an API request at a set time each day to retrieve daily transaction data and record it in the database.

[1645] The server then analyzes the stored transaction information and detects abnormal transactions. Transaction abnormalities are defined in two patterns: fixed cost abnormalities and large transaction amounts. For fixed cost abnormalities, the server calculates the average value and standard deviation of fixed costs such as utility bills and mobile phone bills based on past transaction data. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ±3 x standard deviation), it is detected as an abnormality. On the other hand, for large transaction amounts, an abnormality is detected if a transaction that exceeds the set threshold occurs. If an abnormality is detected, the server sends an alert message to the user's device.

[1646] Terminal handling

[1647] The terminal receives the alert message sent from the server and is responsible for notifying the user. The user can then review the alert message and investigate the transaction details themselves. The interface displayed on the terminal includes an option to select whether to send notifications to family members. If the user allows notifications to be sent to family members, the terminal sends the selection to the server.

[1648] User Action

[1649] The user checks the alert message displayed on the device. If the transaction is recognizable, they can choose not to notify their family. If they are not recognizable, they can choose to allow notification to their family. This selection is sent to the server via the device.

[1650] Additional Server Processing

[1651] If the user allows notification to family members, the server will send an alert message to the pre-registered family members' devices, allowing the family members to receive detailed information about the abnormal transaction and take appropriate measures.

[1652] Specific examples

[1653] Specifically, the following scenarios are possible:

[1654] 1. Normal transaction:

[1655] The server retrieves daily utility bills (10,000 yen) from the financial institution's API.

[1656] This transaction is within the expected range and is not detected as an anomaly.

[1657] 2. Unusual utility bill transactions:

[1658] The server retrieves a utility bill transaction for 50,000 yen from the financial institution's API.

[1659] This transaction is detected as an anomaly because it is far outside the predicted range (average 10,000 yen ± 3 x 5,000 yen).

[1660] The server generates an alert message and sends it to the user's terminal.

[1661] The user confirms the alert message and allows notification to family members.

[1662] The server sends a notification to the family's device, providing them with detailed information.

[1663] 3. Large transfers:

[1664] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[1665] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[1666] The server generates an alert message and notifies the user.

[1667] If the user is not aware of the incident, they can allow their family members to be notified.

[1668] The server sends a notification to the family's devices and steps are taken to take action.

[1669] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1670] Step 1:

[1671] The server retrieves the transaction information

[1672] Input: Request to financial institution API

[1673] Specific operation: The server sends a request to the financial institution's API at a specific time interval (e.g., every day at midnight), using the authentication token to access the API and requesting transaction information for the target senior citizen's financial account.

[1674] Output: Transaction information (JSON format)

[1675] Specific data processing: Transaction information is returned from financial institutions in JSON format, including the transaction date, transaction amount, transaction type (deposit, withdrawal, transfer, etc.), and counterparty.

[1676] Step 2:

[1677] The server saves the transaction information it has acquired into a database.

[1678] Input: Retrieved transaction information (JSON format)

[1679] Specific operation: The server analyzes the acquired transaction information and extracts necessary fields, such as transaction date, transaction amount, and transaction type.

[1680] Output: Stored transaction information (specific table in database)

[1681] Specific data processing: The extracted data is properly saved in the transactions table of the database.

[1682] Step 3:

[1683] The server analyzes the transaction information and detects abnormal transactions.

[1684] Input: Historical transaction data in the database, retrieved new transaction data

[1685] Specific operation: The server calculates the average and standard deviation of fixed costs based on past transaction data. For example, using utility bill data from the past three months, the server calculates the average (e.g., 10,000 yen) and standard deviation (e.g., 5,000 yen).

[1686] Output: Abnormal transactions (list format)

[1687] Specific data calculation: If a new transaction exceeds the average value ±3 × standard deviation, the transaction is added to the list as an anomaly. Large transactions exceeding a set threshold (e.g., 1 million yen) are also added to the list as an anomaly.

[1688] Step 4:

[1689] The server generates an alert for abnormal transactions and sends it to the terminal.

[1690] Input: Detected abnormal transactions (list format)

[1691] What happens: The server creates an alert message with details of the anomalous transaction. For example, it generates a message like, "A utility bill transaction of 50,000 yen has been detected as anomalous. Does this transaction seem familiar?"

