system

Abstract

Provide a system. A device for acquiring biometric data; means for pre-processing the acquired biometric data; means for extracting features from the preprocessed biometric data; A means for inputting the extracted features into an artificial intelligence model to obtain an inference result; a means for outputting the obtained inference result and notifying the user; A system including:

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] Conventional artificial intelligence systems are unable to reflect the user's physical sensations and intuition, making it difficult to make advanced decisions based on animal instincts or the intuition of experienced experts. Furthermore, even in effective earthquake prediction and investment decision-making, accuracy is insufficient because only mechanical data is used. For this reason, there has been a demand for a system that utilizes the user's biometric information to enable more accurate predictions and decision-making. [Means for solving the problem]

[0005] To solve the above-mentioned problems, the present invention provides a system having the following configuration. First, a means for acquiring biometric data is provided. This means acquires data such as heart rate, electrodermal response, and hormone levels in real time. Next, a means for preprocessing the acquired biometric data is provided. Preprocessing involves noise removal and normalization of the data. Furthermore, a means for extracting features from the preprocessed biometric data is provided. For example, sudden fluctuations in heart rate and peaks in electrodermal response are extracted. Then, a means for inputting the extracted features into an artificial intelligence model and obtaining inference results is provided. Finally, a means for outputting the obtained inference results and notifying the user is provided, making it easier for the user to intuitively understand the results. This enables advanced decision-making that reflects the user's physical information, thereby increasing the success rate of earthquake prediction and investment.

[0006] "Biometric data" refers to various data related to the human body, such as heart rate, galvanic skin response, and hormone levels.

[0007] The "preprocessing means" refers to a processing method for performing data adjustments such as noise removal and normalization on the acquired biometric data.

[0008] "Feature extraction means" refers to algorithms or methods that detect significant patterns or variations in pre-processed biometric data.

[0009] An "artificial intelligence model" refers to a model that uses technologies such as machine learning and deep learning to analyze data and make predictions.

[0010] "Means for obtaining inference results" refers to a method for inputting feature data into an artificial intelligence model and outputting predictions or decision-making results as a result.

[0011] "Means for notifying" refers to a method for conveying the inference results to the user visually, audibly, or tactilely.

[0012] "Noise removal" refers to the process of removing unnecessary information and errors contained in biometric data.

[0013] "Normalization" refers to the process of standardizing the range of data, typically scaling it to the range from 0 to 1.

[0014] A "feature" is a data point that indicates an important pattern or indicator, such as a sudden change in heart rate or a peak in electrodermal response. [Brief explanation of the drawings]

[0015] [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

[0016] 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.

[0017] First, the terms used in the following description will be explained.

[0018] 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).

[0019] 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.

[0020] 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.

[0021] 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.

[0022] 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."

[0023] [First embodiment]

[0024] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0025] 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.

[0026] 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).

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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.

[0031] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] 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."

[0036] This invention relates to a system that collects a user's physical information in real time and uses an artificial intelligence (AI) model based on that information to make advanced decisions. In particular, accuracy can be improved in earthquake prediction and investment decision-making by utilizing physical sensations and intuition. Specific embodiments for carrying out the present invention are described below.

[0037] System Configuration

[0038] The system consists of the following main components:

[0039] 1. Sensor device (terminal) - Acquires real-time physical information such as the user's heart rate, galvanic skin response, and hormone levels.

[0040] 2. Data transmission module (terminal) - transmits the acquired biometric data to the server.

[0041] 3. Data preprocessing module (server) - Performs preprocessing such as noise removal and normalization on the received raw data.

[0042] 4. Feature Extraction Module (Server) - Extracts important features from the preprocessed data.

[0043] 5. AI model (server) - receives extracted features as input and outputs inference results.

[0044] 6. Result notification module (terminal) - Notifies the user of the inference results.

[0045] Processing flow

[0046] 1. Biometric data acquisition

[0047] The device captures data such as the user's heart rate, galvanic skin response, and hormone levels every second, using wearable devices and sensors attached to the skin.

[0048] 2. Data transmission

[0049] The device transmits the acquired raw data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[0050] 3. Data Preprocessing

[0051] The server performs filtering operations on the received raw data to remove noise, for example, smoothing the heart rate data using a moving average filter, and normalization, scaling the data to a range from 0 to 1.

[0052] 4. Feature Extraction

[0053] The server extracts features from the preprocessed data, such as sudden fluctuations in heart rate and peaks in skin galvanic responses, thereby capturing the user's physical reactions to precursors to earthquakes or sudden market fluctuations.

[0054] 5. Linking to AI models

[0055] The server inputs the extracted feature data into the AI ​​model, which compares it with the data it has learned so far to determine the probability of an earthquake occurring and whether to "buy," "sell," or "hold" in investment decisions.

[0056] 6. Output and notification of results

[0057] The server receives the inference results of the AI ​​model, converts them into a format that is easy for the user to understand, and sends the results to the device, which then notifies the user in various ways, such as by sound alert, vibration, or display on the screen.

[0058] Specific examples

[0059] Examples of earthquake prediction

[0060] 1. Acquisition of physical information

[0061] The device captures the user's heart rate and electrical skin response every second and sends it to a server.

[0062] 2. Data Preprocessing

[0063] The server removes noise from the received data using a moving average filter and normalizes it.

[0064] 3. Feature Extraction

[0065] The server detects rapid fluctuations in heart rate and peaks in electrodermal response.

[0066] 4. Linking to AI models

[0067] The server inputs these features into an earthquake prediction AI model to infer the probability of an earthquake occurring.

[0068] 5. Outputting the results

[0069] The server converts the probability of an earthquake occurring into "high risk," "medium risk," or "low risk," and sends it to the terminal.

[0070] 6. User Notices

[0071] If the risk is high, the device will notify the user, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[0072] Investment Decision Examples

[0073] 1. Acquisition of physical information

[0074] The device measures the user's heart rate, galvanic skin response, and hormone levels every second and transmits them to a server.

[0075] 2. Data Preprocessing

[0076] The server processes this data for noise reduction and normalization.

[0077] 3. Feature Extraction

[0078] The server detects heart rate variability, peaks in galvanic skin response, and spikes in hormone levels.

[0079] 4. Linking to AI models

[0080] The server inputs these characteristics into an investment decision-making AI model and infers the decision to "buy," "sell," or "hold."

[0081] 5. Outputting the results

[0082] The server receives the inference results, converts them into a format such as "You should buy XX," and sends them to the terminal.

[0083] 6. User Notices

[0084] The terminal notifies the user of the results and suggests specific actions to take.

[0085] The above is a specific embodiment for carrying out the present invention. This system enables advanced decision-making using the user's physical information, thereby increasing the probability of success in earthquake prediction and investment.

[0086] The processing flow will be explained below.

[0087] Step 1:

[0088] Acquiring physical information

[0089] The device collects real-time biometric data such as the user's heart rate, galvanic skin response, and hormone levels using wearable and skin-attached sensors.

[0090] Step 2:

[0091] Sending data

[0092] The device transmits the acquired biometric data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[0093] Step 3:

[0094] Receiving data

[0095] The server receives the biometric data sent from the device and stores it in chronological order.

[0096] Step 4:

[0097] Noise Reduction

[0098] The server performs filtering on the received raw data to remove noise, for example, applying a moving average filter to smooth the heart rate data.

[0099] Step 5:

[0100] Data normalization

[0101] The server normalizes the denoised data, scaling the heart rate data to a range of 0 to 1 to make it consistent with other biometric data.

[0102] Step 6:

[0103] Feature Extraction

[0104] The server extracts important features from the normalized data, applying algorithms to detect, for example, sudden fluctuations in heart rate or peaks in electrodermal response.

[0105] Step 7:

[0106] Entering feature data

[0107] The server inputs the extracted feature data into an AI model that has been trained for earthquake prediction and investment decision-making.

[0108] Step 8:

[0109] AI model inference

[0110] The server inputs the characteristic data into the AI ​​model and performs inference to output the probability of an earthquake occurring and investment decisions such as "buy," "sell," or "hold."

[0111] Step 9:

[0112] Receiving inference results

[0113] The server receives the inference results from the AI ​​model and converts them into a format that is easy for users to understand.

[0114] Step 10:

[0115] Sending the results

[0116] The server then sends the converted inference results to the terminal, which then provides the results to the user as an alert message.

[0117] Step 11:

[0118] User Notification

[0119] The device notifies the user of the inference results by means of voice alerts, vibrations, screen displays, etc., and prompts the user to take specific action.

[0120] The above is a specific flow of the program processing of the present invention, which enables advanced decision-making in real time using the user's physical information.

[0121] Example 1

[0122] 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."

[0123] Existing systems have a problem with low accuracy in earthquake prediction and investment decision-making using users' biometric data. This is because the processes of biometric data acquisition, preprocessing, feature extraction, inference, and notification are performed separately, reducing the overall system efficiency. Another problem is that insufficient noise removal and normalization of biometric data can easily produce inaccurate inference results. To solve these problems, an advanced integrated system is required.

[0124] 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.

[0125] In this invention, the server includes means for acquiring biometric data, means for preprocessing the acquired biometric data, means for extracting features from the preprocessed biometric data, means for inputting the extracted features into an artificial intelligence model to obtain an inference result, means for outputting the obtained inference result and notifying the user, means for using a wearable device or a sensor attached to the skin, means for transmitting data using wireless communication technology such as Bluetooth or Wi-Fi, means for smoothing data using a moving average filter, means for normalizing data to a range from 0 to 1, and means for converting the inference result into "high risk," "medium risk," or "low risk." This enables advanced decision-making using a user's biometric data to be performed accurately in real time.

[0126] "Biometric data" is information obtained from a user's body, including heart rate, galvanic skin response, hormone levels, and the like.

[0127] "Preprocessing" refers to applying filtering, normalization, and other processes to acquired biometric data to remove noise and convert the data into a format that is easier to analyze.

[0128] "Feature extraction" is the process of finding important patterns and variations in pre-processed data and preparing it for input into an artificial intelligence model.

[0129] An "artificial intelligence model" is a software system that uses machine learning algorithms to analyze data and draw inferences.

[0130] "Inference results" are the conclusions or judgments that an AI model provides based on biometric data, such as the probability of an earthquake occurring or decisions about whether to "buy," "sell," or "hold" an investment.

[0131] "Notification" is a means of conveying the inference results to the user, and is implemented in the form of an alert sound, vibration, screen display, etc.

[0132] A "wearable device" is a device worn by a user that can acquire biometric data, such as a smartwatch or activity tracker.

[0133] A "sensor" is a device for measuring biometric data, and includes types that are attached to the skin and types that are worn on the arm.

[0134] "Wireless communication technology" refers to technology for sending and receiving data wirelessly, and includes Bluetooth and Wi-Fi.

[0135] A "moving average filter" is a filtering method used to smooth out fluctuations in data, and is a method of removing noise by taking the average over time.

[0136] "Normalization" is a process performed to unify the scale of data, and involves converting the scale of biometric data to a range of 0 to 1.

[0137] "Risk conversion" is the process of converting inference results into categories such as "high risk," "medium risk," and "low risk," and presenting them in a form that is easy for users to understand.

[0138] The system of the present invention is configured to collect biometric data of a user in real time and make advanced decisions using an artificial intelligence (AI) model. Detailed embodiments of the system are described below.

[0139] System Configuration

[0140] This system mainly consists of the following hardware and software components.

[0141] 1. Sensor device (terminal)

[0142] Sensor devices are used to obtain real-time biometric data such as a user's heart rate, electrodermal response, and hormone levels. Specific examples include wearable devices (smartwatches, activity trackers) and sensors attached to the skin. For example, Apple Watch and general skin electrode sensors can be used.

[0143] 2. Data transmission module (terminal)

[0144] The data transmission module is responsible for transmitting the biometric data acquired from the sensor device to the server, using wireless communication technologies such as Bluetooth and Wi-Fi.

[0145] 3. Data preprocessing module (server)

[0146] The data preprocessing module performs noise removal and normalization on the biometric data received on the server, specifically smoothing the data using a moving average filter and scaling the biometric data to a range of 0 to 1.

[0147] 4. Feature extraction module (server)

[0148] The feature extraction module extracts important features from the pre-processed data, such as rapid heart rate fluctuations or peaks in electrodermal response.

[0149] 5. AI model (server)

[0150] This module inputs the extracted feature data into the AI ​​model to obtain inference results. It uses a deep learning algorithm to compare the results with past data and make inferences.

[0151] 6. Result notification module (terminal)

[0152] This module notifies the user of the inference results. Notifications are given in the form of an alert sound, vibration, screen display, etc.

[0153] Specific examples

[0154] Examples of earthquake prediction

[0155] When a user uses a sensor device (e.g., a smartwatch), their heart rate and electrodermal response are acquired every second. This data is sent to a server via Bluetooth. The server smooths the heart rate data using a moving average filter and normalizes the data. The server then detects sudden heart rate fluctuations and peaks in electrodermal response from the preprocessed data. This feature data is input into an AI earthquake prediction model to infer the probability of an earthquake occurring. Finally, the server converts the inference result into "high risk," "medium risk," or "low risk" and sends it to the device. If the risk is high, the device notifies the user, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[0156] Investment Decision Examples

[0157] The user uses the same sensor device to measure their heart rate, galvanic response, and hormone levels every second. This data is sent to a server, which then denoises and normalizes it. Heart rate fluctuations, peaks in galvanic response, and spikes in hormone levels are then detected. This characteristic data is then input into an AI investment decision-making model, which then infers whether to "buy," "sell," or "hold." The server then converts this inference result into a concrete format, such as "You should buy XX," and sends it to the device. The device then notifies the user, "You should buy XX," and suggests a specific action to take.

[0158] Prompt Sentence Examples

[0159] Example prompts to be input to the generative AI model:

[0160] The user's heart rate fluctuates rapidly and their electrodermal response shows a peak. These can be fed into an AI model to assess the risk of an earthquake occurring.

[0161] This system enables advanced decision-making in real time using the user's biometric data, increasing the chances of successful earthquake prediction and investment.

[0162] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0163] Step 1:

[0164] Biometric data acquisition

[0165] Subject: Device

[0166] Description: The device uses sensor devices to collect real-time biometric data such as the user's heart rate, galvanic skin response, and hormone levels. Input data comes from various sensors worn by the user. The device converts this signal into digital data and updates it every second.

[0167] How it works: The device receives signals from the smartwatch and skin electrodes. For example, the optical heart rate monitor in the Apple Watch measures your heart rate beat-by-beat, and the galvanic skin response sensor captures your stress level. This data is temporarily stored on the device.

[0168] Step 2:

[0169] Sending data

[0170] Subject: Device

[0171] Description: The device transmits the acquired biometric data to the server at regular intervals (e.g., every second). The input data is the biometric data acquired from the sensor device, and the output data is the data packet sent to the server.

[0172] How it works: The device sends data to the server via Bluetooth or Wi-Fi, for example using the low-latency, highly reliable Bluetooth Low Energy (BLE) protocol to ensure that the biometric data reaches the server reliably.

[0173] Step 3:

[0174] Data Preprocessing

[0175] Subject: Server

[0176] Description: The server performs preprocessing on the received biometric data, specifically noise removal and normalization. The input data is the biometric data sent from the device, and the output data is the preprocessed data.

[0177] What it does: The server applies a moving average filter to remove spikes and noise from the data. For example, a 3-second moving average filter is used to smooth out fluctuations in heart rate data. At the same time, it normalizes the data by scaling it to a range of 0 to 1, which makes it easier to extract features and input into AI models.

[0178] Step 4:

[0179] Feature Extraction

[0180] Subject: Server

[0181] Description: The server extracts important features from preprocessed data. The input data is the preprocessed data, and the output data is the extracted feature data.

[0182] How it works: The server detects features such as sudden fluctuations in heart rate data and peaks in galvanic response (GDS). It then uses an algorithm to extract, for example, spikes in heart rate or peaks in GDS, thereby identifying moments of stress or excitement felt by the user.

[0183] Step 5:

[0184] Linking to AI models

[0185] Subject: Server

[0186] Description: The server inputs the extracted feature data into the AI ​​model and obtains the inference result. The input data is the feature data, and the output data is the inference result from the AI ​​model.

[0187] How it works: The server uses a trained deep learning model to input feature data into an AI model. The AI ​​model compares this data with past data to determine the probability of an earthquake occurring and whether to "buy," "sell," or "hold" an investment. The model outputs its inference results in real time.

[0188] Step 6:

[0189] Output and notification of results

[0190] Subject: Server

[0191] Description: The server converts the inference results of the AI ​​model into an easy-to-understand format and sends it to the terminal. The input data is the inference results from the AI ​​model, and the output data is the final result that is notified to the user.

[0192] Specific operation: The server classifies the inference results into "high risk," "medium risk," or "low risk" and sends them to the device in JSON format. The device analyzes the received data and notifies the user with a specific message. For example, it warns the user with an alert sound or vibration saying, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[0193] The above is a detailed description of the specific operations and data flow at each processing step of this system. This detailed flow enables advanced decision-making using the user's biometric data in real time.

[0194] (Application example 1)

[0195] 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."

[0196] Conventional content delivery systems primarily recommend content based on users' preferences and interests, but this approach fails to respond to users' real-time emotional states. This results in a poor user experience and reduces their willingness to consume content. Furthermore, the lack of a mechanism for capturing changes in emotional states in real time makes it difficult to provide users with appropriate feedback.

[0197] 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.

[0198] In this invention, the server includes means for acquiring biometric data, means for preprocessing the acquired biometric data, means for extracting features from the preprocessed biometric data, means for inputting the extracted features into an artificial intelligence model to obtain an inference result, means for recommending optimal content to a user based on the obtained inference result, means for notifying the user of the recommended content and monitoring emotional changes in real time as the user views the content, and means for providing feedback based on the monitored emotional changes. This enables personalized content recommendations based on the user's real-time emotional state and the provision of appropriate feedback.

[0199] Below are definitions of important terms included in the claims, tailored to application examples.

[0200] "Biometric data" refers to physical information such as a user's heart rate, galvanic skin response, and hormone levels.

[0201] "Preprocessing" refers to performing processes such as noise removal and normalization on acquired biometric data.

[0202] "Feature extraction" refers to extracting important features, such as rapid heart rate fluctuations or peaks in electrodermal responses, from preprocessed biometric data.

[0203] An "artificial intelligence model" refers to a machine learning algorithm or neural network that makes inferences or decisions based on given input data.

[0204] "Inference result" refers to the prediction or judgment result that an artificial intelligence model outputs based on input data.

[0205] "Content recommendation" refers to the process of suggesting optimal videos, music, and other media to users based on their biometric data.

[0206] "Changes in emotion" refers to fluctuations in emotional state, such as excitement, joy, or sadness, that a user experiences when viewing content.

[0207] "Providing feedback" refers to suggesting appropriate advice or next actions to the user based on monitored emotional changes.