[1692] Output: Alert message (push notification to device)

[1693] Specific data processing: The generated alert message is sent to the user's device as a push notification.

[1694] Step 5:

[1695] The device receives the alert and notifies the user.

[1696] Input: The alert message sent from the server

[1697] Specific behavior: The device will display the received alert message as a pop-up notification on the screen. When the user taps the pop-up, the app will open and display more information.

[1698] Output: Notification to the user (display on screen)

[1699] Specific data display: Detailed information about the alert is displayed on the screen, and the user is given the option to "notify family members."

[1700] Step 6:

[1701] The user responds to the alert and sends their choice from the device to the server

[1702] Input: Alert message confirmation result (whether to notify family members)

[1703] Specific operation: The user checks the details of the alert and selects "Notify family members" or "Don't notify family members." The selection is sent from the device to the server.

[1704] Output: Selection (whether to notify family members)

[1705] Specific data transmission: User selections are transmitted to the server.

[1706] Step 7:

[1707] The server sends a notification to the family

[1708] Input: User's selection result (allow notification to family members)

[1709] Specific operation: The server checks the pre-registered contact information of family members, creates an alert message, and sends it to the family members' devices. For example, it generates a message such as, "An abnormal transaction has been detected in your parent's account. Please check the details."

[1710] Output: Alert message to family members (push notification to device or email)

[1711] Specific data transmission: The generated alert message will be sent to the family's device so that the family can check the detailed information.

[1712] (Application example 1)

[1713] 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."

[1714] Ensuring the security of transactions in elderly people's financial accounts is important for preventing financial damage, but it is difficult for elderly people themselves to constantly monitor their transactions for abnormalities. There is also a need for a method to quickly and reliably notify family members when abnormal transactions occur. Conventional systems lack the ability to detect abnormalities in real time and provide immediate notification, preventing family members from responding quickly. A new system is needed to resolve these issues and provide peace of mind to elderly people and their families.

[1715] 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.

[1716] In this invention, the server includes: [means for acquiring transaction information on the elderly person's financial account]; [means for analyzing the acquired transaction information and detecting abnormalities]; [means for notifying the user of detected abnormalities and requesting permission for notifying family members]; [means for sending notifications to family members based on the user's permission]; [means for sending notifications in real time when an abnormal transaction is detected]; and [means for sending similar alerts to family members]. This enables real-time detection of abnormal transactions and prompt notification, enabling the elderly person and their family members to prevent financial damage. It also provides a system that reduces the burden on the elderly person themselves and allows family members to respond quickly.

[1717] The term "elderly" generally refers to people aged 60 or over, especially those in need of social and medical support.

[1718] "Financial Account" means a deposit account or other asset management account opened by a customer at a financial institution for the purpose of managing assets.

[1719] "Transaction information" refers to detailed data regarding various financial transactions, such as deposits, withdrawals, transfers, and payments, in a financial account.

[1720] "Real-time" means that data is collected, analyzed, and communicated almost immediately, with very little time delay.

[1721] "Abnormal transactions" refer to transactions that deviate significantly from normal trading patterns, including large transactions and transactions that exceed expected ranges.

[1722] A "server" refers to a computer system that provides processing power and data to multiple clients over a network.

[1723] "Database" means a system for systematically storing and managing information in digital form.

[1724] "Notification" means a message or alert sent to inform a user or interested party of specific information.

[1725] An "alert" refers to a notification method that immediately warns the user when an abnormality or a condition requiring attention occurs.

[1726] "Family" refers to a person related by blood or marriage to an older person and who is responsible for assisting and supporting the older person in managing their life, health, and property.

[1727] To implement this invention, it is necessary to build a system consisting of a server, terminals (such as smartphones or PCs), and users. The main function of the system is to provide a mechanism for monitoring transaction information in elderly people's financial accounts and detecting and notifying abnormal transactions.

[1728] Server Processing

[1729] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and counterparty. The obtained information is stored in a database. The server uses the NumPy library to calculate the mean and standard deviation from past transaction data. It defines the mean value and prediction range (e.g., mean value ± 3 × standard deviation) of fixed costs, and detects newly obtained transactions that exceed the prediction range as anomalous transactions. Large transactions whose transaction amount exceeds a set threshold are also detected as anomalous. When an anomaly is detected, the server uses Pushbullet to send an alert message via push notification to the user's device.

[1730] Terminal handling

[1731] The device receives the alert message sent from the server and notifies the user. The user can check the alert message and check the transaction details themselves. The alert message includes details of the abnormal transaction and an option to notify family members. If the user allows notification to family members, the device sends the information to the server.

[1732] User Action

[1733] The user checks the alert message displayed on the device and decides whether or not the transaction is recognizable. If not, the user is encouraged to allow notifications to family members. This selection is made on the device, and the result is sent to the server. Based on the user's permission, the server can send alert messages to the devices of family members who have been registered in advance.

[1734] Specific examples

[1735] For example, if the average utility bill for the past six months was 10,000 yen with a standard deviation of 5,000 yen, and a new transaction of 20,000 yen occurs, the server will detect this as an anomaly. In this case, the server will send an alert to the user's smartphone stating, "Anomalous transaction detected: A utility bill transaction of 20,000 yen occurred on 2023-10-01." If the user allows notifications, similar alerts will also be sent to the smartphones of family members.

[1736] Example prompts for generative AI models

[1737] "We will design a system that monitors transaction information for elderly people's financial accounts, detects abnormal transactions, and notifies users. An abnormal transaction in the transaction data we received (for example, the average utility bill for the past six months is 10,000 yen, with a standard deviation of 5,000 yen) is one that exceeds the average value ±3 x standard deviation. We will set it up so that a push notification is sent stating, 'Abnormal transaction detected: A utility bill transaction of 50,000 yen occurred on 2023-10-01.' Please create a specific program based on this content."

[1738] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1739] Step 1:

[1740] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. The information obtained includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The input for this process is the API call, and the output is transaction information data in JSON format. The server saves the obtained data in a database.

[1741] Step 2:

[1742] The server extracts historical transaction data from the stored transaction information and calculates the mean and standard deviation of a particular fixed cost (e.g., utility bills). It uses the historical transaction data (in the database) as input and gets the mean and standard deviation as output. It uses the NumPy library to perform statistical operations.

[1743] Step 3:

[1744] The server analyzes newly acquired transaction data and determines whether the transaction falls within the expected range based on the calculated mean and standard deviation. The input is the latest transaction data, and the output is a flag indicating whether the transaction is "anomalous" or "normal." For example, if the amount of a new transaction exceeds the mean value ± 3 x standard deviation, the server detects the transaction as "anomalous."

[1745] Step 4:

[1746] The server uses Pushbullet's API to send an alert message to the user's device for transactions detected as abnormal. The alert includes details of the abnormal transaction (transaction date, amount, and counterparty) and an option to notify family members. The input is the abnormal transaction data, and the output is the alert message.

[1747] Step 5:

[1748] The device receives the alert message sent from the server and notifies the user. The input is the alert message, and the output is a push notification to the user. The user checks the content of the alert message on the device.

[1749] Step 6:

[1750] The user checks the alert message displayed on the terminal and selects how to respond to the abnormal transaction. Specifically, the user selects whether to notify family members. The input is the notified alert information, and the output is the user's selection.

[1751] Step 7:

[1752] When the user allows notifications to family members, the device sends the selection information to the server. The input is the user's selection, and the output is the notification permission data for the server.

[1753] Step 8:

[1754] Based on the user's permission, the server sends an alert message to pre-registered family members' devices. This allows family members to receive information about abnormal transactions and take prompt action. The input is notification permission data from the user, and the output is an alert message for the family members.

[1755] By following the above steps, financial transactions of elderly people can be monitored in real time, and if any abnormalities are detected, the user and their family can be notified promptly, thereby preventing financial damage.

[1756] 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.

[1757] To implement this invention, we will build a system that monitors transaction information on elderly people's financial accounts, detects and notifies abnormal transactions, and combines it with an emotion engine that recognizes user emotions. The system is primarily composed of a server, a terminal (such as a smartphone or PC), the emotion engine, and the user.

[1758] Server Processing

[1759] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, amount, type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database.