[0208] This invention is a system that uses a user's biometric data to recommend optimal content in real time and provide appropriate feedback. This system consists of the following main components:

[0209] System Configuration

[0210] 1. Means of acquiring biometric data

[0211] Means: A wearable device for measuring the user's heart rate or galvanic skin response (e.g., a heart rate sensor or galvanic skin response sensor).

[0212] Hardware used: General wearable devices (e.g., smartwatches, fitness bands, etc.)

[0213] 2. Pretreatment Methods

[0214] Method: Denoise and normalize the acquired biometric data

[0215] Software used: Python numerical calculation libraries (e.g., numpy, pandas), data preprocessing libraries (e.g., scikit-learn)

[0216] 3. Feature Extraction Method

[0217] Method: Extract features such as heart rate variability and electrodermal response peaks from the preprocessed data.

[0218] Software used: Python machine learning library (e.g., scikit-learn)

[0219] 4. Artificial Intelligence Models

[0220] Method: Take the extracted features as input and infer the user's real-time emotional state.

[0221] Software used: Deep learning libraries (e.g., TENSORFLOW (registered trademark), Keras)

[0222] Model: Emotional Inference Model

[0223] 5. Content recommendation method based on inference results

[0224] Method: Recommending optimal content to users based on inference results from an AI model

[0225] Software used: Recommendation systems (e.g., content recommendation algorithms)

[0226] 6. User Notification Methods

[0227] Method: Notify the smartphone of the recommended content so that the user can watch it.

[0228] Hardware used: Smartphone

[0229] Software used: Notification library (e.g., Firebase Cloud Messaging)

[0230] 7. Emotion Monitoring Tools

[0231] Method: While the user is watching the content, heart rate and electrodermal response are collected again to monitor changes in emotions in real time.

[0232] Hardware used: General wearable devices

[0233] Software used: Dedicated SDK (e.g., Apple HealthKit, Fitbit SDK)

[0234] 8. Means of Providing Feedback

[0235] Method: Suggest the next content to watch or specific actions based on changes in emotions

[0236] Software used: Recommendation systems (e.g., feedback generation algorithms)

[0237] Specific examples

[0238] For example, when a user wears a smartwatch and watches a video, the smartwatch measures the user's heart rate and electrodermal response and sends the data to a smartphone. The smartphone then transmits this biometric data to a server at regular intervals. The server then preprocesses the received data and extracts features. An AI model uses this feature data to infer the user's emotional state and recommends the next optimal piece of content based on the inference. The smartphone then notifies the user of the content and provides appropriate feedback by monitoring emotional changes in real time while watching.

[0239] Prompt Sentence Examples

[0240] "Please infer the user's real-time emotional state from their heart rate and electrodermal response based on the content they are currently viewing, and recommend the best content for them to watch next."

[0241] This allows users to enjoy content that is always tailored to their emotional state, resulting in a system that provides an even more entertaining experience.

[0242] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0243] Step 1:

[0244] The user puts on the wearable device.

[0245] The device measures the user's heart rate and electrical skin response in real time, and transmits this data to a smartphone via Bluetooth or other means.

[0246] Input: User's heart rate, galvanic skin response

[0247] Data processing: Acquired from wearable devices

[0248] Output: Biometric data is sent to a smartphone

[0249] Step 2:

[0250] The smartphone receives biometric data from the wearable device and sends it to a server at regular intervals, using Wi-Fi or mobile data for communication.

[0251] Input: Biometric data (heart rate, electrodermal response)

[0252] Data processing: Buffering data at regular intervals

[0253] Output: Biometric data is sent to the server

[0254] Step 3:

[0255] The server performs noise reduction and normalization on the received biometric data, for example, smoothing the heart rate data using a moving average filter.

[0256] Input: Biometric data (heart rate, electrodermal response)

[0257] Data calculation: Noise removal (moving average filter), normalization

[0258] Output: Preprocessed biometric data

[0259] Step 4:

[0260] The server extracts features such as rapid fluctuations in heart rate and peaks in electrodermal response from the preprocessed biometric data.

[0261] Input: Preprocessed biometric data

[0262] Data calculation: feature extraction (rapid fluctuations, peak detection)

[0263] Output: feature data

[0264] Step 5:

[0265] The server inputs the extracted feature data into an artificial intelligence model to infer the user's emotional state.

[0266] Input: feature data

[0267] Data Computation: Emotional Inference with AI Models

[0268] Output: Inference result (emotional state)

[0269] Step 6:

[0270] The server uses a feedback and recommendation algorithm to recommend the most suitable content to the user based on the inference results.

[0271] Input: Inference result (emotional state)

[0272] Data processing: Content selection by recommendation algorithm

[0273] Output: Recommended content

[0274] Step 7:

[0275] The smartphone notifies the user of the recommended content sent from the server using services such as Firebase Cloud Messaging.

[0276] Input: Recommended content

[0277] Data operation: Notification preparation (message generation)

[0278] Output: Content recommendation notification

[0279] Step 8:

[0280] While the user is watching the content, the smartphone again collects biometric data from the wearable device and monitors emotional changes in real time.

[0281] Input: Biometric data during viewing (heart rate, skin galvanic response)

[0282] Data calculation: Real-time monitoring

[0283] Output: User's emotional change data

[0284] Step 9:

[0285] The server uses the emotion change data to provide the next feedback and recommend content, and generates appropriate feedback using an algorithm.

[0286] Input: Emotion change data

[0287] Data calculation: feedback generation, next content recommendation

[0288] Output: Feedback to the user, next content recommendation

[0289] The above are the specific processing steps for implementing the present invention. This system allows users to enjoy content that is always tailored to their emotional state, improving their entertainment experience.

[0290] 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.

[0291] This invention relates to a system that collects a user's physical information and emotional state in real time and uses an artificial intelligence (AI) model and an emotion engine to make advanced decisions based on that information. In particular, accuracy can be improved in earthquake prediction and investment decision-making by combining physical sensations, intuition, and emotional state. Specific embodiments for implementing the present invention are described below.

[0292] System Configuration

[0293] The system consists of the following main components:

[0294] 1. Sensor device (terminal) - Acquires real-time physical information such as the user's heart rate, galvanic skin response, and hormone levels.

[0295] 2. Data transmission module (terminal) - transmits the acquired biometric data to the server.

[0296] 3. Data preprocessing module (server) - Performs preprocessing such as noise removal and normalization on the received raw data.

[0297] 4. Feature Extraction Module (Server) - Extracts important features from the preprocessed data.

[0298] 5. Emotion Recognition Module (Server) - Uses an emotion engine to infer the user's emotional state from biometric data.

[0299] 6. AI model (server) - receives extracted features and estimated emotional states as input and outputs inference results.

[0300] 7. Result correction module (server) - Corrects the inference results based on the emotional state.

[0301] 8. Result notification module (terminal) - Notifies the user of the inference results.

[0302] Processing flow

[0303] 1. Biometric data acquisition

[0304] The device captures second-by-second physical information such as the user's heart rate, galvanic skin response, and hormone levels using wearable devices and sensors attached to the skin.

[0305] 2. Data transmission

[0306] The device transmits the acquired raw data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[0307] 3. Data preprocessing and feature extraction

[0308] The server performs filtering operations on the received raw data to remove noise, for example, smoothing the heart rate data using a moving average filter, and normalization, scaling the data to a range from 0 to 1.

[0309] Extracting important features from normalized data, for example detecting sudden fluctuations in heart rate or peaks in electrodermal response.

[0310] 4. Emotional State Estimation

[0311] The server uses an emotion recognition module to estimate the user's emotional state from the preprocessed biometric data, for example, by using cosine similarity and emotion models to assess the user's stress state and happiness.

[0312] 5. Input to the AI ​​model

[0313] The server inputs the extracted feature data and estimated emotional state into an AI model, which then uses this information to make predictions for earthquake prediction and investment decision-making.

[0314] 6. Correction of results

[0315] The server adjusts the inference results of the AI ​​model based on the emotional state, enabling more accurate decision-making by adjusting the impact of the emotional state on the results.

[0316] 7. Output and notification of results

[0317] The server converts the final inference results into a format that is easy for the user to understand, and then sends the results to the device, which then notifies the user in various ways, such as by sound alert, vibration, or display on the screen.

[0318] Specific examples

[0319] Examples of earthquake prediction

[0320] 1. Acquisition of physical information

[0321] The device captures the user's heart rate and electrical skin response every second and sends it to a server.

[0322] 2. Data preprocessing and feature extraction

[0323] The server removes noise from the received data using a moving average filter and normalizes it.

[0324] Detects rapid fluctuations in heart rate and peaks in electrodermal response.

[0325] 3. Emotional State Estimation

[0326] The server uses an emotion engine to infer the user's emotional state (e.g., fear or tension) from the pre-processed data.

[0327] 4. Input to the AI ​​model

[0328] The server inputs the extracted feature data and emotional state into an earthquake prediction AI model to infer the probability of an earthquake occurring.

[0329] 5. Correction of results

[0330] The server will take into account the user's emotional state to correct the inference results, for example by weighting and adjusting the results if the emotional state has a strong influence on the predictions.

[0331] 6. Output and notification of results

[0332] The server converts the probability of an earthquake occurring into "high risk," "medium risk," or "low risk," and sends it to the terminal.

[0333] If the risk is high, the device will notify the user, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[0334] Investment Decision Examples

[0335] 1. Acquisition of physical information

[0336] The device measures the user's heart rate, galvanic skin response, and hormone levels every second and transmits them to a server.

[0337] 2. Data preprocessing and feature extraction

[0338] The server processes this data for noise reduction and normalization.

[0339] Detects heart rate variability, peaks in electrodermal response, and surges in hormone levels.

[0340] 3. Emotional State Estimation

[0341] The server uses an emotion engine to estimate the user's emotional state (e.g., anxiety or depression) from the preprocessed data.

[0342] 4. Input to the AI ​​model

[0343] The server inputs these characteristics and emotional states into an investment decision-making AI model to infer decisions on whether to buy, sell, or hold.

[0344] 5. Correction of results

[0345] The server corrects the inference results based on the user's emotional state in order to minimize the impact of the user's emotional state on investment decisions.

[0346] 6. Output and notification of results

[0347] The server receives the inference result, converts it into a format such as "You should buy XX," and sends it to the terminal.

[0348] The terminal notifies the user of the results and prompts them to take specific action.

[0349] The above is a specific embodiment for carrying out the present invention. This system enables advanced decision-making in real time by utilizing the user's physical information and emotional state, thereby increasing the probability of success in earthquake prediction and investment.

[0350] The processing flow will be explained below.

[0351] Step 1:

[0352] Acquiring physical information

[0353] The device collects real-time biometric data, such as the user's heart rate, galvanic skin response, and hormone levels, using wearable and skin-attached sensors.

[0354] Step 2:

[0355] Sending data

[0356] The device transmits the acquired biometric data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[0357] Step 3:

[0358] Receiving data

[0359] The server receives the biometric data sent from the device and stores it in chronological order.

[0360] Step 4:

[0361] Noise Reduction

[0362] The server performs filtering on the received raw data using a moving average filter or other method to remove noise, thereby smoothing the heart rate data.

[0363] Step 5:

[0364] Data normalization

[0365] The server performs a normalization process to scale the denoised data to the range of 0 to 1, thereby maintaining data consistency.

[0366] Step 6:

[0367] Feature Extraction

[0368] The server extracts important features from the normalized data, such as detecting sudden fluctuations in heart rate or peaks in electrodermal response.

[0369] Step 7:

[0370] Emotional state estimation

[0371] The server uses an emotion recognition module to estimate the user's emotional state from preprocessed biometric data, and utilizes an emotion engine to assess the user's stress state and happiness.

[0372] Step 8:

[0373] Input of feature data and emotional state

[0374] The server inputs the extracted feature data and estimated emotional state into an AI model, which uses this information to make predictions for earthquake prediction and investment decision-making.

[0375] Step 9:

[0376] AI model inference

[0377] The server inputs the characteristic data and emotional state into the AI ​​model, and infers the probability of an earthquake occurring and investment decisions such as "buy," "sell," or "hold."

[0378] Step 10:

[0379] Receiving inference results

[0380] The server receives the inference results from the AI ​​model, which are then further corrected rather than used as is.

[0381] Step 11:

[0382] Correcting the results

[0383] The server adjusts the inference results of the AI ​​model based on the emotional state, enabling more accurate decision-making by adjusting the impact of the emotional state on the results.

[0384] Step 12:

[0385] Sending the results

[0386] The server then sends the corrected inference results to the terminal, where they are converted into a format that is easy for the user to understand.

[0387] Step 13:

[0388] User Notification

[0389] The device notifies the user of the corrected inference results by means of voice alerts, vibrations, screen displays, etc., and prompts the user to take specific action.

[0390] The above is a specific flow of the program processing of the present invention, which enables advanced decision-making in real time by utilizing the user's physical information and emotional state.

[0391] Example 2

[0392] 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."

[0393] Conventional earthquake prediction and investment decision-making systems have the problem of low prediction accuracy because they do not take into account the user's emotional state or physical information. Furthermore, it is difficult to collect data and make decisions in real time, making it difficult to prompt users to take immediate action. Furthermore, data preprocessing and feature extraction are insufficient, making it difficult to obtain reliable results.

[0394] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0395] In this invention, the server includes means for acquiring a user's physical information in real time, means for transmitting the acquired physical information to the server, means for removing noise from the physical information received by the server and normalizing the data, means for extracting features from the preprocessed data, means for estimating an emotional state using an emotion recognition module, means for inputting the feature data and the emotional state into an artificial intelligence model to obtain an inference result, means for correcting the inference result based on the emotional state, and means for notifying the user of the corrected inference result. This enables highly accurate predictions that reflect the user's physical information and emotional state in real time, and allows the user to be quickly notified of specific actions.

[0396] "Physical information" refers to biometric data such as a user's heart rate, galvanic skin response, and hormone levels.

[0397] "Real-time" refers to data being collected and processed almost immediately.

[0398] "Server" refers to the computer system that receives, pre-processes, extracts features, estimates emotional states, and performs inference on biometric data.

[0399] "Terminal" refers to a device for collecting and transmitting a user's physical information to a server.

[0400] "Noise reduction" refers to the process of removing unwanted signals and disturbances from raw data.

[0401] "Normalization" refers to the process of scaling data to a particular range.

[0402] "Feature extraction" refers to the process of finding important indicators and patterns from preprocessed data.

[0403] "Emotion Recognition Module" refers to software or algorithms for inferring a user's emotional state based on biometric data.

[0404] An "artificial intelligence model" refers to a model that uses machine learning algorithms to make inferences.

[0405] "Inference results" refer to the predictions and judgments that an artificial intelligence model outputs based on input data.

[0406] "Correction" refers to the process of adjusting the initial inference result by taking into account factors such as emotional state.

[0407] "Notification" refers to the system informing the user of inference results and actions.

[0408] This invention is a system that collects a user's physical information and emotional state in real time, and uses an artificial intelligence (AI) model and emotion engine to make advanced decisions based on that information. In particular, the accuracy of earthquake prediction and investment decision-making can be improved by combining physical sensations, intuition, and emotional state.

[0409] Hardware and software used

[0410] The devices used are wearable or skin-attached sensor devices that can capture biometric data such as heart rate, galvanic skin response, and hormone levels within seconds.

[0411] Specifically, smartwatches and biosensors are used as terminals, and these devices transmit data to a server via wireless communication technologies such as Bluetooth and Wi-Fi.

[0412] On the server, Python's pandas library and sklearn's StandardScaler are used to remove noise and normalize the biometric data. OpenAI's sentiment analysis API is used to estimate emotional states. Furthermore, an AI model built with TensorFlow is used to infer the probability of earthquake occurrence and investment decision-making.

[0413] Data processing and calculation

[0414] 1. Biometric data acquisition

[0415] The device captures the user's heart rate, galvanic skin response, and hormone levels every second. Specifically, the smartwatch measures heart rate and galvanic skin response, while sensors under the skin periodically measure hormone levels.

[0416] 2. Data transmission

[0417] The device transmits the acquired biometric data to a smartphone via Bluetooth, and the smartphone then transmits the data to a server using Wi-Fi.

[0418] 3. Data preprocessing and feature extraction

[0419] The server performs noise removal and data normalization on the received raw data. For example, it uses Python's pandas library to smooth the data with a moving average filter and sklearn's StandardScaler to normalize the data to the range of 0 to 1. It then extracts important features, such as rapid fluctuations in heart rate and peaks in electrodermal responses.

[0420] 4. Emotional State Estimation

[0421] The server estimates the user's emotional state from the preprocessed data using an emotion recognition module. For example, the server uses OpenAI's emotion analysis API to evaluate the user's stress level and happiness from the normalized data.

[0422] 5. Input to the AI ​​model

[0423] The server inputs the extracted feature data and estimated emotional state into an AI model to perform inference. For example, the server inputs the data into an earthquake prediction model built using TensorFlow to predict the probability of an earthquake occurring.

[0424] 6. Correction of results

[0425] The server corrects the inference results of the AI ​​model based on the emotional state, adjusting the influence of emotional states such as high stress if they have a strong impact on the results.

[0426] 7. Output and notification of results

[0427] The server converts the final inference results into a user-friendly format and sends them to the device, which notifies the user of the results by means of an alert sound, vibration, or on-screen display.

[0428] Examples and prompts

[0429] Specific examples

[0430] Examples of earthquake prediction:

[0431] The device captures the user's heart rate and galvanic skin response every second and sends them to a server. The server smooths and normalizes the data, detecting sudden fluctuations in heart rate and peaks in galvanic skin response. An emotion engine estimates fear and tension, and inputs the data into an earthquake prediction AI model. The server then classifies the risk of an earthquake occurrence as "high risk" or similar, and notifies the device of the results.

[0432] Examples of investment decisions:

[0433] The device acquires the user's heart rate, galvanic response, and hormone levels, and sends them to a server. The server preprocesses this data and detects heart rate fluctuations, peaks in galvanic response, and sudden spikes in hormone levels. An emotion engine estimates feelings of anxiety and depression, and inputs the data into an investment decision-making AI model. The server then makes a decision to "buy," "sell," or "hold," and notifies the user of the result.

[0434] Prompt Sentence Examples

[0435] Earthquake prediction prompt:

[0436] "Get the user's heart rate and electrodermal response data every second to estimate their emotional state, such as fear or tension. Then, based on this, assess the risk of an earthquake occurring and notify them if the risk is high."

[0437] Investment Decision Prompt:

[0438] "Collect data on the user's heart rate, galvanic skin response, and hormone levels to estimate their emotional state, such as anxiety or depression. Based on that, make investment decisions to buy, sell, or hold, and notify the user of the results."

[0439] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0440] Step 1:

[0441] Biometric data acquisition

[0442] The device collects real-time physical information every second, such as the user's heart rate, galvanic skin response, and hormone levels. Specifically, this data is collected by a smartwatch or sub-skin sensors.