[1760] The server then analyzes the transaction information and detects abnormal transactions. Transaction anomalies are defined as follows:

[1761] 1. Fixed cost anomalies: Based on past transaction data, the average value and standard deviation of fixed costs such as utility bills and mobile phone bills are calculated. If the newly acquired transaction amount exceeds the predicted range (for example, the average value ± 3 × standard deviation), it is detected as an anomaly.

[1762] 2. Large Transactions: If a large transaction occurs that exceeds a set threshold, it is detected as an anomaly.

[1763] When an anomaly is detected, the server generates an alert message, which includes details of the anomalous transaction, the type of anomaly, and a prompt to notify family members. Furthermore, an emotion engine recognizes the user's emotions and tailors the notification accordingly.

[1764] Emotion engine processing

[1765] The emotion engine recognizes emotions from the user's facial expressions, voice, text input, etc. The emotion engine performs the following tasks:

[1766] 1. Emotion recognition: Recognizes user emotions in real time through the device's camera, microphone, and keyboard input.

[1767] 2. Sentiment analysis: Send feedback to the server based on the recognized emotion (surprise, anxiety, confusion, etc.).

[1768] 3. Notification adjustment: If the user feels anxious or confused, we recommend notifying their family members. We may also change the notification content to be more gentle.

[1769] Terminal handling

[1770] The device receives the alert message sent by the server and notifies the user. The notification is coordinated as follows:

[1771] 1. Normal notification: If an abnormal transaction is detected, the user will be notified with detailed information.

[1772] 2. Adjusted Notification: Based on the analysis results of the emotion engine, notifications are sent in a gentler tone. For example, a message like, "I'm a little worried, are you okay?"

[1773] 3. Recommend notifying family: If the user is feeling anxious or confused, the option to recommend notifying family will be automatically selected.

[1774] The user checks the alert message and selects whether to notify their family. If the user allows notification to their family, the information is sent from the device to the server.

[1775] Additional Server Processing

[1776] After receiving the user's authorization, the server retrieves the family member's contact information from a database and sends an alert message to the family member, including details of the abnormal transaction, the user's selection, and suggestions for how to proceed.

[1777] Specific examples

[1778] Specifically, the following scenarios are possible:

[1779] 1. Normal transaction:

[1780] The server retrieves daily utility bills (10,000 yen) from the financial institution API.

[1781] This transaction is within the expected range and is not detected as an anomaly.

[1782] 2. Unusual utility bill transactions:

[1783] The server retrieves a utility bill transaction for 50,000 yen from the financial institution API.

[1784] This transaction is detected as an anomaly because it is far outside the expected range (average 10,000 yen ± 30,000 yen).

[1785] The emotion engine detects the user's surprise or anxiety and generates a gentle alert recommending notifying family members.

[1786] The server generates the adjusted alert and sends it to the user's device.

[1787] The user confirms the notification and allows notification to family members.

[1788] The server sends tailored alerts to family devices, allowing them to respond quickly.

[1789] 3. Large transfers:

[1790] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[1791] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[1792] The emotion engine analyzes the user's emotions and generates appropriate notification content.

[1793] The server generates an alert containing details of the abnormal transaction and notifies the user.

[1794] If the user is not aware of the incident, they can allow their family members to be notified.

[1795] The server sends a notification to the family's devices and action is taken to address the issue.

[1796] This system allows seniors to manage transaction anomalies themselves while ensuring their financial security. The introduction of an emotion engine also reduces the psychological burden on users, allowing family members to intervene quickly if necessary, further enhancing the security of seniors' daily lives.

[1797] The processing flow will be explained below.

[1798] Server Processing

[1799] Step 1:

[1800] The server accesses the financial institution API at regular intervals (e.g., 9:00 AM every day).

[1801] The server sends a request for transaction data using authentication information for the financial institution API.

[1802] If authentication is successful, the financial institution API returns the user's transaction information.

[1803] The server stores the acquired transaction information in a database.

[1804] Step 2:

[1805] The server retrieves the latest transaction information from the database.

[1806] The server filters the newly acquired transaction information to identify eligible transactions.

[1807] The transactions covered include the date, transaction amount, transaction type, and counterparty.

[1808] Step 3:

[1809] The server categorizes the transaction information into fixed cost categories.

[1810] The server analyzes past transaction data to identify fixed expense categories such as utility bills and cell phone bills.

[1811] Check if the new transaction falls into the fixed cost category.

[1812] Step 4:

[1813] The server detects anomalies in fixed costs.

[1814] The server calculates the mean and standard deviation for the fixed cost categories.

[1815] If the new transaction amount exceeds the expected range (mean ± 3 x standard deviation), it is marked as an anomaly.

[1816] Step 5:

[1817] The server detects anomalies in large transactions.

[1818] The server sets a large amount threshold (e.g., 1 million yen) for all transactions.

[1819] If the transaction amount exceeds this threshold, it is marked as an anomaly.

[1820] Step 6:

[1821] If an anomaly is detected, the server generates an alert message.

[1822] The alert message will include details of the abnormal transaction, the type of abnormality, and permission to notify family members.

[1823] Emotion engine processing

[1824] Step 1:

[1825] The emotion engine captures the user's facial expressions and voice in real time.

[1826] Data is collected through cameras and microphones.

[1827] Step 2:

[1828] The emotion engine analyzes the acquired data and recognizes the user's emotions.

[1829] The emotion engine uses facial recognition algorithms and voice analysis to identify the user's emotions.

[1830] If the user's emotion corresponds to anxiety, surprise, confusion, etc., the information is sent to the server.

[1831] Step 3:

[1832] The emotion engine sends feedback to the server to adjust the notification content.

[1833] If the user is feeling anxious or confused, change the notification message to a more gentle one.

[1834] An option to recommend notifying family members will also be added.

[1835] Terminal handling

[1836] Step 1:

[1837] The terminal receives the alert message from the server.

[1838] Notify the user that new notifications are available.

[1839] The device will display the notification content in a pop-up or notification bar.

[1840] Step 2:

[1841] The terminal waits for the user's action.

[1842] The user opens the notification to view the details of the alert message.

[1843] The message will include transaction details, a description of the anomaly, and an option to notify family members.

[1844] Step 3:

[1845] When the user presses the button to allow notifications to family members,

[1846] The terminal transmits the user's authorization information to the server.

[1847] Additional Server Processing

[1848] Step 1:

[1849] The server receives the user's authorization information.

[1850] The server retrieves contact information for registered family members from the database.

[1851] Step 2:

[1852] The server sends an alert message to the family.

[1853] A message is sent to the family member's device containing details of the abnormal transaction, the user's selection, and suggestions on how to proceed.

[1854] If the message is sent successfully, a record of it is saved in the database.

[1855] Specific processing examples

[1856] Specifically, the following scenarios are possible:

[1857] 1. Normal transaction:

[1858] The server retrieves daily utility bills (10,000 yen) from the financial institution API.

[1859] This transaction is within the expected range and is not detected as an anomaly.

[1860] 2. Unusual utility bill transactions:

[1861] The server retrieves a utility bill transaction for 50,000 yen from the financial institution API.

[1862] This transaction is detected as an anomaly because it is far outside the expected range (average 10,000 yen ± 30,000 yen).

[1863] The emotion engine detects the user's surprise or anxiety and generates a gentle alert recommending notifying family members.

[1864] The server generates the adjusted alert and sends it to the user's device.

[1865] The user confirms the notification and allows notification to family members.

[1866] The server sends tailored alerts to family devices, allowing them to respond quickly.

[1867] 3. Large transfers:

[1868] The server retrieves a remittance transaction of 2 million yen from the financial institution API.

[1869] This transaction exceeds the threshold (1 million yen), so it is automatically detected as an anomaly.

[1870] The emotion engine analyzes the user's emotions and generates appropriate notification content.

[1871] The server generates an alert containing details of the abnormal transaction and notifies the user.

[1872] If the user is not aware of the incident, they can allow their family members to be notified.

[1873] The server sends notifications to family members' devices so that countermeasures can be taken quickly.

[1874] Example 2

[1875] 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."

[1876] There is a need for a method and system that can respond quickly and appropriately when abnormal financial transactions occur among elderly people. Furthermore, it is necessary to consider the user's feelings when notifying them of abnormal transactions. Conventional systems have had problems such as causing unnecessary anxiety to users and delaying notification to family members. The purpose of this invention is to solve these problems and simultaneously ensure the financial security and psychological peace of mind of elderly people.