[0443] Input: User's physical information

[0444] Output: Acquired biometric data (heart rate, galvanic skin response, hormone levels)

[0445] Step 2:

[0446] Sending data

[0447] The device transmits the acquired biometric data to a server at regular intervals. Specifically, the data is transferred to a smartphone via Bluetooth, and the smartphone then transmits the data to the server using Wi-Fi.

[0448] Input: Biometric data (heart rate, galvanic skin response, hormone levels)

[0449] Output: Biometric data sent to the server

[0450] Step 3:

[0451] Data Preprocessing

[0452] The server performs noise removal and normalization on the received biometric data. Specifically, it smooths the data with a moving average filter using Python's pandas library and normalizes the data to the range of 0 to 1 using sklearn's StandardScaler.

[0453] Input: Received biometric data

[0454] Output: Denoised normalized data

[0455] Step 4:

[0456] Feature Extraction

[0457] The server extracts important features from the denoised and normalised data, specifically detecting rapid fluctuations in heart rate and peaks in electrodermal response.

[0458] Input: Denoised normalized data

[0459] Output: Extracted feature data (heart rate spikes, electrodermal response peaks)

[0460] Step 5:

[0461] Emotional state estimation

[0462] The server uses an emotion recognition module to estimate the user's emotional state from the preprocessed data. For example, it uses OpenAI's emotion analysis API to evaluate the user's stress level or happiness from the data.

[0463] Input: Extracted feature data

[0464] Output: Estimated emotional state (stress, happiness)

[0465] Step 6:

[0466] Input to the AI ​​model

[0467] The server inputs the extracted feature data and estimated emotional state into a generative AI model to perform inference for earthquake prediction and investment decision-making. Specifically, the server inputs the data into an AI model using TensorFlow.

[0468] Input: extracted feature data, estimated emotional state

[0469] Output: Initial inference results (probability of earthquake occurrence, investment decision)

[0470] Step 7:

[0471] Correcting the results

[0472] The server corrects the initial inference results of the AI ​​model based on the emotional state, adjusting its influence if the emotional state has a strong influence on the results.

[0473] Input: Initial inference result, estimated emotional state

[0474] Output: Corrected inference results

[0475] Step 8:

[0476] Notification of results

[0477] The server converts the corrected inference results into a user-friendly format and sends them to the device, which notifies the user of the results by sound alerts, vibrations, or a display on the screen.

[0478] Input: Corrected inference results

[0479] Output: The result reported to the user

[0480] (Application example 2)

[0481] 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."

[0482] Conventional smart home security systems primarily detect anomalies based on environmental data, without taking into account the user's physical condition or emotional state. As a result, they sometimes fail to detect actual anomalies, or sometimes generate frequent false alarms. This invention aims to improve the accuracy of anomaly detection by combining the user's biometric information and emotional state, providing a more reliable security experience.

[0483] 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.

[0484] In this invention, the server includes means for acquiring biometric data, means for preprocessing the acquired biometric data, means for extracting features from the preprocessed biometric data, means for inputting the extracted features into an artificial intelligence model to obtain an inference result, means for outputting the obtained inference result and notifying the user, means for estimating the emotional state of the user and detecting an abnormality, and means for notifying the user of a specific countermeasure based on the abnormality detection result. This enables an abnormality detection that reflects the user's physical condition and emotional state.

[0485] "Biometric data" is data indicative of a user's physiological state, such as a user's heart rate, galvanic skin response, or hormone levels.

[0486] "Preprocessing" is a process that removes noise from biometric data and prepares the data in a form that is easier to analyze.

[0487] "Feature extraction" is the process of extracting important patterns and trends from pre-processed data.

[0488] An "artificial intelligence model" is an algorithm or system that learns from large amounts of data and makes appropriate inferences about unknown data.

[0489] "Inference results" are predictions or judgments obtained from data processed by an artificial intelligence model.

[0490] "Notification means" refers to a means for notifying the user of inference results or anomaly detection results, particularly via a smartphone or other device.

[0491] "Emotional state" refers to the psychological state such as anxiety, tension, or stress felt by the user.

[0492] "Anomaly detection" is the process of recognizing safety risks and anomalies based on a person's biometric data and emotional state.

[0493] "Countermeasures" are specific actions or procedures that a user should take in response to a detected abnormality.

[0494] This invention is a security service-related system that collects a user's physical information and emotional state in real time and makes advanced decisions based on that data using an artificial intelligence model and emotion engine. In particular, it includes technology to improve the accuracy of anomaly detection in smart home security systems.

[0495] System Configuration

[0496] The system consists of the following main components:

[0497] 1. Wearable devices: collect real-time biometric data such as the user's heart rate, galvanic skin response, and hormone levels.

[0498] 2. Smartphone: A device for transmitting biometric data acquired from wearable devices to a server.

[0499] 3. Server: Includes the following software modules:

[0500] Data preprocessing module: Removes noise and normalizes raw data (software used: Azure (registered trademark) Data Factory).

[0501] Feature Extraction Module: Extracts important features from the preprocessed data.

[0502] Emotion recognition module: Estimates emotional state from biometric data (software used: Google® Cloud AI).

[0503] AI model: Detects anomalies based on extracted features and emotional states (software used: TensorFlow).

[0504] Result notification module: Notifies the user of the results of anomaly detection via their smartphone (software used: Firebase Cloud Messaging).

[0505] Processing flow

[0506] The server first receives biometric data acquired from the wearable device, denoises and normalizes it using Azure Data Factory, and then extracts features from the preprocessed data, such as sudden fluctuations in heart rate and peaks in electrodermal responses.

[0507] The emotion recognition module uses Google Cloud AI to infer the user's emotional state from pre-processed data and identify states such as anxiety or tension. These features and emotional state are then fed into a TensorFlow-powered AI model for anomaly detection.

[0508] As a result, if an abnormality is detected, a notification will be sent to the user's smartphone via Firebase Cloud Messaging with instructions on how to deal with the problem. The user can then take appropriate action based on the notification they receive.

[0509] Specific examples

[0510] For example, in a "smart home security system equipped with emotion sensors," if a user experiences a sudden rise in heart rate and a peak in their electrodermal response, the emotion recognition module will determine that the user is in a state of high anxiety, and this state will be registered as an abnormality in the system. The AI ​​model will also analyze environmental data (for example, whether the door is open or closed, and the sound level in the room), and if it finally detects an abnormality, it will send a notification to the user's smartphone saying, "The door is not closed. Please check."

[0511] Prompt Sentence Examples

[0512] An example of a prompt is "Design a smart home security system that monitors heart rate and galvanic skin response in real time, and detects and notifies users if anxiety levels are high."

[0513] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0514] Step 1:

[0515] Biometric data acquisition

[0516] The terminal (wearable device) collects real-time data on the user's heart rate, skin galvanic response, and hormone levels. This data is collected from sensors and received by a smartphone. The input is the user's biometric data, and the output is the raw data received by the smartphone. Specifically, data is recorded every second and sent to the smartphone via Bluetooth or other means.

[0517] Step 2:

[0518] Sending data

[0519] The device (smartphone) sends the received biometric data to a server using Wi-Fi or Bluetooth. The server receives this data and supplies it to the next processing step. The input is the acquired raw data, and the output is the data sent to the server. Specifically, the smartphone application packets the data at intervals and sends it.

[0520] Step 3:

[0521] Data Preprocessing

[0522] The raw data sent to the server is denoised and normalised using Azure Data Factory. This purifies the data and makes it easier to analyse. The input is the raw data sent to the server, and the output is the denoised and normalised data. Specific operations include smoothing the heart rate data using a moving average filter and scaling the data from 0 to 1.

[0523] Step 4:

[0524] Feature Extraction

[0525] Important features such as sudden changes in heart rate and peaks in electrodermal responses are extracted from data that has been preprocessed on the server. The input is the preprocessed data, and the output is the extracted features. Specifically, an algorithm is used to analyze the temporal fluctuations of the data and identify sudden changes and peak values.

[0526] Step 5:

[0527] Emotional state estimation

[0528] The extracted feature data is input into Google Cloud AI's emotion recognition module, which estimates the user's emotional state. The input is the extracted feature data, and the output is the estimated emotional state. Specifically, stress and anxiety states are visualized using techniques such as cosine similarity.

[0529] Step 6:

[0530] Input to AI model and anomaly detection

[0531] The estimated emotional state and feature data are input into an AI model (TensorFlow) to perform anomaly detection. The input is the feature data and emotional state, and the output is the anomaly detection result. Specifically, the trained AI model analyzes this data and determines whether there is an anomaly.

[0532] Step 7:

[0533] Result notification

[0534] If an abnormality is detected, the server uses Firebase Cloud Messaging to send a notification to the user's smartphone. The input is the abnormality detection result, and the output is the notification that is displayed on the user's smartphone. Specifically, a message such as "The door is not closed. Please check" is sent.

[0535] Step 8:

[0536] User Action

[0537] The user receives the notification and takes appropriate action based on the specific countermeasures shown. The input is the notification displayed on the smartphone, and the output is the user's action. Specifically, the user checks the door and takes action such as closing it if necessary.

[0538] This enables highly accurate anomaly detection and notification based on the user's biometric information and emotional state.

[0539] 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.

[0540] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (registered trademark) (Internet search engine).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0541] 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.

[0542] [Second embodiment]

[0543] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0544] 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.

[0545] 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).

[0546] 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.

[0547] 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.

[0548] 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).

[0549] 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.

[0550] 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.

[0551] 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.

[0552] 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.

[0553] 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.

[0554] 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."

[0555] This invention relates to a system that collects a user's physical information in real time and uses an artificial intelligence (AI) model based on that information to make advanced decisions. In particular, accuracy can be improved in earthquake prediction and investment decision-making by utilizing physical sensations and intuition. Specific embodiments for carrying out the present invention are described below.

[0556] System Configuration

[0557] The system consists of the following main components:

[0558] 1. Sensor device (terminal) - Acquires real-time physical information such as the user's heart rate, galvanic skin response, and hormone levels.

[0559] 2. Data transmission module (terminal) - transmits the acquired biometric data to the server.

[0560] 3. Data preprocessing module (server) - Performs preprocessing such as noise removal and normalization on the received raw data.

[0561] 4. Feature Extraction Module (Server) - Extracts important features from the preprocessed data.

[0562] 5. AI model (server) - receives extracted features as input and outputs inference results.

[0563] 6. Result notification module (terminal) - Notifies the user of the inference results.

[0564] Processing flow

[0565] 1. Biometric data acquisition

[0566] The device captures data such as the user's heart rate, galvanic skin response, and hormone levels every second, using wearable devices and sensors attached to the skin.

[0567] 2. Data transmission

[0568] The device transmits the acquired raw data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[0569] 3. Data Preprocessing

[0570] The server performs filtering operations on the received raw data to remove noise, for example, smoothing the heart rate data using a moving average filter, and normalization, scaling the data to a range from 0 to 1.

[0571] 4. Feature Extraction

[0572] The server extracts features from the preprocessed data, such as sudden fluctuations in heart rate and peaks in skin galvanic responses, thereby capturing the user's physical reactions to precursors to earthquakes or sudden market fluctuations.

[0573] 5. Linking to AI models

[0574] The server inputs the extracted feature data into the AI ​​model, which compares it with the data it has learned so far to determine the probability of an earthquake occurring and whether to "buy," "sell," or "hold" in investment decisions.

[0575] 6. Output and notification of results

[0576] The server receives the inference results of the AI ​​model, converts them into a format that is easy for the user to understand, and sends the results to the device, which then notifies the user in various ways, such as by sound alert, vibration, or display on the screen.

[0577] Specific examples

[0578] Examples of earthquake prediction

[0579] 1. Acquisition of physical information

[0580] The device captures the user's heart rate and electrical skin response every second and sends it to a server.

[0581] 2. Data Preprocessing

[0582] The server removes noise from the received data using a moving average filter and normalizes it.

[0583] 3. Feature Extraction

[0584] The server detects rapid fluctuations in heart rate and peaks in electrodermal response.

[0585] 4. Linking to AI models

[0586] The server inputs these features into an earthquake prediction AI model to infer the probability of an earthquake occurring.

[0587] 5. Outputting the results

[0588] The server converts the probability of an earthquake occurring into "high risk," "medium risk," or "low risk," and sends it to the terminal.

[0589] 6. User Notices

[0590] If the risk is high, the device will notify the user, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[0591] Investment Decision Examples

[0592] 1. Acquisition of physical information

[0593] The device measures the user's heart rate, galvanic skin response, and hormone levels every second and transmits them to a server.

[0594] 2. Data Preprocessing

[0595] The server processes this data for noise reduction and normalization.

[0596] 3. Feature Extraction

[0597] The server detects heart rate variability, peaks in galvanic skin response, and spikes in hormone levels.

[0598] 4. Linking to AI models

[0599] The server inputs these characteristics into an investment decision-making AI model and infers the decision to "buy," "sell," or "hold."

[0600] 5. Outputting the results

[0601] The server receives the inference results, converts them into a format such as "You should buy XX," and sends them to the terminal.

[0602] 6. User Notices

[0603] The terminal notifies the user of the results and suggests specific actions to take.

[0604] The above is a specific embodiment for carrying out the present invention. This system enables advanced decision-making using the user's physical information, thereby increasing the probability of success in earthquake prediction and investment.

[0605] The processing flow will be explained below.

[0606] Step 1:

[0607] Acquiring physical information

[0608] The device collects real-time biometric data such as the user's heart rate, galvanic skin response, and hormone levels using wearable and skin-attached sensors.

[0609] Step 2:

[0610] Sending data

[0611] The device transmits the acquired biometric data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[0612] Step 3:

[0613] Receiving data

[0614] The server receives the biometric data sent from the device and stores it in chronological order.

[0615] Step 4:

[0616] Noise Reduction

[0617] The server performs filtering on the received raw data to remove noise, for example, applying a moving average filter to smooth the heart rate data.

[0618] Step 5:

[0619] Data normalization

[0620] The server normalizes the denoised data, scaling the heart rate data to a range of 0 to 1 to make it consistent with other biometric data.

[0621] Step 6:

[0622] Feature Extraction

[0623] The server extracts important features from the normalized data, applying algorithms to detect, for example, sudden fluctuations in heart rate or peaks in electrodermal response.

[0624] Step 7:

[0625] Entering feature data

[0626] The server inputs the extracted feature data into an AI model that has been trained for earthquake prediction and investment decision-making.

[0627] Step 8:

[0628] AI model inference

[0629] The server inputs the characteristic data into the AI ​​model and performs inference to output the probability of an earthquake occurring and investment decisions such as "buy," "sell," or "hold."

[0630] Step 9:

[0631] Receiving inference results

[0632] The server receives the inference results from the AI ​​model and converts them into a format that is easy for users to understand.

[0633] Step 10:

[0634] Sending the results

[0635] The server then sends the converted inference results to the terminal, which then provides the results to the user as an alert message.

[0636] Step 11:

[0637] User Notification

[0638] The device notifies the user of the inference results by means of voice alerts, vibrations, screen displays, etc., and prompts the user to take specific action.

[0639] The above is a specific flow of the program processing of the present invention, which enables advanced decision-making in real time using the user's physical information.

[0640] Example 1

[0641] 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."

[0642] Existing systems have a problem with low accuracy in earthquake prediction and investment decision-making using users' biometric data. This is because the processes of biometric data acquisition, preprocessing, feature extraction, inference, and notification are performed separately, reducing the overall system efficiency. Another problem is that insufficient noise removal and normalization of biometric data can easily produce inaccurate inference results. To solve these problems, an advanced integrated system is required.

[0643] 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.

[0644] In this invention, the server includes means for acquiring biometric data, means for preprocessing the acquired biometric data, means for extracting features from the preprocessed biometric data, means for inputting the extracted features into an artificial intelligence model to obtain an inference result, means for outputting the obtained inference result and notifying the user, means for using a wearable device or a sensor attached to the skin, means for transmitting data using wireless communication technology such as Bluetooth or Wi-Fi, means for smoothing data using a moving average filter, means for normalizing data to a range from 0 to 1, and means for converting the inference result into "high risk," "medium risk," or "low risk." This enables advanced decision-making using a user's biometric data to be performed accurately in real time.

[0645] "Biometric data" is information obtained from a user's body, including heart rate, galvanic skin response, hormone levels, and the like.

[0646] "Preprocessing" refers to applying filtering, normalization, and other processes to acquired biometric data to remove noise and convert the data into a format that is easier to analyze.

[0647] "Feature extraction" is the process of finding important patterns and variations in pre-processed data and preparing it for input into an artificial intelligence model.

[0648] An "artificial intelligence model" is a software system that uses machine learning algorithms to analyze data and draw inferences.

[0649] "Inference results" are the conclusions or judgments that an AI model provides based on biometric data, such as the probability of an earthquake occurring or decisions about whether to "buy," "sell," or "hold" an investment.

[0650] "Notification" is a means of conveying the inference results to the user, and is implemented in the form of an alert sound, vibration, screen display, etc.

[0651] A "wearable device" is a device worn by a user that can acquire biometric data, such as a smartwatch or activity tracker.

[0652] A "sensor" is a device for measuring biometric data, and includes types that are attached to the skin and types that are worn on the arm.

[0653] "Wireless communication technology" refers to technology for sending and receiving data wirelessly, and includes Bluetooth and Wi-Fi.

[0654] A "moving average filter" is a filtering method used to smooth out fluctuations in data, and is a method of removing noise by taking the average over time.

[0655] "Normalization" is a process performed to unify the scale of data, and involves converting the scale of biometric data to a range of 0 to 1.

[0656] "Risk conversion" is the process of converting inference results into categories such as "high risk," "medium risk," and "low risk," and presenting them in a form that is easy for users to understand.

[0657] The system of the present invention is configured to collect biometric data of a user in real time and make advanced decisions using an artificial intelligence (AI) model. Detailed embodiments of the system are described below.

[0658] System Configuration

[0659] This system mainly consists of the following hardware and software components.

[0660] 1. Sensor device (terminal)

[0661] Sensor devices are used to obtain real-time biometric data such as a user's heart rate, electrodermal response, and hormone levels. Specific examples include wearable devices (smartwatches, activity trackers) and sensors attached to the skin. For example, Apple Watch and general skin electrode sensors can be used.

[0662] 2. Data transmission module (terminal)

[0663] The data transmission module is responsible for transmitting the biometric data acquired from the sensor device to the server, using wireless communication technologies such as Bluetooth and Wi-Fi.

[0664] 3. Data preprocessing module (server)

[0665] The data preprocessing module performs noise removal and normalization on the biometric data received on the server, specifically smoothing the data using a moving average filter and scaling the biometric data to a range of 0 to 1.

[0666] 4. Feature extraction module (server)

[0667] The feature extraction module extracts important features from the pre-processed data, such as rapid heart rate fluctuations or peaks in electrodermal response.

[0668] 5. AI model (server)

[0669] This module inputs the extracted feature data into the AI ​​model to obtain inference results. It uses a deep learning algorithm to compare the results with past data and make inferences.