[1877] The identification process by the identification processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means. In this invention, the server includes a means for acquiring transaction information of the elderly person's financial account, a means for analyzing the acquired transaction information and detecting abnormalities, and a means for recognizing the user's emotions and adjusting the notification content. This makes it possible to notify the user of abnormal transactions in a manner that takes into consideration the user's emotions, and also to notify family members at an appropriate time.

[1878] "Elderly" refers to people who are older, especially those aged 65 or older, and who require special consideration in financial transactions and daily life.

[1879] "Financial account" is a general term for deposit accounts, savings accounts, fixed-term deposit accounts, etc. opened at banks, credit unions, and other financial institutions, and is a basic means for recording and managing financial transactions.

[1880] "Transaction information" refers to detailed data relating to financial transactions such as deposits, withdrawals, transfers, and debits related to a financial account, including the date, amount, type, and counterparty of the transaction.

[1881] "Abnormal transactions" refer to transactions that significantly deviate from normal trading patterns, including, for example, transactions that exceed the mean and standard deviation of a particular fixed cost or large transactions that exceed a set threshold.

[1882] "Emotion engine" is a general term for software or hardware that recognizes and analyzes emotions in real time from a user's facial expressions, voice, text input, etc.

[1883] "Adjusting the notification content" refers to a process of changing the wording and content of the notification message depending on the user's emotional state, so as not to make the user feel excessively anxious.

[1884] "Family" refers to a person who has a certain kinship relationship with the user and who has the role of supporting the user in emergencies or abnormal situations.

[1885] "Server" means the main computer system that manages the entire system and acquires, analyzes, and stores transaction information, and generates and transmits notifications.

[1886] A "terminal" refers to a device such as a smartphone or PC that is directly operated by a user and that receives notifications sent from a server.

[1887] "Notification" refers to a message sent by the system to the user or family member informing them of an abnormal transaction, and may be sent in the form of voice, text, push notification, etc.

[1888] To implement this invention, it is necessary to build a system that monitors transaction information on elderly people's financial accounts, detects abnormal transactions, and notifies them, as well as an emotion engine that recognizes user emotions. This system is primarily composed of a server, a terminal (such as a smartphone or PC), an emotion engine, and a user.

[1889] Server Processing

[1890] The server periodically obtains transaction information for the elderly person's financial account through the financial institution's API. This transaction information includes the transaction date, transaction amount, transaction type (deposit, withdrawal, transfer, etc.), and the transaction partner. The obtained information is stored in a database on the server. The server analyzes this data and runs an algorithm to detect abnormal transactions.

[1891] Specifically, the server detects abnormal transactions through the following steps:

[1892] 1. Fixed cost anomaly detection: Based on past transaction data, the average value and standard deviation of fixed costs such as utility bills and mobile phone bills are calculated. If the newly acquired transaction amount exceeds the predicted range of the average value ±3 x standard deviation, it is detected as an anomaly.

[1893] 2. Large transaction detection: The server detects an abnormality when a large transaction that exceeds a set threshold occurs.

[1894] When an anomaly is detected, the server sends feedback to the emotion engine, and also generates alert messages and adjusts the notification content as needed.

[1895] Emotion engine processing

[1896] The emotion engine recognizes emotions in real time from the user's facial expressions, voice, text input, etc. For this purpose, the device requires input devices such as a camera and microphone. The emotion engine provides the following functions:

[1897] 1. Emotion recognition: Recognizes the user's emotions through the device's camera, microphone, and keyboard input.

[1898] 2. Sentiment analysis: Send feedback to the server based on the recognized emotion (surprise, anxiety, confusion, etc.).

[1899] 3. Notification adjustment: If the user feels anxious or confused, we recommend notifying their family members. We may also change the notification content to be more gentle.

[1900] Terminal handling

[1901] The device receives the alert message sent by the server and notifies the user. The notification is arranged in the following pattern:

[1902] 1. Normal notification: If an abnormal transaction is detected, the user will be notified of the details.