[0670] 6. Result notification module (terminal)

[0671] This module notifies the user of the inference results. Notifications are given in the form of an alert sound, vibration, screen display, etc.

[0672] Specific examples

[0673] Examples of earthquake prediction

[0674] When a user uses a sensor device (e.g., a smartwatch), their heart rate and electrodermal response are acquired every second. This data is sent to a server via Bluetooth. The server smooths the heart rate data using a moving average filter and normalizes the data. The server then detects sudden heart rate fluctuations and peaks in electrodermal response from the preprocessed data. This feature data is input into an AI earthquake prediction model to infer the probability of an earthquake occurring. Finally, the server converts the inference result into "high risk," "medium risk," or "low risk" and sends it to the device. If the risk is high, the device notifies the user, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[0675] Investment Decision Examples

[0676] The user uses the same sensor device to measure their heart rate, galvanic response, and hormone levels every second. This data is sent to a server, which then denoises and normalizes it. Heart rate fluctuations, peaks in galvanic response, and spikes in hormone levels are then detected. This characteristic data is then input into an AI investment decision-making model, which then infers whether to "buy," "sell," or "hold." The server then converts this inference result into a concrete format, such as "You should buy XX," and sends it to the device. The device then notifies the user, "You should buy XX," and suggests a specific action to take.

[0677] Prompt Sentence Examples

[0678] Example prompts to be input to the generative AI model:

[0679] The user's heart rate fluctuates rapidly and their electrodermal response shows a peak. These can be fed into an AI model to assess the risk of an earthquake occurring.

[0680] This system enables advanced decision-making in real time using the user's biometric data, increasing the chances of successful earthquake prediction and investment.

[0681] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0682] Step 1:

[0683] Biometric data acquisition

[0684] Subject: Device

[0685] Description: The device uses sensor devices to collect real-time biometric data such as the user's heart rate, galvanic skin response, and hormone levels. Input data comes from various sensors worn by the user. The device converts this signal into digital data and updates it every second.

[0686] How it works: The device receives signals from the smartwatch and skin electrodes. For example, the optical heart rate monitor in the Apple Watch measures your heart rate beat-by-beat, and the galvanic skin response sensor captures your stress level. This data is temporarily stored on the device.

[0687] Step 2:

[0688] Sending data

[0689] Subject: Device

[0690] Description: The device transmits the acquired biometric data to the server at regular intervals (e.g., every second). The input data is the biometric data acquired from the sensor device, and the output data is the data packet sent to the server.

[0691] How it works: The device sends data to the server via Bluetooth or Wi-Fi, for example using the low-latency, highly reliable Bluetooth Low Energy (BLE) protocol to ensure that the biometric data reaches the server reliably.

[0692] Step 3:

[0693] Data Preprocessing

[0694] Subject: Server

[0695] Description: The server performs preprocessing on the received biometric data, specifically noise removal and normalization. The input data is the biometric data sent from the device, and the output data is the preprocessed data.

[0696] What it does: The server applies a moving average filter to remove spikes and noise from the data. For example, a 3-second moving average filter is used to smooth out fluctuations in heart rate data. At the same time, it normalizes the data by scaling it to a range of 0 to 1, which makes it easier to extract features and input into AI models.

[0697] Step 4:

[0698] Feature Extraction

[0699] Subject: Server

[0700] Description: The server extracts important features from preprocessed data. The input data is the preprocessed data, and the output data is the extracted feature data.

[0701] How it works: The server detects features such as sudden fluctuations in heart rate data and peaks in galvanic response (GDS). It then uses an algorithm to extract, for example, spikes in heart rate or peaks in GDS, thereby identifying moments of stress or excitement felt by the user.

[0702] Step 5:

[0703] Linking to AI models

[0704] Subject: Server

[0705] Description: The server inputs the extracted feature data into the AI ​​model and obtains the inference result. The input data is the feature data, and the output data is the inference result from the AI ​​model.

[0706] How it works: The server uses a trained deep learning model to input feature data into an AI model. The AI ​​model compares this data with past data to determine the probability of an earthquake occurring and whether to "buy," "sell," or "hold" an investment. The model outputs its inference results in real time.

[0707] Step 6:

[0708] Output and notification of results

[0709] Subject: Server

[0710] Description: The server converts the inference results of the AI ​​model into an easy-to-understand format and sends it to the terminal. The input data is the inference results from the AI ​​model, and the output data is the final result that is notified to the user.

[0711] Specific operation: The server classifies the inference results into "high risk," "medium risk," or "low risk" and sends them to the device in JSON format. The device analyzes the received data and notifies the user with a specific message. For example, it warns the user with an alert sound or vibration saying, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[0712] The above is a detailed description of the specific operations and data flow at each processing step of this system. This detailed flow enables advanced decision-making using the user's biometric data in real time.

[0713] (Application example 1)

[0714] 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."

[0715] Conventional content delivery systems primarily recommend content based on users' preferences and interests, but this approach fails to respond to users' real-time emotional states. This results in a poor user experience and reduces their willingness to consume content. Furthermore, the lack of a mechanism for capturing changes in emotional states in real time makes it difficult to provide users with appropriate feedback.

[0716] 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.

[0717] In this invention, the server includes means for acquiring biometric data, means for preprocessing the acquired biometric data, means for extracting features from the preprocessed biometric data, means for inputting the extracted features into an artificial intelligence model to obtain an inference result, means for recommending optimal content to a user based on the obtained inference result, means for notifying the user of the recommended content and monitoring emotional changes in real time as the user views the content, and means for providing feedback based on the monitored emotional changes. This enables personalized content recommendations based on the user's real-time emotional state and the provision of appropriate feedback.

[0718] Below are definitions of important terms included in the claims, tailored to application examples.

[0719] "Biometric data" refers to physical information such as a user's heart rate, galvanic skin response, and hormone levels.

[0720] "Preprocessing" refers to performing processes such as noise removal and normalization on acquired biometric data.

[0721] "Feature extraction" refers to extracting important features, such as rapid heart rate fluctuations or peaks in electrodermal responses, from preprocessed biometric data.

[0722] An "artificial intelligence model" refers to a machine learning algorithm or neural network that makes inferences or decisions based on given input data.

[0723] "Inference result" refers to the prediction or judgment result that an artificial intelligence model outputs based on input data.

[0724] "Content recommendation" refers to the process of suggesting optimal videos, music, and other media to users based on their biometric data.

[0725] "Changes in emotion" refers to fluctuations in emotional state, such as excitement, joy, or sadness, that a user experiences when viewing content.

[0726] "Providing feedback" refers to suggesting appropriate advice or next actions to the user based on monitored emotional changes.

[0727] This invention is a system that uses a user's biometric data to recommend optimal content in real time and provide appropriate feedback. This system consists of the following main components:

[0728] System Configuration

[0729] 1. Means of acquiring biometric data

[0730] Means: A wearable device for measuring the user's heart rate or galvanic skin response (e.g., a heart rate sensor or galvanic skin response sensor).

[0731] Hardware used: General wearable devices (e.g., smartwatches, fitness bands, etc.)

[0732] 2. Pretreatment Methods

[0733] Method: Denoise and normalize the acquired biometric data

[0734] Software used: Python numerical calculation libraries (e.g., numpy, pandas), data preprocessing libraries (e.g., scikit-learn)

[0735] 3. Feature Extraction Method

[0736] Method: Extract features such as heart rate variability and electrodermal response peaks from the preprocessed data.

[0737] Software used: Python machine learning library (e.g., scikit-learn)

[0738] 4. Artificial Intelligence Models

[0739] Method: Take the extracted features as input and infer the user's real-time emotional state.

[0740] Software used: Deep learning libraries (e.g., TensorFlow, Keras)

[0741] Model: Emotional Inference Model

[0742] 5. Content recommendation method based on inference results

[0743] Method: Recommending optimal content to users based on inference results from an AI model

[0744] Software used: Recommendation systems (e.g., content recommendation algorithms)

[0745] 6. User Notification Methods

[0746] Method: Notify the smartphone of the recommended content so that the user can watch it.

[0747] Hardware used: Smartphone

[0748] Software used: Notification library (e.g., Firebase Cloud Messaging)

[0749] 7. Emotion Monitoring Tools

[0750] Method: While the user is watching the content, heart rate and electrodermal response are collected again to monitor changes in emotions in real time.

[0751] Hardware used: General wearable devices

[0752] Software used: Dedicated SDK (e.g., Apple HealthKit, Fitbit SDK)

[0753] 8. Means of Providing Feedback

[0754] Method: Suggest the next content to watch or specific actions based on changes in emotions

[0755] Software used: Recommendation systems (e.g., feedback generation algorithms)

[0756] Specific examples

[0757] For example, when a user wears a smartwatch and watches a video, the smartwatch measures the user's heart rate and electrodermal response and sends the data to a smartphone. The smartphone then transmits this biometric data to a server at regular intervals. The server then preprocesses the received data and extracts features. An AI model uses this feature data to infer the user's emotional state and recommends the next optimal piece of content based on the inference. The smartphone then notifies the user of the content and provides appropriate feedback by monitoring emotional changes in real time while watching.

[0758] Prompt Sentence Examples

[0759] "Please infer the user's real-time emotional state from their heart rate and electrodermal response based on the content they are currently viewing, and recommend the best content for them to watch next."

[0760] This allows users to enjoy content that is always tailored to their emotional state, resulting in a system that provides an even more entertaining experience.

[0761] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0762] Step 1:

[0763] The user puts on the wearable device.

[0764] The device measures the user's heart rate and electrical skin response in real time, and transmits this data to a smartphone via Bluetooth or other means.

[0765] Input: User's heart rate, galvanic skin response

[0766] Data processing: Acquired from wearable devices

[0767] Output: Biometric data is sent to a smartphone

[0768] Step 2:

[0769] The smartphone receives biometric data from the wearable device and sends it to a server at regular intervals, using Wi-Fi or mobile data for communication.

[0770] Input: Biometric data (heart rate, electrodermal response)

[0771] Data processing: Buffering data at regular intervals

[0772] Output: Biometric data is sent to the server

[0773] Step 3:

[0774] The server performs noise reduction and normalization on the received biometric data, for example, smoothing the heart rate data using a moving average filter.

[0775] Input: Biometric data (heart rate, electrodermal response)

[0776] Data calculation: Noise removal (moving average filter), normalization

[0777] Output: Preprocessed biometric data

[0778] Step 4:

[0779] The server extracts features such as rapid fluctuations in heart rate and peaks in electrodermal response from the preprocessed biometric data.

[0780] Input: Preprocessed biometric data

[0781] Data calculation: feature extraction (rapid fluctuations, peak detection)

[0782] Output: feature data

[0783] Step 5:

[0784] The server inputs the extracted feature data into an artificial intelligence model to infer the user's emotional state.

[0785] Input: feature data

[0786] Data Computation: Emotional Inference with AI Models

[0787] Output: Inference result (emotional state)

[0788] Step 6:

[0789] The server uses a feedback and recommendation algorithm to recommend the most suitable content to the user based on the inference results.

[0790] Input: Inference result (emotional state)

[0791] Data processing: Content selection by recommendation algorithm

[0792] Output: Recommended content

[0793] Step 7:

[0794] The smartphone notifies the user of the recommended content sent from the server using services such as Firebase Cloud Messaging.

[0795] Input: Recommended content

[0796] Data operation: Notification preparation (message generation)

[0797] Output: Content recommendation notification

[0798] Step 8:

[0799] While the user is watching the content, the smartphone again collects biometric data from the wearable device and monitors emotional changes in real time.

[0800] Input: Biometric data during viewing (heart rate, skin galvanic response)

[0801] Data calculation: Real-time monitoring

[0802] Output: User's emotional change data

[0803] Step 9:

[0804] The server uses the emotion change data to provide the next feedback and recommend content, and generates appropriate feedback using an algorithm.

[0805] Input: Emotion change data

[0806] Data calculation: feedback generation, next content recommendation

[0807] Output: Feedback to the user, next content recommendation

[0808] The above are the specific processing steps for implementing the present invention. This system allows users to enjoy content that is always tailored to their emotional state, improving their entertainment experience.

[0809] 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.

[0810] This invention relates to a system that collects a user's physical information and emotional state in real time and uses an artificial intelligence (AI) model and an emotion engine to make advanced decisions based on that information. In particular, accuracy can be improved in earthquake prediction and investment decision-making by combining physical sensations, intuition, and emotional state. Specific embodiments for implementing the present invention are described below.

[0811] System Configuration

[0812] The system consists of the following main components:

[0813] 1. Sensor device (terminal) - Acquires real-time physical information such as the user's heart rate, galvanic skin response, and hormone levels.

[0814] 2. Data transmission module (terminal) - transmits the acquired biometric data to the server.

[0815] 3. Data preprocessing module (server) - Performs preprocessing such as noise removal and normalization on the received raw data.

[0816] 4. Feature Extraction Module (Server) - Extracts important features from the preprocessed data.

[0817] 5. Emotion Recognition Module (Server) - Uses an emotion engine to infer the user's emotional state from biometric data.

[0818] 6. AI model (server) - receives extracted features and estimated emotional states as input and outputs inference results.

[0819] 7. Result correction module (server) - Corrects the inference results based on the emotional state.

[0820] 8. Result notification module (terminal) - Notifies the user of the inference results.

[0821] Processing flow

[0822] 1. Biometric data acquisition

[0823] The device captures second-by-second physical information such as the user's heart rate, galvanic skin response, and hormone levels using wearable devices and sensors attached to the skin.

[0824] 2. Data transmission

[0825] The device transmits the acquired raw data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[0826] 3. Data preprocessing and feature extraction

[0827] The server performs filtering operations on the received raw data to remove noise, for example, smoothing the heart rate data using a moving average filter, and normalization, scaling the data to a range from 0 to 1.

[0828] Extracting important features from normalized data, for example detecting sudden fluctuations in heart rate or peaks in electrodermal response.

[0829] 4. Emotional State Estimation

[0830] The server uses an emotion recognition module to estimate the user's emotional state from the preprocessed biometric data, for example, by using cosine similarity and emotion models to assess the user's stress state and happiness.

[0831] 5. Input to the AI ​​model

[0832] The server inputs the extracted feature data and estimated emotional state into an AI model, which then uses this information to make predictions for earthquake prediction and investment decision-making.

[0833] 6. Correction of results

[0834] The server adjusts the inference results of the AI ​​model based on the emotional state, enabling more accurate decision-making by adjusting the impact of the emotional state on the results.

[0835] 7. Output and notification of results

[0836] The server converts the final inference results into a format that is easy for the user to understand, and then sends the results to the device, which then notifies the user in various ways, such as by sound alert, vibration, or display on the screen.

[0837] Specific examples

[0838] Examples of earthquake prediction

[0839] 1. Acquisition of physical information

[0840] The device captures the user's heart rate and electrical skin response every second and sends it to a server.

[0841] 2. Data preprocessing and feature extraction

[0842] The server removes noise from the received data using a moving average filter and normalizes it.

[0843] Detects rapid fluctuations in heart rate and peaks in electrodermal response.

[0844] 3. Emotional State Estimation

[0845] The server uses an emotion engine to infer the user's emotional state (e.g., fear or tension) from the pre-processed data.

[0846] 4. Input to the AI ​​model

[0847] The server inputs the extracted feature data and emotional state into an earthquake prediction AI model to infer the probability of an earthquake occurring.

[0848] 5. Correction of results

[0849] The server will take into account the user's emotional state to correct the inference results, for example by weighting and adjusting the results if the emotional state has a strong influence on the predictions.

[0850] 6. Output and notification of results

[0851] The server converts the probability of an earthquake occurring into "high risk," "medium risk," or "low risk," and sends it to the terminal.

[0852] If the risk is high, the device will notify the user, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[0853] Investment Decision Examples

[0854] 1. Acquisition of physical information

[0855] The device measures the user's heart rate, galvanic skin response, and hormone levels every second and transmits them to a server.

[0856] 2. Data preprocessing and feature extraction

[0857] The server processes this data for noise reduction and normalization.

[0858] Detects heart rate variability, peaks in electrodermal response, and surges in hormone levels.

[0859] 3. Emotional State Estimation

[0860] The server uses an emotion engine to estimate the user's emotional state (e.g., anxiety or depression) from the preprocessed data.

[0861] 4. Input to the AI ​​model

[0862] The server inputs these characteristics and emotional states into an investment decision-making AI model to infer decisions on whether to buy, sell, or hold.

[0863] 5. Correction of results

[0864] The server corrects the inference results based on the user's emotional state in order to minimize the impact of the user's emotional state on investment decisions.

[0865] 6. Output and notification of results

[0866] The server receives the inference result, converts it into a format such as "You should buy XX," and sends it to the terminal.

[0867] The terminal notifies the user of the results and prompts them to take specific action.

[0868] The above is a specific embodiment for carrying out the present invention. This system enables advanced decision-making in real time by utilizing the user's physical information and emotional state, thereby increasing the probability of success in earthquake prediction and investment.

[0869] The processing flow will be explained below.

[0870] Step 1:

[0871] Acquiring physical information

[0872] The device collects real-time biometric data, such as the user's heart rate, galvanic skin response, and hormone levels, using wearable and skin-attached sensors.

[0873] Step 2:

[0874] Sending data

[0875] The device transmits the acquired biometric data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[0876] Step 3:

[0877] Receiving data

[0878] The server receives the biometric data sent from the device and stores it in chronological order.

[0879] Step 4:

[0880] Noise Reduction

[0881] The server performs filtering on the received raw data using a moving average filter or other method to remove noise, thereby smoothing the heart rate data.

[0882] Step 5:

[0883] Data normalization

[0884] The server performs a normalization process to scale the denoised data to the range of 0 to 1, thereby maintaining data consistency.

[0885] Step 6:

[0886] Feature Extraction

[0887] The server extracts important features from the normalized data, such as detecting sudden fluctuations in heart rate or peaks in electrodermal response.

[0888] Step 7:

[0889] Emotional state estimation

[0890] The server uses an emotion recognition module to estimate the user's emotional state from preprocessed biometric data, and utilizes an emotion engine to assess the user's stress state and happiness.

[0891] Step 8:

[0892] Input of feature data and emotional state

[0893] The server inputs the extracted feature data and estimated emotional state into an AI model, which uses this information to make predictions for earthquake prediction and investment decision-making.

[0894] Step 9:

[0895] AI model inference

[0896] The server inputs the characteristic data and emotional state into the AI ​​model, and infers the probability of an earthquake occurring and investment decisions such as "buy," "sell," or "hold."

[0897] Step 10:

[0898] Receiving inference results

[0899] The server receives the inference results from the AI ​​model, which are then further corrected rather than used as is.

[0900] Step 11:

[0901] Correcting the results

[0902] The server adjusts the inference results of the AI ​​model based on the emotional state, enabling more accurate decision-making by adjusting the impact of the emotional state on the results.

[0903] Step 12:

[0904] Sending the results

[0905] The server then sends the corrected inference results to the terminal, where they are converted into a format that is easy for the user to understand.