[1903] 2. Adjusted Notification: Based on the analysis results of the emotion engine, notifications are sent in a gentler tone. For example, a message like, "I'm a little worried, are you okay?"

[1904] 3. Recommend notifying family: If the user is feeling anxious or confused, the option to recommend notifying family will be automatically selected.

[1905] The user checks the alert message and selects whether to notify their family. If the user allows notification to their family, the information is sent from the device to the server.

[1906] Additional Server Processing

[1907] After receiving the user's authorization, the server retrieves the family member's contact information from a database and sends an alert message to the family member, including details of the abnormal transaction, the user's selection, and suggestions for how to proceed.

[1908] Specific examples

[1909] The following example further illustrates the operation of the system:

[1910] Regular transactions

[1911] The server retrieves a daily utility bill (10,000 yen) from a financial institution API. This transaction is within the expected range and is not detected as an anomaly.

[1912] Unusual utility bill transactions

[1913] The server receives a utility bill transaction of 50,000 yen. This transaction is detected as an anomaly because it is far outside the predicted range (average 10,000 yen ± 30,000 yen). The emotion engine detects the user's surprise or anxiety and generates an alert in a gentle manner recommending notifying family members. The server generates an adjusted alert and sends it to the user's device. The user confirms the notification and allows the notification to family members. The server then sends the adjusted alert to the family members' devices, allowing them to respond quickly.

[1914] Large remittances

[1915] The server receives a remittance transaction of 2 million yen. Because this transaction exceeds the threshold (1 million yen), it is automatically detected as an anomaly. The emotion engine analyzes the user's emotions and generates appropriate notification content. The server generates an alert with details of the anomalous transaction and sends it to the device. If the user does not recognize the transaction, they can allow notification to family members. The server sends a notification to the family member's device, and action is taken to take measures.

[1916] Prompt Sentence Examples

[1917] By inputting the following prompts into the generative AI model, we can build specific system functions:

[1918] "Create a system that collects transaction information from senior citizens' financial accounts, detects abnormal transactions, and notifies them. Add a function that recognizes the user's emotions and adjusts the content of notifications accordingly."

[1919] This system allows seniors to manage transaction anomalies themselves while ensuring their financial security. The introduction of an emotion engine also reduces the psychological burden on users, allowing family members to intervene quickly if necessary, further enhancing the security of seniors' daily lives.

[1920] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1921] Step 1:

[1922] The server obtains transaction information for the elderly person's financial account through the financial institution's API. The input is the request information to the API, and the output is the obtained transaction details. Specifically, the server sends a request to the endpoint "https: / / api.bank.com / transactions" and receives the transaction data in JSON format.

[1923] Step 2:

[1924] The server saves the acquired transaction information in a database. The input is the transaction details data acquired in step 1, and the output is the transaction record saved in the database. Specifically, the server uses the INSERT statement to insert the transaction data it received into the "transactions" table in the database called "transaction_db".

[1925] Step 3:

[1926] The server analyzes the transaction information and detects abnormal transactions. The input to this step is the transaction data stored in the database. The output is the abnormal transaction detection results. The server processes the data in the following steps:

[1927] 1. Fixed cost anomaly detection: The server calculates the average and standard deviation of fixed costs such as utility bills and mobile phone bills based on past transaction data. Specifically, it executes the SQL query "SELECT AVG(amount), STDDEV(amount) FROM transactions WHERE type='utility'" to obtain the results. If the newly acquired transaction amount exceeds the predicted range of the average value ±3 x standard deviation, it is detected as an anomaly.

[1928] 2. Detection of large transactions: If a large transaction that exceeds a set threshold (for example, 1 million yen) occurs, the server detects it as an anomaly.

[1929] Step 4:

[1930] If an anomaly is detected, the server generates an alert message. The input to this step is the abnormal transaction detection result obtained in step 3. The output is the generated alert message. Specifically, the server generates the message "Your electricity bill this month is unusually high. Please review it."

[1931] Step 5:

[1932] The emotion engine recognizes emotions in real time from the user's facial expressions, voice, text input, etc. The inputs for this step are the user's facial images, voice data, and text input obtained from the device. The output is recognized emotion data. Specifically, the device captures the user's facial expressions using the camera and analyzes them using an emotion recognition model (e.g., Google Cloud Vision API). It also recognizes audio from the microphone and analyzes the audio data using an emotion analysis API (e.g., IBM Watson Tone Analyzer).