[0906] Step 13:

[0907] User Notification

[0908] The device notifies the user of the corrected inference results by means of voice alerts, vibrations, screen displays, etc., and prompts the user to take specific action.

[0909] The above is a specific flow of the program processing of the present invention, which enables advanced decision-making in real time by utilizing the user's physical information and emotional state.

[0910] Example 2

[0911] 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."

[0912] Conventional earthquake prediction and investment decision-making systems have the problem of low prediction accuracy because they do not take into account the user's emotional state or physical information. Furthermore, it is difficult to collect data and make decisions in real time, making it difficult to prompt users to take immediate action. Furthermore, data preprocessing and feature extraction are insufficient, making it difficult to obtain reliable results.

[0913] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0914] In this invention, the server includes means for acquiring a user's physical information in real time, means for transmitting the acquired physical information to the server, means for removing noise from the physical information received by the server and normalizing the data, means for extracting features from the preprocessed data, means for estimating an emotional state using an emotion recognition module, means for inputting the feature data and the emotional state into an artificial intelligence model to obtain an inference result, means for correcting the inference result based on the emotional state, and means for notifying the user of the corrected inference result. This enables highly accurate predictions that reflect the user's physical information and emotional state in real time, and allows the user to be quickly notified of specific actions.

[0915] "Physical information" refers to biometric data such as a user's heart rate, galvanic skin response, and hormone levels.

[0916] "Real-time" refers to data being collected and processed almost immediately.

[0917] "Server" refers to the computer system that receives, pre-processes, extracts features, estimates emotional states, and performs inference on biometric data.

[0918] "Terminal" refers to a device for collecting and transmitting a user's physical information to a server.

[0919] "Noise reduction" refers to the process of removing unwanted signals and disturbances from raw data.

[0920] "Normalization" refers to the process of scaling data to a particular range.

[0921] "Feature extraction" refers to the process of finding important indicators and patterns from preprocessed data.

[0922] "Emotion Recognition Module" refers to software or algorithms for inferring a user's emotional state based on biometric data.

[0923] An "artificial intelligence model" refers to a model that uses machine learning algorithms to make inferences.

[0924] "Inference results" refer to the predictions and judgments that an artificial intelligence model outputs based on input data.

[0925] "Correction" refers to the process of adjusting the initial inference result by taking into account factors such as emotional state.

[0926] "Notification" refers to the system informing the user of inference results and actions.

[0927] This invention is a system that collects a user's physical information and emotional state in real time, and uses an artificial intelligence (AI) model and emotion engine to make advanced decisions based on that information. In particular, the accuracy of earthquake prediction and investment decision-making can be improved by combining physical sensations, intuition, and emotional state.

[0928] Hardware and software used

[0929] The devices used are wearable or skin-attached sensor devices that can capture biometric data such as heart rate, galvanic skin response, and hormone levels within seconds.

[0930] Specifically, smartwatches and biosensors are used as terminals, and these devices transmit data to a server via wireless communication technologies such as Bluetooth and Wi-Fi.

[0931] On the server, Python's pandas library and sklearn's StandardScaler are used to remove noise and normalize the biometric data. OpenAI's sentiment analysis API is used to estimate emotional states. Furthermore, an AI model built with TensorFlow is used to infer the probability of earthquake occurrence and investment decisions.

[0932] Data processing and calculation

[0933] 1. Biometric data acquisition

[0934] The device captures the user's heart rate, galvanic skin response, and hormone levels every second. Specifically, the smartwatch measures heart rate and galvanic skin response, while sensors under the skin periodically measure hormone levels.

[0935] 2. Data transmission

[0936] The device transmits the acquired biometric data to a smartphone via Bluetooth, and the smartphone then transmits the data to a server using Wi-Fi.

[0937] 3. Data preprocessing and feature extraction

[0938] The server performs noise removal and data normalization on the received raw data. For example, it uses Python's pandas library to smooth the data with a moving average filter and sklearn's StandardScaler to normalize the data to the range of 0 to 1. It then extracts important features, such as rapid fluctuations in heart rate and peaks in electrodermal responses.

[0939] 4. Emotional State Estimation

[0940] The server estimates the user's emotional state from the preprocessed data using an emotion recognition module. For example, the server uses OpenAI's emotion analysis API to evaluate the user's stress level and happiness from the normalized data.

[0941] 5. Input to the AI ​​model

[0942] The server inputs the extracted feature data and estimated emotional state into an AI model to perform inference. For example, the server inputs the data into an earthquake prediction model built using TensorFlow to predict the probability of an earthquake occurring.

[0943] 6. Correction of results

[0944] The server corrects the inference results of the AI ​​model based on the emotional state, adjusting the influence of emotional states such as high stress if they have a strong impact on the results.

[0945] 7. Output and notification of results

[0946] The server converts the final inference results into a user-friendly format and sends them to the device, which notifies the user of the results by means of an alert sound, vibration, or on-screen display.

[0947] Examples and prompts

[0948] Specific examples

[0949] Examples of earthquake prediction:

[0950] The device captures the user's heart rate and galvanic skin response every second and sends them to a server. The server smooths and normalizes the data, detecting sudden fluctuations in heart rate and peaks in galvanic skin response. An emotion engine estimates fear and tension, and inputs the data into an earthquake prediction AI model. The server then classifies the risk of an earthquake occurrence as "high risk" or similar, and notifies the device of the results.

[0951] Examples of investment decisions:

[0952] The device acquires the user's heart rate, galvanic response, and hormone levels, and sends them to a server. The server preprocesses this data and detects heart rate fluctuations, peaks in galvanic response, and sudden spikes in hormone levels. An emotion engine estimates feelings of anxiety and depression, and inputs the data into an investment decision-making AI model. The server then makes a decision to "buy," "sell," or "hold," and notifies the user of the result.

[0953] Prompt Sentence Examples

[0954] Earthquake prediction prompt:

[0955] "Get the user's heart rate and electrodermal response data every second to estimate their emotional state, such as fear or tension. Then, based on this, assess the risk of an earthquake occurring and notify them if the risk is high."

[0956] Investment Decision Prompt:

[0957] "Collect data on the user's heart rate, galvanic skin response, and hormone levels to estimate their emotional state, such as anxiety or depression. Based on that, make investment decisions to buy, sell, or hold, and notify the user of the results."

[0958] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0959] Step 1:

[0960] Biometric data acquisition

[0961] The device collects real-time physical information every second, such as the user's heart rate, galvanic skin response, and hormone levels. Specifically, this data is collected by a smartwatch or sub-skin sensors.

[0962] Input: User's physical information

[0963] Output: Acquired biometric data (heart rate, galvanic skin response, hormone levels)

[0964] Step 2:

[0965] Sending data

[0966] The device transmits the acquired biometric data to a server at regular intervals. Specifically, the data is transferred to a smartphone via Bluetooth, and the smartphone then transmits the data to the server using Wi-Fi.

[0967] Input: Biometric data (heart rate, galvanic skin response, hormone levels)

[0968] Output: Biometric data sent to the server

[0969] Step 3:

[0970] Data Preprocessing

[0971] The server performs noise removal and normalization on the received biometric data. Specifically, it smooths the data with a moving average filter using Python's pandas library and normalizes the data to the range of 0 to 1 using sklearn's StandardScaler.

[0972] Input: Received biometric data

[0973] Output: Denoised normalized data

[0974] Step 4:

[0975] Feature Extraction

[0976] The server extracts important features from the denoised and normalised data, specifically detecting rapid fluctuations in heart rate and peaks in electrodermal response.

[0977] Input: Denoised normalized data

[0978] Output: Extracted feature data (heart rate spikes, electrodermal response peaks)

[0979] Step 5:

[0980] Emotional state estimation

[0981] The server uses an emotion recognition module to estimate the user's emotional state from the preprocessed data. For example, it uses OpenAI's emotion analysis API to evaluate the user's stress level or happiness from the data.

[0982] Input: Extracted feature data

[0983] Output: Estimated emotional state (stress, happiness)

[0984] Step 6:

[0985] Input to the AI ​​model

[0986] The server inputs the extracted feature data and estimated emotional state into a generative AI model to perform inference for earthquake prediction and investment decision-making. Specifically, the server inputs the data into an AI model using TensorFlow.

[0987] Input: extracted feature data, estimated emotional state

[0988] Output: Initial inference results (probability of earthquake occurrence, investment decision)

[0989] Step 7:

[0990] Correcting the results

[0991] The server corrects the initial inference results of the AI ​​model based on the emotional state, adjusting its influence if the emotional state has a strong influence on the results.

[0992] Input: Initial inference result, estimated emotional state

[0993] Output: Corrected inference results

[0994] Step 8:

[0995] Notification of results

[0996] The server converts the corrected inference results into a user-friendly format and sends them to the device, which notifies the user of the results by sound alerts, vibrations, or a display on the screen.

[0997] Input: Corrected inference results

[0998] Output: The result reported to the user

[0999] (Application example 2)

[1000] 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."

[1001] Conventional smart home security systems primarily detect anomalies based on environmental data, without taking into account the user's physical condition or emotional state. As a result, they sometimes fail to detect actual anomalies, or sometimes generate frequent false alarms. This invention aims to improve the accuracy of anomaly detection by combining the user's biometric information and emotional state, providing a more reliable security experience.

[1002] 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.

[1003] In this invention, the server includes means for acquiring biometric data, means for preprocessing the acquired biometric data, means for extracting features from the preprocessed biometric data, means for inputting the extracted features into an artificial intelligence model to obtain an inference result, means for outputting the obtained inference result and notifying the user, means for estimating the emotional state of the user and detecting an abnormality, and means for notifying the user of a specific countermeasure based on the abnormality detection result. This enables an abnormality detection that reflects the user's physical condition and emotional state.

[1004] "Biometric data" is data indicative of a user's physiological state, such as a user's heart rate, galvanic skin response, or hormone levels.

[1005] "Preprocessing" is a process that removes noise from biometric data and prepares the data in a form that is easier to analyze.

[1006] "Feature extraction" is the process of extracting important patterns and trends from pre-processed data.

[1007] An "artificial intelligence model" is an algorithm or system that learns from large amounts of data and makes appropriate inferences about unknown data.

[1008] "Inference results" are predictions or judgments obtained from data processed by an artificial intelligence model.

[1009] "Notification means" refers to a means for notifying the user of inference results or anomaly detection results, particularly via a smartphone or other device.

[1010] "Emotional state" refers to the psychological state such as anxiety, tension, or stress felt by the user.

[1011] "Anomaly detection" is the process of recognizing safety risks and anomalies based on a person's biometric data and emotional state.

[1012] "Countermeasures" are specific actions or procedures that a user should take in response to a detected abnormality.

[1013] This invention is a security service-related system that collects a user's physical information and emotional state in real time and makes advanced decisions based on that data using an artificial intelligence model and emotion engine. In particular, it includes technology to improve the accuracy of anomaly detection in smart home security systems.

[1014] System Configuration

[1015] The system consists of the following main components:

[1016] 1. Wearable devices: collect real-time biometric data such as the user's heart rate, galvanic skin response, and hormone levels.

[1017] 2. Smartphone: A device for transmitting biometric data acquired from wearable devices to a server.

[1018] 3. Server: Includes the following software modules:

[1019] Data preprocessing module: Removes noise and normalizes raw data (software used: Azure Data Factory).

[1020] Feature Extraction Module: Extracts important features from the preprocessed data.

[1021] Emotion recognition module: Estimates emotional state from biometric data (software used: Google Cloud AI).

[1022] AI model: Detects anomalies based on extracted features and emotional states (software used: TensorFlow).

[1023] Result notification module: Notifies the user of the results of anomaly detection via their smartphone (software used: Firebase Cloud Messaging).

[1024] Processing flow

[1025] The server first receives biometric data acquired from the wearable device, denoises and normalizes it using Azure Data Factory, and then extracts features from the preprocessed data, such as sudden fluctuations in heart rate and peaks in electrodermal responses.

[1026] The emotion recognition module uses Google Cloud AI to infer the user's emotional state from pre-processed data and identify states such as anxiety or tension. These features and emotional state are then fed into a TensorFlow-powered AI model for anomaly detection.

[1027] As a result, if an abnormality is detected, a notification will be sent to the user's smartphone via Firebase Cloud Messaging with instructions on how to deal with the problem. The user can then take appropriate action based on the notification they receive.

[1028] Specific examples

[1029] For example, in a "smart home security system equipped with emotion sensors," if a user experiences a sudden rise in heart rate and a peak in their electrodermal response, the emotion recognition module will determine that the user is in a state of high anxiety, and this state will be registered as an abnormality in the system. The AI ​​model will also analyze environmental data (for example, whether the door is open or closed, and the sound level in the room), and if it finally detects an abnormality, it will send a notification to the user's smartphone saying, "The door is not closed. Please check."

[1030] Prompt Sentence Examples

[1031] An example of a prompt is "Design a smart home security system that monitors heart rate and galvanic skin response in real time, and detects and notifies users if anxiety levels are high."

[1032] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1033] Step 1:

[1034] Biometric data acquisition

[1035] The terminal (wearable device) collects real-time data on the user's heart rate, skin galvanic response, and hormone levels. This data is collected from sensors and received by a smartphone. The input is the user's biometric data, and the output is the raw data received by the smartphone. Specifically, data is recorded every second and sent to the smartphone via Bluetooth or other means.

[1036] Step 2:

[1037] Sending data

[1038] The device (smartphone) sends the received biometric data to a server using Wi-Fi or Bluetooth. The server receives this data and supplies it to the next processing step. The input is the acquired raw data, and the output is the data sent to the server. Specifically, the smartphone application packets the data at intervals and sends it.

[1039] Step 3:

[1040] Data Preprocessing

[1041] The raw data sent to the server is denoised and normalised using Azure Data Factory. This purifies the data and makes it easier to analyse. The input is the raw data sent to the server, and the output is the denoised and normalised data. Specific operations include smoothing the heart rate data using a moving average filter and scaling the data from 0 to 1.

[1042] Step 4:

[1043] Feature Extraction

[1044] Important features such as sudden changes in heart rate and peaks in electrodermal responses are extracted from data that has been preprocessed on the server. The input is the preprocessed data, and the output is the extracted features. Specifically, an algorithm is used to analyze the temporal fluctuations of the data and identify sudden changes and peak values.

[1045] Step 5:

[1046] Emotional state estimation

[1047] The extracted feature data is input into Google Cloud AI's emotion recognition module, which estimates the user's emotional state. The input is the extracted feature data, and the output is the estimated emotional state. Specifically, stress and anxiety states are visualized using techniques such as cosine similarity.

[1048] Step 6:

[1049] Input to AI model and anomaly detection

[1050] The estimated emotional state and feature data are input into an AI model (TensorFlow) to perform anomaly detection. The input is the feature data and emotional state, and the output is the anomaly detection result. Specifically, the trained AI model analyzes this data and determines whether there is an anomaly.

[1051] Step 7:

[1052] Result notification

[1053] If an abnormality is detected, the server uses Firebase Cloud Messaging to send a notification to the user's smartphone. The input is the abnormality detection result, and the output is the notification that is displayed on the user's smartphone. Specifically, a message such as "The door is not closed. Please check" is sent.

[1054] Step 8:

[1055] User Action

[1056] The user receives the notification and takes appropriate action based on the specific countermeasures shown. The input is the notification displayed on the smartphone, and the output is the user's action. Specifically, the user checks the door and takes action such as closing it if necessary.

[1057] This enables highly accurate anomaly detection and notification based on the user's biometric information and emotional state.

[1058] 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.

[1059] 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.

[1060] 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.

[1061] [Third embodiment]

[1062] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1063] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[1064] 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).

[1065] 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.

[1066] 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.

[1067] 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).

[1068] 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.

[1069] 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.

[1070] 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.

[1071] 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.

[1072] 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.

[1073] 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."

[1074] This invention relates to a system that collects a user's physical information in real time and uses an artificial intelligence (AI) model based on that information to make advanced decisions. In particular, accuracy can be improved in earthquake prediction and investment decision-making by utilizing physical sensations and intuition. Specific embodiments for carrying out the present invention are described below.

[1075] System Configuration

[1076] The system consists of the following main components:

[1077] 1. Sensor device (terminal) - Acquires real-time physical information such as the user's heart rate, galvanic skin response, and hormone levels.

[1078] 2. Data transmission module (terminal) - transmits the acquired biometric data to the server.

[1079] 3. Data preprocessing module (server) - Performs preprocessing such as noise removal and normalization on the received raw data.

[1080] 4. Feature Extraction Module (Server) - Extracts important features from the preprocessed data.

[1081] 5. AI model (server) - receives extracted features as input and outputs inference results.

[1082] 6. Result notification module (terminal) - Notifies the user of the inference results.

[1083] Processing flow

[1084] 1. Biometric data acquisition

[1085] The device captures data such as the user's heart rate, galvanic skin response, and hormone levels every second, using wearable devices and sensors attached to the skin.

[1086] 2. Data transmission

[1087] The device transmits the acquired raw data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[1088] 3. Data Preprocessing

[1089] The server performs filtering operations on the received raw data to remove noise, for example, smoothing the heart rate data using a moving average filter, and normalization, scaling the data to a range from 0 to 1.

[1090] 4. Feature Extraction

[1091] The server extracts features from the preprocessed data, such as sudden fluctuations in heart rate and peaks in skin galvanic responses, thereby capturing the user's physical reactions to precursors to earthquakes or sudden market fluctuations.

[1092] 5. Linking to AI models

[1093] The server inputs the extracted feature data into the AI ​​model, which compares it with the data it has learned so far to determine the probability of an earthquake occurring and whether to "buy," "sell," or "hold" in investment decisions.

[1094] 6. Output and notification of results

[1095] The server receives the inference results of the AI ​​model, converts them into a format that is easy for the user to understand, and sends the results to the device, which then notifies the user in various ways, such as by sound alert, vibration, or display on the screen.

[1096] Specific examples

[1097] Examples of earthquake prediction

[1098] 1. Acquisition of physical information

[1099] The device captures the user's heart rate and electrical skin response every second and sends it to a server.

[1100] 2. Data Preprocessing

[1101] The server removes noise from the received data using a moving average filter and normalizes it.

[1102] 3. Feature Extraction

[1103] The server detects rapid fluctuations in heart rate and peaks in electrodermal response.

[1104] 4. Linking to AI models

[1105] The server inputs these features into an earthquake prediction AI model to infer the probability of an earthquake occurring.

[1106] 5. Outputting the results

[1107] The server converts the probability of an earthquake occurring into "high risk," "medium risk," or "low risk," and sends it to the terminal.

[1108] 6. User Notices

[1109] If the risk is high, the device will notify the user, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[1110] Investment Decision Examples

[1111] 1. Acquisition of physical information

[1112] The device measures the user's heart rate, galvanic skin response, and hormone levels every second and transmits them to a server.

[1113] 2. Data Preprocessing

[1114] The server processes this data for noise reduction and normalization.

[1115] 3. Feature Extraction

[1116] The server detects heart rate variability, peaks in galvanic skin response, and spikes in hormone levels.

[1117] 4. Linking to AI models

[1118] The server inputs these characteristics into an investment decision-making AI model and infers the decision to "buy," "sell," or "hold."

[1119] 5. Outputting the results

[1120] The server receives the inference results, converts them into a format such as "You should buy XX," and sends them to the terminal.

[1121] 6. User Notices

[1122] The terminal notifies the user of the results and suggests specific actions to take.

[1123] The above is a specific embodiment for carrying out the present invention. This system enables advanced decision-making using the user's physical information, thereby increasing the probability of success in earthquake prediction and investment.

[1124] The processing flow will be explained below.

[1125] Step 1:

[1126] Acquiring physical information

[1127] The device collects real-time biometric data such as the user's heart rate, galvanic skin response, and hormone levels using wearable and skin-attached sensors.

[1128] Step 2:

[1129] Sending data

[1130] The device transmits the acquired biometric data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[1131] Step 3:

[1132] Receiving data

[1133] The server receives the biometric data sent from the device and stores it in chronological order.

[1134] Step 4:

[1135] Noise Reduction

[1136] The server performs filtering on the received raw data to remove noise, for example, applying a moving average filter to smooth the heart rate data.

[1137] Step 5:

[1138] Data normalization

[1139] The server normalizes the denoised data, scaling the heart rate data to a range of 0 to 1 to make it consistent with other biometric data.

[1140] Step 6:

[1141] Feature Extraction

[1142] The server extracts important features from the normalized data, applying algorithms to detect, for example, sudden fluctuations in heart rate or peaks in electrodermal response.

[1143] Step 7:

[1144] Entering feature data

[1145] The server inputs the extracted feature data into an AI model that has been trained for earthquake prediction and investment decision-making.

[1146] Step 8:

[1147] AI model inference

[1148] The server inputs the characteristic data into the AI ​​model and performs inference to output the probability of an earthquake occurring and investment decisions such as "buy," "sell," or "hold."

[1149] Step 9:

[1150] Receiving inference results

[1151] The server receives the inference results from the AI ​​model and converts them into a format that is easy for users to understand.

[1152] Step 10:

[1153] Sending the results

[1154] The server then sends the converted inference results to the terminal, which then provides the results to the user as an alert message.

[1155] Step 11:

[1156] User Notification

[1157] The device notifies the user of the inference results by means of voice alerts, vibrations, screen displays, etc., and prompts the user to take specific action.

[1158] The above is a specific flow of the program processing of the present invention, which enables advanced decision-making in real time using the user's physical information.

[1159] Example 1

[1160] 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."

[1161] Existing systems have a problem with low accuracy in earthquake prediction and investment decision-making using users' biometric data. This is because the processes of biometric data acquisition, preprocessing, feature extraction, inference, and notification are performed separately, reducing the overall system efficiency. Another problem is that insufficient noise removal and normalization of biometric data can easily produce inaccurate inference results. To solve these problems, an advanced integrated system is required.

[1162] 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.

[1163] In this invention, the server includes means for acquiring biometric data, means for preprocessing the acquired biometric data, means for extracting features from the preprocessed biometric data, means for inputting the extracted features into an artificial intelligence model to obtain an inference result, means for outputting the obtained inference result and notifying the user, means for using a wearable device or a sensor attached to the skin, means for transmitting data using wireless communication technology such as Bluetooth or Wi-Fi, means for smoothing data using a moving average filter, means for normalizing data to a range from 0 to 1, and means for converting the inference result into "high risk," "medium risk," or "low risk." This enables advanced decision-making using a user's biometric data to be performed accurately in real time.

[1164] "Biometric data" is information obtained from a user's body, including heart rate, galvanic skin response, hormone levels, and the like.

[1165] "Preprocessing" refers to applying filtering, normalization, and other processes to acquired biometric data to remove noise and convert the data into a format that is easier to analyze.

[1166] "Feature extraction" is the process of finding important patterns and variations in pre-processed data and preparing it for input into an artificial intelligence model.

[1167] An "artificial intelligence model" is a software system that uses machine learning algorithms to analyze data and draw inferences.

[1168] "Inference results" are the conclusions or judgments that an AI model provides based on biometric data, such as the probability of an earthquake occurring or decisions about whether to "buy," "sell," or "hold" an investment.

[1169] "Notification" is a means of conveying the inference results to the user, and is implemented in the form of an alert sound, vibration, screen display, etc.

[1170] A "wearable device" is a device worn by a user that can acquire biometric data, such as a smartwatch or activity tracker.

[1171] A "sensor" is a device for measuring biometric data, and includes types that are attached to the skin and types that are worn on the arm.

[1172] "Wireless communication technology" refers to technology for sending and receiving data wirelessly, and includes Bluetooth and Wi-Fi.

[1173] A "moving average filter" is a filtering method used to smooth out fluctuations in data, and is a method of removing noise by taking the average over time.

[1174] "Normalization" is a process performed to unify the scale of data, and involves converting the scale of biometric data to a range of 0 to 1.

[1175] "Risk conversion" is the process of converting inference results into categories such as "high risk," "medium risk," and "low risk," and presenting them in a form that is easy for users to understand.

[1176] The system of the present invention is configured to collect biometric data of a user in real time and make advanced decisions using an artificial intelligence (AI) model. Detailed embodiments of the system are described below.

[1177] System Configuration

[1178] This system mainly consists of the following hardware and software components.

[1179] 1. Sensor device (terminal)

[1180] Sensor devices are used to obtain real-time biometric data such as a user's heart rate, electrodermal response, and hormone levels. Specific examples include wearable devices (smartwatches, activity trackers) and sensors attached to the skin. For example, Apple Watch and general skin electrode sensors can be used.

[1181] 2. Data transmission module (terminal)

[1182] The data transmission module is responsible for transmitting the biometric data acquired from the sensor device to the server, using wireless communication technologies such as Bluetooth and Wi-Fi.

[1183] 3. Data preprocessing module (server)

[1184] The data preprocessing module performs noise removal and normalization on the biometric data received on the server, specifically smoothing the data using a moving average filter and scaling the biometric data to a range of 0 to 1.

[1185] 4. Feature extraction module (server)

[1186] The feature extraction module extracts important features from the pre-processed data, such as rapid heart rate fluctuations or peaks in electrodermal response.

[1187] 5. AI model (server)

[1188] This module inputs the extracted feature data into the AI ​​model to obtain inference results. It uses a deep learning algorithm to compare the results with past data and make inferences.

[1189] 6. Result notification module (terminal)

[1190] This module notifies the user of the inference results. Notifications are given in the form of an alert sound, vibration, screen display, etc.

[1191] Specific examples

[1192] Examples of earthquake prediction

[1193] When a user uses a sensor device (e.g., a smartwatch), their heart rate and electrodermal response are acquired every second. This data is sent to a server via Bluetooth. The server smooths the heart rate data using a moving average filter and normalizes the data. The server then detects sudden heart rate fluctuations and peaks in electrodermal response from the preprocessed data. This feature data is input into an AI earthquake prediction model to infer the probability of an earthquake occurring. Finally, the server converts the inference result into "high risk," "medium risk," or "low risk" and sends it to the device. If the risk is high, the device notifies the user, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[1194] Investment Decision Examples

[1195] The user uses the same sensor device to measure their heart rate, galvanic response, and hormone levels every second. This data is sent to a server, which then denoises and normalizes it. Heart rate fluctuations, peaks in galvanic response, and spikes in hormone levels are then detected. This characteristic data is then input into an AI investment decision-making model, which then infers whether to "buy," "sell," or "hold." The server then converts this inference result into a concrete format, such as "You should buy XX," and sends it to the device. The device then notifies the user, "You should buy XX," and suggests a specific action to take.

[1196] Prompt Sentence Examples

[1197] Example prompts to be input to the generative AI model:

[1198] The user's heart rate fluctuates rapidly and their electrodermal response shows a peak. These can be fed into an AI model to assess the risk of an earthquake occurring.

[1199] This system enables advanced decision-making in real time using the user's biometric data, increasing the chances of successful earthquake prediction and investment.

[1200] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1201] Step 1:

[1202] Biometric data acquisition

[1203] Subject: Device

[1204] Description: The device uses sensor devices to collect real-time biometric data such as the user's heart rate, galvanic skin response, and hormone levels. Input data comes from various sensors worn by the user. The device converts this signal into digital data and updates it every second.

[1205] How it works: The device receives signals from the smartwatch and skin electrodes. For example, the optical heart rate monitor in the Apple Watch measures your heart rate beat-by-beat, and the galvanic skin response sensor captures your stress level. This data is temporarily stored on the device.

[1206] Step 2:

[1207] Sending data

[1208] Subject: Device

[1209] Description: The device transmits the acquired biometric data to the server at regular intervals (e.g., every second). The input data is the biometric data acquired from the sensor device, and the output data is the data packet sent to the server.

[1210] How it works: The device sends data to the server via Bluetooth or Wi-Fi, for example using the low-latency, highly reliable Bluetooth Low Energy (BLE) protocol to ensure that the biometric data reaches the server reliably.

[1211] Step 3:

[1212] Data Preprocessing

[1213] Subject: Server

[1214] Description: The server performs preprocessing on the received biometric data, specifically noise removal and normalization. The input data is the biometric data sent from the device, and the output data is the preprocessed data.

[1215] What it does: The server applies a moving average filter to remove spikes and noise from the data. For example, a 3-second moving average filter is used to smooth out fluctuations in heart rate data. At the same time, it normalizes the data by scaling it to a range of 0 to 1, which makes it easier to extract features and input into AI models.

[1216] Step 4:

[1217] Feature Extraction

[1218] Subject: Server

[1219] Description: The server extracts important features from preprocessed data. The input data is the preprocessed data, and the output data is the extracted feature data.

[1220] How it works: The server detects features such as sudden fluctuations in heart rate data and peaks in galvanic response (GDS). It then uses an algorithm to extract, for example, spikes in heart rate or peaks in GDS, thereby identifying moments of stress or excitement felt by the user.

[1221] Step 5:

[1222] Linking to AI models

[1223] Subject: Server

[1224] Description: The server inputs the extracted feature data into the AI ​​model and obtains the inference result. The input data is the feature data, and the output data is the inference result from the AI ​​model.

[1225] How it works: The server uses a trained deep learning model to input feature data into an AI model. The AI ​​model compares this data with past data to determine the probability of an earthquake occurring and whether to "buy," "sell," or "hold" an investment. The model outputs its inference results in real time.

[1226] Step 6:

[1227] Output and notification of results

[1228] Subject: Server

[1229] Description: The server converts the inference results of the AI ​​model into an easy-to-understand format and sends it to the terminal. The input data is the inference results from the AI ​​model, and the output data is the final result that is notified to the user.

[1230] Specific operation: The server classifies the inference results into "high risk," "medium risk," or "low risk" and sends them to the device in JSON format. The device analyzes the received data and notifies the user with a specific message. For example, it warns the user with an alert sound or vibration saying, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[1231] The above is a detailed description of the specific operations and data flow at each processing step of this system. This detailed flow enables advanced decision-making using the user's biometric data in real time.

[1232] (Application example 1)

[1233] 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."

[1234] Conventional content delivery systems primarily recommend content based on users' preferences and interests, but this approach fails to respond to users' real-time emotional states. This results in a poor user experience and reduces their willingness to consume content. Furthermore, the lack of a mechanism for capturing changes in emotional states in real time makes it difficult to provide users with appropriate feedback.

[1235] 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.

[1236] In this invention, the server includes means for acquiring biometric data, means for preprocessing the acquired biometric data, means for extracting features from the preprocessed biometric data, means for inputting the extracted features into an artificial intelligence model to obtain an inference result, means for recommending optimal content to a user based on the obtained inference result, means for notifying the user of the recommended content and monitoring emotional changes in real time as the user views the content, and means for providing feedback based on the monitored emotional changes. This enables personalized content recommendations based on the user's real-time emotional state and the provision of appropriate feedback.

[1237] Below are definitions of important terms included in the claims, tailored to application examples.

[1238] "Biometric data" refers to physical information such as a user's heart rate, galvanic skin response, and hormone levels.

[1239] "Preprocessing" refers to performing processes such as noise removal and normalization on acquired biometric data.

[1240] "Feature extraction" refers to extracting important features, such as rapid heart rate fluctuations or peaks in electrodermal responses, from preprocessed biometric data.

[1241] An "artificial intelligence model" refers to a machine learning algorithm or neural network that makes inferences or decisions based on given input data.

[1242] "Inference result" refers to the prediction or judgment result that an artificial intelligence model outputs based on input data.

[1243] "Content recommendation" refers to the process of suggesting optimal videos, music, and other media to users based on their biometric data.

[1244] "Changes in emotion" refers to fluctuations in emotional state, such as excitement, joy, or sadness, that a user experiences when viewing content.

[1245] "Providing feedback" refers to suggesting appropriate advice or next actions to the user based on monitored emotional changes.

[1246] This invention is a system that uses a user's biometric data to recommend optimal content in real time and provide appropriate feedback. This system consists of the following main components:

[1247] System Configuration

[1248] 1. Means of acquiring biometric data

[1249] Means: A wearable device for measuring the user's heart rate or galvanic skin response (e.g., a heart rate sensor or galvanic skin response sensor).

[1250] Hardware used: General wearable devices (e.g., smartwatches, fitness bands, etc.)

[1251] 2. Pretreatment Methods

[1252] Method: Denoise and normalize the acquired biometric data

[1253] Software used: Python numerical calculation libraries (e.g., numpy, pandas), data preprocessing libraries (e.g., scikit-learn)

[1254] 3. Feature Extraction Method

[1255] Method: Extract features such as heart rate variability and electrodermal response peaks from the preprocessed data.

[1256] Software used: Python machine learning library (e.g., scikit-learn)

[1257] 4. Artificial Intelligence Models

[1258] Method: Take the extracted features as input and infer the user's real-time emotional state.

[1259] Software used: Deep learning libraries (e.g., TensorFlow, Keras)

[1260] Model: Emotional Inference Model

[1261] 5. Content recommendation method based on inference results

[1262] Method: Recommending optimal content to users based on inference results from an AI model

[1263] Software used: Recommendation systems (e.g., content recommendation algorithms)

[1264] 6. User Notification Methods

[1265] Method: Notify the smartphone of the recommended content so that the user can watch it.

[1266] Hardware used: Smartphone

[1267] Software used: Notification library (e.g., Firebase Cloud Messaging)

[1268] 7. Emotion Monitoring Tools

[1269] Method: While the user is watching the content, heart rate and electrodermal response are collected again to monitor changes in emotions in real time.

[1270] Hardware used: General wearable devices

[1271] Software used: Dedicated SDK (e.g., Apple HealthKit, Fitbit SDK)

[1272] 8. Means of Providing Feedback

[1273] Method: Suggest the next content to watch or specific actions based on changes in emotions

[1274] Software used: Recommendation systems (e.g., feedback generation algorithms)

[1275] Specific examples

[1276] For example, when a user wears a smartwatch and watches a video, the smartwatch measures the user's heart rate and electrodermal response and sends the data to a smartphone. The smartphone then transmits this biometric data to a server at regular intervals. The server then preprocesses the received data and extracts features. An AI model uses this feature data to infer the user's emotional state and recommends the next optimal piece of content based on the inference. The smartphone then notifies the user of the content and provides appropriate feedback by monitoring emotional changes in real time while watching.

[1277] Prompt Sentence Examples

[1278] "Please infer the user's real-time emotional state from their heart rate and electrodermal response based on the content they are currently viewing, and recommend the best content for them to watch next."

[1279] This allows users to enjoy content that is always tailored to their emotional state, resulting in a system that provides an even more entertaining experience.

[1280] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1281] Step 1:

[1282] The user puts on the wearable device.

[1283] The device measures the user's heart rate and electrical skin response in real time, and transmits this data to a smartphone via Bluetooth or other means.

[1284] Input: User's heart rate, galvanic skin response

[1285] Data processing: Acquired from wearable devices

[1286] Output: Biometric data is sent to a smartphone

[1287] Step 2:

[1288] The smartphone receives biometric data from the wearable device and sends it to a server at regular intervals, using Wi-Fi or mobile data for communication.

[1289] Input: Biometric data (heart rate, electrodermal response)

[1290] Data processing: Buffering data at regular intervals

[1291] Output: Biometric data is sent to the server

[1292] Step 3:

[1293] The server performs noise reduction and normalization on the received biometric data, for example, smoothing the heart rate data using a moving average filter.

[1294] Input: Biometric data (heart rate, electrodermal response)

[1295] Data calculation: Noise removal (moving average filter), normalization

[1296] Output: Preprocessed biometric data

[1297] Step 4:

[1298] The server extracts features such as rapid fluctuations in heart rate and peaks in electrodermal response from the preprocessed biometric data.

[1299] Input: Preprocessed biometric data

[1300] Data calculation: feature extraction (rapid fluctuations, peak detection)

[1301] Output: feature data

[1302] Step 5:

[1303] The server inputs the extracted feature data into an artificial intelligence model to infer the user's emotional state.

[1304] Input: feature data

[1305] Data Computation: Emotional Inference with AI Models

[1306] Output: Inference result (emotional state)

[1307] Step 6:

[1308] The server uses a feedback and recommendation algorithm to recommend the most suitable content to the user based on the inference results.

[1309] Input: Inference result (emotional state)

[1310] Data processing: Content selection by recommendation algorithm

[1311] Output: Recommended content

[1312] Step 7:

[1313] The smartphone notifies the user of the recommended content sent from the server using services such as Firebase Cloud Messaging.

[1314] Input: Recommended content

[1315] Data operation: Notification preparation (message generation)

[1316] Output: Content recommendation notification

[1317] Step 8:

[1318] While the user is watching the content, the smartphone again collects biometric data from the wearable device and monitors emotional changes in real time.

[1319] Input: Biometric data during viewing (heart rate, skin galvanic response)

[1320] Data calculation: Real-time monitoring

[1321] Output: User's emotional change data

[1322] Step 9:

[1323] The server uses the emotion change data to provide the next feedback and recommend content, and generates appropriate feedback using an algorithm.

[1324] Input: Emotion change data

[1325] Data calculation: feedback generation, next content recommendation

[1326] Output: Feedback to the user, next content recommendation

[1327] The above are the specific processing steps for implementing the present invention. This system allows users to enjoy content that is always tailored to their emotional state, improving their entertainment experience.

[1328] 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.

[1329] This invention relates to a system that collects a user's physical information and emotional state in real time and uses an artificial intelligence (AI) model and an emotion engine to make advanced decisions based on that information. In particular, accuracy can be improved in earthquake prediction and investment decision-making by combining physical sensations, intuition, and emotional state. Specific embodiments for implementing the present invention are described below.

[1330] System Configuration

[1331] The system consists of the following main components:

[1332] 1. Sensor device (terminal) - Acquires real-time physical information such as the user's heart rate, galvanic skin response, and hormone levels.