[1933] Step 6:

[1934] The emotion engine adjusts the alert message based on the recognized emotion data. The input to this step is the emotion data obtained in step 5 and the alert message generated in step 4. The output is the adjusted alert message. Specifically, the server changes the message to a gentler one, such as "Your electricity bill seems a little high. Please check it," based on feedback from the emotion engine.

[1935] Step 7:

[1936] The device notifies the user of the adjusted alert message. The input to this step is the alert message adjusted in step 6. The output is a notification to the user. Specifically, the device uses a push notification to display the message "I'm a little worried, are you okay?"

[1937] Step 8:

[1938] The user checks the alert message and selects whether to notify family members. The input to this step is the alert message displayed in step 7. The output is the user's selection information. Specifically, the user clicks the "Notify family members" button displayed on the terminal.

[1939] Step 9:

[1940] The server receives the user's permission information and sends an alert message to the family. The input to this step is the user's selection information obtained in step 8. The output is an alert message to the family. Specifically, the server executes the SQL query "SELECT contact_info FROM family_contacts WHERE user_id='12345'" to obtain the family's contact information and sends a message via email or SMS saying, "Your parent has an unusual transaction. Please check on them."

[1941] (Application example 2)

[1942] 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."

[1943] In modern society, there has been an increase in fraudulent access to elderly people's financial accounts and fraudulent transactions. This not only threatens the financial security of elderly people, but also increases psychological burdens such as anxiety and confusion. Current systems are capable of detecting and notifying abnormal transactions, but they are unable to respond appropriately based on the user's emotions, which can cause excessive stress to the user. In addition, notification to family members is not incorporated into the process, resulting in a lack of cooperation between elderly people and their families.

[1944] 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.

[1945] In this invention, the server includes: [means for acquiring transaction information for the elderly person's financial account;] [means for analyzing the acquired transaction information and detecting abnormalities;] [means for notifying the user of the detected abnormality, adjusting the content of the notification using emotion analysis means that recognizes the user's emotion, and requesting permission for notifying family members; and [means for sending notifications to family members based on the user's permission.] This ensures the safety of the elderly person's financial account while realizing appropriate notifications that take emotion into consideration, and enables the elderly person and their family members to intervene quickly if necessary.

[1946] "Elderly" refers to people over a certain age, primarily those considered at high risk of financial fraud and unauthorized access.

[1947] "Financial account" refers to an account opened with a bank or financial institution for the purpose of depositing money or conducting transactions.

[1948] "Transaction Information" refers to data regarding transaction dates, amounts, types, and counterparties obtained through a financial institution's API.

[1949] "Analysis" refers to the process of evaluating acquired transaction information and detecting anomalies based on patterns and regularities.

[1950] An "abnormal transaction" refers to a transaction that is deemed to be unnatural and different from the norm, and requires safety confirmation.

[1951] "User" refers to the elderly person who uses this system or someone acting on their behalf.

[1952] "Emotion analysis means" refers to technology for recognizing a user's psychological state from facial expressions, voice, text input, etc.

[1953] "Adjusting notification content" refers to softening or emphasizing alert or notification content based on the user's current emotional state.

[1954] "Permission" refers to the act of a user deciding whether or not to send notifications to family members.

[1955] "Family" refers to the user's children, siblings, and other close relatives who are responsible for providing support as needed.

[1956] A "server" refers to a computer system that primarily acquires, analyzes, stores, and notifies data.

[1957] To implement this invention, it is necessary to bui...

Claims

1. a means for obtaining transaction information for the senior citizen's financial account; A means for analyzing the acquired transaction information and detecting anomalies; means for notifying the user of the detected abnormality and requesting permission to notify family members; means for sending notifications to family members based on the user's permission; A system including:

2. 2. The system of claim 1, further comprising means for calculating the mean and standard deviation of a particular fixed cost from the transaction information and detecting transactions that exceed the expected range as an anomaly.

3. The system according to claim 1 , further comprising means for detecting an abnormality when a transaction amount in the transaction information exceeds a set threshold.

Images