[1333] 2. Data transmission module (terminal) - transmits the acquired biometric data to the server.

[1334] 3. Data preprocessing module (server) - Performs preprocessing such as noise removal and normalization on the received raw data.

[1335] 4. Feature Extraction Module (Server) - Extracts important features from the preprocessed data.

[1336] 5. Emotion Recognition Module (Server) - Uses an emotion engine to infer the user's emotional state from biometric data.

[1337] 6. AI model (server) - receives extracted features and estimated emotional states as input and outputs inference results.

[1338] 7. Result correction module (server) - Corrects the inference results based on the emotional state.

[1339] 8. Result notification module (terminal) - Notifies the user of the inference results.

[1340] Processing flow

[1341] 1. Biometric data acquisition

[1342] The device captures second-by-second physical information such as the user's heart rate, galvanic skin response, and hormone levels using wearable devices and sensors attached to the skin.

[1343] 2. Data transmission

[1344] The device transmits the acquired raw data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[1345] 3. Data preprocessing and feature extraction

[1346] The server performs filtering operations on the received raw data to remove noise, for example, smoothing the heart rate data using a moving average filter, and normalization, scaling the data to a range from 0 to 1.

[1347] Extracting important features from normalized data, for example detecting sudden fluctuations in heart rate or peaks in electrodermal response.

[1348] 4. Emotional State Estimation

[1349] The server uses an emotion recognition module to estimate the user's emotional state from the preprocessed biometric data, for example, by using cosine similarity and emotion models to assess the user's stress state and happiness.

[1350] 5. Input to the AI ​​model

[1351] The server inputs the extracted feature data and estimated emotional state into an AI model, which then uses this information to make predictions for earthquake prediction and investment decision-making.

[1352] 6. Correction of results

[1353] The server adjusts the inference results of the AI ​​model based on the emotional state, enabling more accurate decision-making by adjusting the impact of the emotional state on the results.

[1354] 7. Output and notification of results

[1355] The server converts the final inference results into a format that is easy for the user to understand, and then sends the results to the device, which then notifies the user in various ways, such as by sound alert, vibration, or display on the screen.

[1356] Specific examples

[1357] Examples of earthquake prediction

[1358] 1. Acquisition of physical information

[1359] The device captures the user's heart rate and electrical skin response every second and sends it to a server.

[1360] 2. Data preprocessing and feature extraction

[1361] The server removes noise from the received data using a moving average filter and normalizes it.

[1362] Detects rapid fluctuations in heart rate and peaks in electrodermal response.

[1363] 3. Emotional State Estimation

[1364] The server uses an emotion engine to infer the user's emotional state (e.g., fear or tension) from the pre-processed data.

[1365] 4. Input to the AI ​​model

[1366] The server inputs the extracted feature data and emotional state into an earthquake prediction AI model to infer the probability of an earthquake occurring.

[1367] 5. Correction of results

[1368] The server will take into account the user's emotional state to correct the inference results, for example by weighting and adjusting the results if the emotional state has a strong influence on the predictions.

[1369] 6. Output and notification of results

[1370] The server converts the probability of an earthquake occurring into "high risk," "medium risk," or "low risk," and sends it to the terminal.

[1371] If the risk is high, the device will notify the user, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[1372] Investment Decision Examples

[1373] 1. Acquisition of physical information

[1374] The device measures the user's heart rate, galvanic skin response, and hormone levels every second and transmits them to a server.

[1375] 2. Data preprocessing and feature extraction

[1376] The server processes this data for noise reduction and normalization.

[1377] Detects heart rate variability, peaks in electrodermal response, and surges in hormone levels.

[1378] 3. Emotional State Estimation

[1379] The server uses an emotion engine to estimate the user's emotional state (e.g., anxiety or depression) from the preprocessed data.

[1380] 4. Input to the AI ​​model

[1381] The server inputs these characteristics and emotional states into an investment decision-making AI model to infer decisions on whether to buy, sell, or hold.

[1382] 5. Correction of results

[1383] The server corrects the inference results based on the user's emotional state in order to minimize the impact of the user's emotional state on investment decisions.

[1384] 6. Output and notification of results

[1385] The server receives the inference result, converts it into a format such as "You should buy XX," and sends it to the terminal.

[1386] The terminal notifies the user of the results and prompts them to take specific action.

[1387] The above is a specific embodiment for carrying out the present invention. This system enables advanced decision-making in real time by utilizing the user's physical information and emotional state, thereby increasing the probability of success in earthquake prediction and investment.

[1388] The processing flow will be explained below.

[1389] Step 1:

[1390] Acquiring physical information

[1391] The device collects real-time biometric data, such as the user's heart rate, galvanic skin response, and hormone levels, using wearable and skin-attached sensors.

[1392] Step 2:

[1393] Sending data

[1394] The device transmits the acquired biometric data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[1395] Step 3:

[1396] Receiving data

[1397] The server receives the biometric data sent from the device and stores it in chronological order.

[1398] Step 4:

[1399] Noise Reduction

[1400] The server performs filtering on the received raw data using a moving average filter or other method to remove noise, thereby smoothing the heart rate data.

[1401] Step 5:

[1402] Data normalization

[1403] The server performs a normalization process to scale the denoised data to the range of 0 to 1, thereby maintaining data consistency.

[1404] Step 6:

[1405] Feature Extraction

[1406] The server extracts important features from the normalized data, such as detecting sudden fluctuations in heart rate or peaks in electrodermal response.

[1407] Step 7:

[1408] Emotional state estimation

[1409] The server uses an emotion recognition module to estimate the user's emotional state from preprocessed biometric data, and utilizes an emotion engine to assess the user's stress state and happiness.

[1410] Step 8:

[1411] Input of feature data and emotional state

[1412] The server inputs the extracted feature data and estimated emotional state into an AI model, which uses this information to make predictions for earthquake prediction and investment decision-making.

[1413] Step 9:

[1414] AI model inference

[1415] The server inputs the characteristic data and emotional state into the AI ​​model, and infers the probability of an earthquake occurring and investment decisions such as "buy," "sell," or "hold."

[1416] Step 10:

[1417] Receiving inference results

[1418] The server receives the inference results from the AI ​​model, which are then further corrected rather than used as is.

[1419] Step 11:

[1420] Correcting the results

[1421] The server adjusts the inference results of the AI ​​model based on the emotional state, enabling more accurate decision-making by adjusting the impact of the emotional state on the results.

[1422] Step 12:

[1423] Sending the results

[1424] The server then sends the corrected inference results to the terminal, where they are converted into a format that is easy for the user to understand.

[1425] Step 13:

[1426] User Notification

[1427] The device notifies the user of the corrected inference results by means of voice alerts, vibrations, screen displays, etc., and prompts the user to take specific action.

[1428] The above is a specific flow of the program processing of the present invention, which enables advanced decision-making in real time by utilizing the user's physical information and emotional state.

[1429] Example 2

[1430] 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."

[1431] Conventional earthquake prediction and investment decision-making systems have the problem of low prediction accuracy because they do not take into account the user's emotional state or physical information. Furthermore, it is difficult to collect data and make decisions in real time, making it difficult to prompt users to take immediate action. Furthermore, data preprocessing and feature extraction are insufficient, making it difficult to obtain reliable results.

[1432] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1433] In this invention, the server includes means for acquiring a user's physical information in real time, means for transmitting the acquired physical information to the server, means for removing noise from the physical information received by the server and normalizing the data, means for extracting features from the preprocessed data, means for estimating an emotional state using an emotion recognition module, means for inputting the feature data and the emotional state into an artificial intelligence model to obtain an inference result, means for correcting the inference result based on the emotional state, and means for notifying the user of the corrected inference result. This enables highly accurate predictions that reflect the user's physical information and emotional state in real time, and allows the user to be quickly notified of specific actions.

[1434] "Physical information" refers to biometric data such as a user's heart rate, galvanic skin response, and hormone levels.

[1435] "Real-time" refers to data being collected and processed almost immediately.

[1436] "Server" refers to the computer system that receives, pre-processes, extracts features, estimates emotional states, and performs inference on biometric data.

[1437] "Terminal" refers to a device for collecting and transmitting a user's physical information to a server.

[1438] "Noise reduction" refers to the process of removing unwanted signals and disturbances from raw data.

[1439] "Normalization" refers to the process of scaling data to a particular range.

[1440] "Feature extraction" refers to the process of finding important indicators and patterns from preprocessed data.

[1441] "Emotion Recognition Module" refers to software or algorithms for inferring a user's emotional state based on biometric data.

[1442] An "artificial intelligence model" refers to a model that uses machine learning algorithms to make inferences.

[1443] "Inference results" refer to the predictions and judgments that an artificial intelligence model outputs based on input data.

[1444] "Correction" refers to the process of adjusting the initial inference result by taking into account factors such as emotional state.

[1445] "Notification" refers to the system informing the user of inference results and actions.

[1446] This invention is a system that collects a user's physical information and emotional state in real time, and uses an artificial intelligence (AI) model and emotion engine to make advanced decisions based on that information. In particular, the accuracy of earthquake prediction and investment decision-making can be improved by combining physical sensations, intuition, and emotional state.

[1447] Hardware and software used

[1448] The devices used are wearable or skin-attached sensor devices that can capture biometric data such as heart rate, galvanic skin response, and hormone levels within seconds.

[1449] Specifically, smartwatches and biosensors are used as terminals, and these devices transmit data to a server via wireless communication technologies such as Bluetooth and Wi-Fi.

[1450] On the server, Python's pandas library and sklearn's StandardScaler are used to remove noise and normalize the biometric data. OpenAI's sentiment analysis API is used to estimate emotional states. Furthermore, an AI model built with TensorFlow is used to infer the probability of earthquake occurrence and investment decisions.

[1451] Data processing and calculation

[1452] 1. Biometric data acquisition

[1453] The device captures the user's heart rate, galvanic skin response, and hormone levels every second. Specifically, the smartwatch measures heart rate and galvanic skin response, while sensors under the skin periodically measure hormone levels.

[1454] 2. Data transmission

[1455] The device transmits the acquired biometric data to a smartphone via Bluetooth, and the smartphone then transmits the data to a server using Wi-Fi.

[1456] 3. Data preprocessing and feature extraction

[1457] The server performs noise removal and data normalization on the received raw data. For example, it uses Python's pandas library to smooth the data with a moving average filter and sklearn's StandardScaler to normalize the data to the range of 0 to 1. It then extracts important features, such as rapid fluctuations in heart rate and peaks in electrodermal responses.

[1458] 4. Emotional State Estimation

[1459] The server estimates the user's emotional state from the preprocessed data using an emotion recognition module. For example, the server uses OpenAI's emotion analysis API to evaluate the user's stress level and happiness from the normalized data.

[1460] 5. Input to the AI ​​model

[1461] The server inputs the extracted feature data and estimated emotional state into an AI model to perform inference. For example, the server inputs the data into an earthquake prediction model built using TensorFlow to predict the probability of an earthquake occurring.

[1462] 6. Correction of results

[1463] The server corrects the inference results of the AI ​​model based on the emotional state, adjusting the influence of emotional states such as high stress if they have a strong impact on the results.

[1464] 7. Output and notification of results

[1465] The server converts the final inference results into a user-friendly format and sends them to the device, which notifies the user of the results by means of an alert sound, vibration, or on-screen display.

[1466] Examples and prompts

[1467] Specific examples

[1468] Examples of earthquake prediction:

[1469] The device captures the user's heart rate and galvanic skin response every second and sends them to a server. The server smooths and normalizes the data, detecting sudden fluctuations in heart rate and peaks in galvanic skin response. An emotion engine estimates fear and tension, and inputs the data into an earthquake prediction AI model. The server then classifies the risk of an earthquake occurrence as "high risk" or similar, and notifies the device of the results.

[1470] Examples of investment decisions:

[1471] The device acquires the user's heart rate, galvanic response, and hormone levels, and sends them to a server. The server preprocesses this data and detects heart rate fluctuations, peaks in galvanic response, and sudden spikes in hormone levels. An emotion engine estimates feelings of anxiety and depression, and inputs the data into an investment decision-making AI model. The server then makes a decision to "buy," "sell," or "hold," and notifies the user of the result.

[1472] Prompt Sentence Examples

[1473] Earthquake prediction prompt:

[1474] "Get the user's heart rate and electrodermal response data every second to estimate their emotional state, such as fear or tension. Then, based on this, assess the risk of an earthquake occurring and notify them if the risk is high."

[1475] Investment Decision Prompt:

[1476] "Collect data on the user's heart rate, galvanic skin response, and hormone levels to estimate their emotional state, such as anxiety or depression. Based on that, make investment decisions to buy, sell, or hold, and notify the user of the results."

[1477] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1478] Step 1:

[1479] Biometric data acquisition

[1480] The device collects real-time physical information every second, such as the user's heart rate, galvanic skin response, and hormone levels. Specifically, this data is collected by a smartwatch or sub-skin sensors.

[1481] Input: User's physical information

[1482] Output: Acquired biometric data (heart rate, galvanic skin response, hormone levels)

[1483] Step 2:

[1484] Sending data

[1485] The device transmits the acquired biometric data to a server at regular intervals. Specifically, the data is transferred to a smartphone via Bluetooth, and the smartphone then transmits the data to the server using Wi-Fi.

[1486] Input: Biometric data (heart rate, galvanic skin response, hormone levels)

[1487] Output: Biometric data sent to the server

[1488] Step 3:

[1489] Data Preprocessing

[1490] The server performs noise removal and normalization on the received biometric data. Specifically, it smooths the data with a moving average filter using Python's pandas library and normalizes the data to the range of 0 to 1 using sklearn's StandardScaler.

[1491] Input: Received biometric data

[1492] Output: Denoised normalized data

[1493] Step 4:

[1494] Feature Extraction

[1495] The server extracts important features from the denoised and normalised data, specifically detecting rapid fluctuations in heart rate and peaks in electrodermal response.

[1496] Input: Denoised normalized data

[1497] Output: Extracted feature data (heart rate spikes, electrodermal response peaks)

[1498] Step 5:

[1499] Emotional state estimation

[1500] The server uses an emotion recognition module to estimate the user's emotional state from the preprocessed data. For example, it uses OpenAI's emotion analysis API to evaluate the user's stress level or happiness from the data.

[1501] Input: Extracted feature data

[1502] Output: Estimated emotional state (stress, happiness)

[1503] Step 6:

[1504] Input to the AI ​​model

[1505] The server inputs the extracted feature data and estimated emotional state into a generative AI model to perform inference for earthquake prediction and investment decision-making. Specifically, the server inputs the data into an AI model using TensorFlow.

[1506] Input: extracted feature data, estimated emotional state

[1507] Output: Initial inference results (probability of earthquake occurrence, investment decision)

[1508] Step 7:

[1509] Correcting the results

[1510] The server corrects the initial inference results of the AI ​​model based on the emotional state, adjusting its influence if the emotional state has a strong influence on the results.

[1511] Input: Initial inference result, estimated emotional state

[1512] Output: Corrected inference results

[1513] Step 8:

[1514] Notification of results

[1515] The server converts the corrected inference results into a user-friendly format and sends them to the device, which notifies the user of the results by sound alerts, vibrations, or a display on the screen.

[1516] Input: Corrected inference results

[1517] Output: The result reported to the user

[1518] (Application example 2)

[1519] 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."

[1520] Conventional smart home security systems primarily detect anomalies based on environmental data, without taking into account the user's physical condition or emotional state. As a result, they sometimes fail to detect actual anomalies, or sometimes generate frequent false alarms. This invention aims to improve the accuracy of anomaly detection by combining the user's biometric information and emotional state, providing a more reliable security experience.

[1521] 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.

[1522] In this invention, the server includes means for acquiring biometric data, means for preprocessing the acquired biometric data, means for extracting features from the preprocessed biometric data, means for inputting the extracted features into an artificial intelligence model to obtain an inference result, means for outputting the obtained inference result and notifying the user, means for estimating the emotional state of the user and detecting an abnormality, and means for notifying the user of a specific countermeasure based on the abnormality detection result. This enables an abnormality detection that reflects the user's physical condition and emotional state.

[1523] "Biometric data" is data indicative of a user's physiological state, such as a user's heart rate, galvanic skin response, or hormone levels.

[1524] "Preprocessing" is a process that removes noise from biometric data and prepares the data in a form that is easier to analyze.

[1525] "Feature extraction" is the process of extracting important patterns and trends from pre-processed data.

[1526] An "artificial intelligence model" is an algorithm or system that learns from large amounts of data and makes appropriate inferences about unknown data.

[1527] "Inference results" are predictions or judgments obtained from data processed by an artificial intelligence model.

[1528] "Notification means" refers to a means for notifying the user of inference results or anomaly detection results, particularly via a smartphone or other device.

[1529] "Emotional state" refers to the psychological state such as anxiety, tension, or stress felt by the user.

[1530] "Anomaly detection" is the process of recognizing safety risks and anomalies based on a person's biometric data and emotional state.

[1531] "Countermeasures" are specific actions or procedures that a user should take in response to a detected abnormality.

[1532] This invention is a security service-related system that collects a user's physical information and emotional state in real time and makes advanced decisions based on that data using an artificial intelligence model and emotion engine. In particular, it includes technology to improve the accuracy of anomaly detection in smart home security systems.

[1533] System Configuration

[1534] The system consists of the following main components:

[1535] 1. Wearable devices: collect real-time biometric data such as the user's heart rate, galvanic skin response, and hormone levels.

[1536] 2. Smartphone: A device for transmitting biometric data acquired from wearable devices to a server.

[1537] 3. Server: Includes the following software modules:

[1538] Data preprocessing module: Removes noise and normalizes raw data (software used: Azure Data Factory).

[1539] Feature Extraction Module: Extracts important features from the preprocessed data.

[1540] Emotion recognition module: Estimates emotional state from biometric data (software used: Google Cloud AI).

[1541] AI model: Detects anomalies based on extracted features and emotional states (software used: TensorFlow).

[1542] Result notification module: Notifies the user of the results of anomaly detection via their smartphone (software used: Firebase Cloud Messaging).

[1543] Processing flow

[1544] The server first receives biometric data acquired from the wearable device, denoises and normalizes it using Azure Data Factory, and then extracts features from the preprocessed data, such as sudden fluctuations in heart rate and peaks in electrodermal responses.

[1545] The emotion recognition module uses Google Cloud AI to infer the user's emotional state from pre-processed data and identify states such as anxiety or tension. These features and emotional state are then fed into a TensorFlow-powered AI model for anomaly detection.

[1546] As a result, if an abnormality is detected, a notification will be sent to the user's smartphone via Firebase Cloud Messaging with instructions on how to deal with the problem. The user can then take appropriate action based on the notification they receive.

[1547] Specific examples

[1548] For example, in a "smart home security system equipped with emotion sensors," if a user experiences a sudden rise in heart rate and a peak in their electrodermal response, the emotion recognition module will determine that the user is in a state of high anxiety, and this state will be registered as an abnormality in the system. The AI ​​model will also analyze environmental data (for example, whether the door is open or closed, and the sound level in the room), and if it finally detects an abnormality, it will send a notification to the user's smartphone saying, "The door is not closed. Please check."

[1549] Prompt Sentence Examples

[1550] An example of a prompt is "Design a smart home security system that monitors heart rate and galvanic skin response in real time, and detects and notifies users if anxiety levels are high."

[1551] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1552] Step 1:

[1553] Biometric data acquisition

[1554] The terminal (wearable device) collects real-time data on the user's heart rate, skin galvanic response, and hormone levels. This data is collected from sensors and received by a smartphone. The input is the user's biometric data, and the output is the raw data received by the smartphone. Specifically, data is recorded every second and sent to the smartphone via Bluetooth or other means.

[1555] Step 2:

[1556] Sending data

[1557] The device (smartphone) sends the received biometric data to a server using Wi-Fi or Bluetooth. The server receives this data and supplies it to the next processing step. The input is the acquired raw data, and the output is the data sent to the server. Specifically, the smartphone application packets the data at intervals and sends it.

[1558] Step 3:

[1559] Data Preprocessing

[1560] The raw data sent to the server is denoised and normalised using Azure Data Factory. This purifies the data and makes it easier to analyse. The input is the raw data sent to the server, and the output is the denoised and normalised data. Specific operations include smoothing the heart rate data using a moving average filter and scaling the data from 0 to 1.

[1561] Step 4:

[1562] Feature Extraction

[1563] Important features such as sudden changes in heart rate and peaks in electrodermal responses are extracted from data that has been preprocessed on the server. The input is the preprocessed data, and the output is the extracted features. Specifically, an algorithm is used to analyze the temporal fluctuations of the data and identify sudden changes and peak values.

[1564] Step 5:

[1565] Emotional state estimation

[1566] The extracted feature data is input into Google Cloud AI's emotion recognition module, which estimates the user's emotional state. The input is the extracted feature data, and the output is the estimated emotional state. Specifically, stress and anxiety states are visualized using techniques such as cosine similarity.

[1567] Step 6:

[1568] Input to AI model and anomaly detection

[1569] The estimated emotional state and feature data are input into an AI model (TensorFlow) to perform anomaly detection. The input is the feature data and emotional state, and the output is the anomaly detection result. Specifically, the trained AI model analyzes this data and determines whether there is an anomaly.

[1570] Step 7:

[1571] Result notification

[1572] If an abnormality is detected, the server uses Firebase Cloud Messaging to send a notification to the user's smartphone. The input is the abnormality detection result, and the output is the notification that is displayed on the user's smartphone. Specifically, a message such as "The door is not closed. Please check" is sent.

[1573] Step 8:

[1574] User Action

[1575] The user receives the notification and takes appropriate action based on the specific countermeasures shown. The input is the notification displayed on the smartphone, and the output is the user's action. Specifically, the user checks the door and takes action such as closing it if necessary.

[1576] This enables highly accurate anomaly detection and notification based on the user's biometric information and emotional state.

[1577] 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.

[1578] 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.

[1579] 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.

[1580] [Fourth embodiment]

[1581] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1582] 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.

[1583] 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).

[1584] 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.

[1585] 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.

[1586] 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).

[1587] 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.

[1588] 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.

[1589] 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.

[1590] 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.

[1591] 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.

[1592] 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.

[1593] 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."

[1594] This invention relates to a system that collects a user's physical information in real time and uses an artificial intelligence (AI) model based on that information to make advanced decisions. In particular, accuracy can be improved in earthquake prediction and investment decision-making by utilizing physical sensations and intuition. Specific embodiments for carrying out the present invention are described below.

[1595] System Configuration

[1596] The system consists of the following main components:

[1597] 1. Sensor device (terminal) - Acquires real-time physical information such as the user's heart rate, galvanic skin response, and hormone levels.

[1598] 2. Data transmission module (terminal) - transmits the acquired biometric data to the server.

[1599] 3. Data preprocessing module (server) - Performs preprocessing such as noise removal and normalization on the received raw data.

[1600] 4. Feature Extraction Module (Server) - Extracts important features from the preprocessed data.

[1601] 5. AI model (server) - receives extracted features as input and outputs inference results.

[1602] 6. Result notification module (terminal) - Notifies the user of the inference results.

[1603] Processing flow

[1604] 1. Biometric data acquisition

[1605] The device captures data such as the user's heart rate, galvanic skin response, and hormone levels every second, using wearable devices and sensors attached to the skin.

[1606] 2. Data transmission

[1607] The device transmits the acquired raw data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[1608] 3. Data Preprocessing

[1609] The server performs filtering operations on the received raw data to remove noise, for example, smoothing the heart rate data using a moving average filter, and normalization, scaling the data to a range from 0 to 1.

[1610] 4. Feature Extraction

[1611] The server extracts features from the preprocessed data, such as sudden fluctuations in heart rate and peaks in skin galvanic responses, thereby capturing the user's physical reactions to precursors to earthquakes or sudden market fluctuations.

[1612] 5. Linking to AI models

[1613] The server inputs the extracted feature data into the AI ​​model, which compares it with the data it has learned so far to determine the probability of an earthquake occurring and whether to "buy," "sell," or "hold" in investment decisions.

[1614] 6. Output and notification of results

[1615] The server receives the inference results of the AI ​​model, converts them into a format that is easy for the user to understand, and sends the results to the device, which then notifies the user in various ways, such as by sound alert, vibration, or display on the screen.

[1616] Specific examples

[1617] Examples of earthquake prediction

[1618] 1. Acquisition of physical information

[1619] The device captures the user's heart rate and electrical skin response every second and sends it to a server.

[1620] 2. Data Preprocessing

[1621] The server removes noise from the received data using a moving average filter and normalizes it.

[1622] 3. Feature Extraction

[1623] The server detects rapid fluctuations in heart rate and peaks in electrodermal response.

[1624] 4. Linking to AI models

[1625] The server inputs these features into an earthquake prediction AI model to infer the probability of an earthquake occurring.

[1626] 5. Outputting the results

[1627] The server converts the probability of an earthquake occurring into "high risk," "medium risk," or "low risk," and sends it to the terminal.

[1628] 6. User Notices

[1629] If the risk is high, the device will notify the user, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[1630] Investment Decision Examples

[1631] 1. Acquisition of physical information

[1632] The device measures the user's heart rate, galvanic skin response, and hormone levels every second and transmits them to a server.

[1633] 2. Data Preprocessing

[1634] The server processes this data for noise reduction and normalization.

[1635] 3. Feature Extraction

[1636] The server detects heart rate variability, peaks in galvanic skin response, and spikes in hormone levels.

[1637] 4. Linking to AI models

[1638] The server inputs these characteristics into an investment decision-making AI model and infers the decision to "buy," "sell," or "hold."

[1639] 5. Outputting the results

[1640] The server receives the inference results, converts them into a format such as "You should buy XX," and sends them to the terminal.

[1641] 6. User Notices

[1642] The terminal notifies the user of the results and suggests specific actions to take.

[1643] The above is a specific embodiment for carrying out the present invention. This system enables advanced decision-making using the user's physical information, thereby increasing the probability of success in earthquake prediction and investment.

[1644] The processing flow will be explained below.

[1645] Step 1:

[1646] Acquiring physical information

[1647] The device collects real-time biometric data such as the user's heart rate, galvanic skin response, and hormone levels using wearable and skin-attached sensors.

[1648] Step 2:

[1649] Sending data

[1650] The device transmits the acquired biometric data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[1651] Step 3:

[1652] Receiving data

[1653] The server receives the biometric data sent from the device and stores it in chronological order.

[1654] Step 4:

[1655] Noise Reduction

[1656] The server performs filtering on the received raw data to remove noise, for example, applying a moving average filter to smooth the heart rate data.

[1657] Step 5:

[1658] Data normalization

[1659] The server normalizes the denoised data, scaling the heart rate data to a range of 0 to 1 to make it consistent with other biometric data.

[1660] Step 6:

[1661] Feature Extraction

[1662] The server extracts important features from the normalized data, applying algorithms to detect, for example, sudden fluctuations in heart rate or peaks in electrodermal response.

[1663] Step 7:

[1664] Entering feature data

[1665] The server inputs the extracted feature data into an AI model that has been trained for earthquake prediction and investment decision-making.

[1666] Step 8:

[1667] AI model inference

[1668] The server inputs the characteristic data into the AI ​​model and performs inference to output the probability of an earthquake occurring and investment decisions such as "buy," "sell," or "hold."

[1669] Step 9:

[1670] Receiving inference results

[1671] The server receives the inference results from the AI ​​model and converts them into a format that is easy for users to understand.

[1672] Step 10:

[1673] Sending the results

[1674] The server then sends the converted inference results to the terminal, which then provides the results to the user as an alert message.

[1675] Step 11:

[1676] User Notification

[1677] The device notifies the user of the inference results by means of voice alerts, vibrations, screen displays, etc., and prompts the user to take specific action.

[1678] The above is a specific flow of the program processing of the present invention, which enables advanced decision-making in real time using the user's physical information.

[1679] Example 1

[1680] 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."

[1681] Existing systems have a problem with low accuracy in earthquake prediction and investment decision-making using users' biometric data. This is because the processes of biometric data acquisition, preprocessing, feature extraction, inference, and notification are performed separately, reducing the overall system efficiency. Another problem is that insufficient noise removal and normalization of biometric data can easily produce inaccurate inference results. To solve these problems, an advanced integrated system is required.

[1682] 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.

[1683] In this invention, the server includes means for acquiring biometric data, means for preprocessing the acquired biometric data, means for extracting features from the preprocessed biometric data, means for inputting the extracted features into an artificial intelligence model to obtain an inference result, means for outputting the obtained inference result and notifying the user, means for using a wearable device or a sensor attached to the skin, means for transmitting data using wireless communication technology such as Bluetooth or Wi-Fi, means for smoothing data using a moving average filter, means for normalizing data to a range from 0 to 1, and means for converting the inference result into "high risk," "medium risk," or "low risk." This enables advanced decision-making using a user's biometric data to be performed accurately in real time.

[1684] "Biometric data" is information obtained from a user's body, including heart rate, galvanic skin response, hormone levels, and the like.

[1685] "Preprocessing" refers to applying filtering, normalization, and other processes to acquired biometric data to remove noise and convert the data into a format that is easier to analyze.

[1686] "Feature extraction" is the process of finding important patterns and variations in pre-processed data and preparing it for input into an artificial intelligence model.

[1687] An "artificial intelligence model" is a software system that uses machine learning algorithms to analyze data and draw inferences.

[1688] "Inference results" are the conclusions or judgments that an AI model provides based on biometric data, such as the probability of an earthquake occurring or decisions about whether to "buy," "sell," or "hold" an investment.

[1689] "Notification" is a means of conveying the inference results to the user, and is implemented in the form of an alert sound, vibration, screen display, etc.

[1690] A "wearable device" is a device worn by a user that can acquire biometric data, such as a smartwatch or activity tracker.

[1691] A "sensor" is a device for measuring biometric data, and includes types that are attached to the skin and types that are worn on the arm.

[1692] "Wireless communication technology" refers to technology for sending and receiving data wirelessly, and includes Bluetooth and Wi-Fi.

[1693] A "moving average filter" is a filtering method used to smooth out fluctuations in data, and is a method of removing noise by taking the average over time.

[1694] "Normalization" is a process performed to unify the scale of data, and involves converting the scale of biometric data to a range of 0 to 1.

[1695] "Risk conversion" is the process of converting inference results into categories such as "high risk," "medium risk," and "low risk," and presenting them in a form that is easy for users to understand.

[1696] The system of the present invention is configured to collect biometric data of a user in real time and make advanced decisions using an artificial intelligence (AI) model. Detailed embodiments of the system are described below.

[1697] System Configuration

[1698] This system mainly consists of the following hardware and software components.

[1699] 1. Sensor device (terminal)

[1700] Sensor devices are used to obtain real-time biometric data such as a user's heart rate, electrodermal response, and hormone levels. Specific examples include wearable devices (smartwatches, activity trackers) and sensors attached to the skin. For example, Apple Watch and general skin electrode sensors can be used.

[1701] 2. Data transmission module (terminal)

[1702] The data transmission module is responsible for transmitting the biometric data acquired from the sensor device to the server, using wireless communication technologies such as Bluetooth and Wi-Fi.

[1703] 3. Data preprocessing module (server)

[1704] The data preprocessing module performs noise removal and normalization on the biometric data received on the server, specifically smoothing the data using a moving average filter and scaling the biometric data to a range of 0 to 1.

[1705] 4. Feature extraction module (server)

[1706] The feature extraction module extracts important features from the pre-processed data, such as rapid heart rate fluctuations or peaks in electrodermal response.

[1707] 5. AI model (server)

[1708] This module inputs the extracted feature data into the AI ​​model to obtain inference results. It uses a deep learning algorithm to compare the results with past data and make inferences.

[1709] 6. Result notification module (terminal)

[1710] This module notifies the user of the inference results. Notifications are given in the form of an alert sound, vibration, screen display, etc.

[1711] Specific examples

[1712] Examples of earthquake prediction

[1713] When a user uses a sensor device (e.g., a smartwatch), their heart rate and electrodermal response are acquired every second. This data is sent to a server via Bluetooth. The server smooths the heart rate data using a moving average filter and normalizes the data. The server then detects sudden heart rate fluctuations and peaks in electrodermal response from the preprocessed data. This feature data is input into an AI earthquake prediction model to infer the probability of an earthquake occurring. Finally, the server converts the inference result into "high risk," "medium risk," or "low risk" and sends it to the device. If the risk is high, the device notifies the user, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[1714] Investment Decision Examples

[1715] The user uses the same sensor device to measure their heart rate, galvanic response, and hormone levels every second. This data is sent to a server, which then denoises and normalizes it. Heart rate fluctuations, peaks in galvanic response, and spikes in hormone levels are then detected. This characteristic data is then input into an AI investment decision-making model, which then infers whether to "buy," "sell," or "hold." The server then converts this inference result into a concrete format, such as "You should buy XX," and sends it to the device. The device then notifies the user, "You should buy XX," and suggests a specific action to take.

[1716] Prompt Sentence Examples

[1717] Example prompts to be input to the generative AI model:

[1718] The user's heart rate fluctuates rapidly and their electrodermal response shows a peak. These can be fed into an AI model to assess the risk of an earthquake occurring.

[1719] This system enables advanced decision-making in real time using the user's biometric data, increasing the chances of successful earthquake prediction and investment.

[1720] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1721] Step 1:

[1722] Biometric data acquisition

[1723] Subject: Device

[1724] Description: The device uses sensor devices to collect real-time biometric data such as the user's heart rate, galvanic skin response, and hormone levels. Input data comes from various sensors worn by the user. The device converts this signal into digital data and updates it every second.

[1725] How it works: The device receives signals from the smartwatch and skin electrodes. For example, the optical heart rate monitor in the Apple Watch measures your heart rate beat-by-beat, and the galvanic skin response sensor captures your stress level. This data is temporarily stored on the device.

[1726] Step 2:

[1727] Sending data

[1728] Subject: Device

[1729] Description: The device transmits the acquired biometric data to the server at regular intervals (e.g., every second). The input data is the biometric data acquired from the sensor device, and the output data is the data packet sent to the server.

[1730] How it works: The device sends data to the server via Bluetooth or Wi-Fi, for example using the low-latency, highly reliable Bluetooth Low Energy (BLE) protocol to ensure that the biometric data reaches the server reliably.

[1731] Step 3:

[1732] Data Preprocessing

[1733] Subject: Server

[1734] Description: The server performs preprocessing on the received biometric data, specifically noise removal and normalization. The input data is the biometric data sent from the device, and the output data is the preprocessed data.

[1735] What it does: The server applies a moving average filter to remove spikes and noise from the data. For example, a 3-second moving average filter is used to smooth out fluctuations in heart rate data. At the same time, it normalizes the data by scaling it to a range of 0 to 1, which makes it easier to extract features and input into AI models.

[1736] Step 4:

[1737] Feature Extraction

[1738] Subject: Server

[1739] Description: The server extracts important features from preprocessed data. The input data is the preprocessed data, and the output data is the extracted feature data.

[1740] How it works: The server detects features such as sudden fluctuations in heart rate data and peaks in galvanic response (GDS). It then uses an algorithm to extract, for example, spikes in heart rate or peaks in GDS, thereby identifying moments of stress or excitement felt by the user.

[1741] Step 5:

[1742] Linking to AI models

[1743] Subject: Server

[1744] Description: The server inputs the extracted feature data into the AI ​​model and obtains the inference result. The input data is the feature data, and the output data is the inference result from the AI ​​model.

[1745] How it works: The server uses a trained deep learning model to input feature data into an AI model. The AI ​​model compares this data with past data to determine the probability of an earthquake occurring and whether to "buy," "sell," or "hold" an investment. The model outputs its inference results in real time.

[1746] Step 6:

[1747] Output and notification of results

[1748] Subject: Server

[1749] Description: The server converts the inference results of the AI ​​model into an easy-to-understand format and sends it to the terminal. The input data is the inference results from the AI ​​model, and the output data is the final result that is notified to the user.

[1750] Specific operation: The server classifies the inference results into "high risk," "medium risk," or "low risk" and sends them to the device in JSON format. The device analyzes the received data and notifies the user with a specific message. For example, it warns the user with an alert sound or vibration saying, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[1751] The above is a detailed description of the specific operations and data flow at each processing step of this system. This detailed flow enables advanced decision-making using the user's biometric data in real time.

[1752] (Application example 1)

[1753] 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."

[1754] Conventional content delivery systems primarily recommend content based on users' preferences and interests, but this approach fails to respond to users' real-time emotional states. This results in a poor user experience and reduces their willingness to consume content. Furthermore, the lack of a mechanism for capturing changes in emotional states in real time makes it difficult to provide users with appropriate feedback.

[1755] 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.

[1756] In this invention, the server includes means for acquiring biometric data, means for preprocessing the acquired biometric data, means for extracting features from the preprocessed biometric data, means for inputting the extracted features into an artificial intelligence model to obtain an inference result, means for recommending optimal content to a user based on the obtained inference result, means for notifying the user of the recommended content and monitoring emotional changes in real time as the user views the content, and means for providing feedback based on the monitored emotional changes. This enables personalized content recommendations based on the user's real-time emotional state and the provision of appropriate feedback.

[1757] Below are definitions of important terms included in the claims, tailored to application examples.

[1758] "Biometric data" refers to physical information such as a user's heart rate, galvanic skin response, and hormone levels.

[1759] "Preprocessing" refers to performing processes such as noise removal and normalization on acquired biometric data.

[1760] "Feature extraction" refers to extracting important features, such as rapid heart rate fluctuations or peaks in electrodermal responses, from preprocessed biometric data.

[1761] An "artificial intelligence model" refers to a machine learning algorithm or neural network that makes inferences or decisions based on given input data.

[1762] "Inference result" refers to the prediction or judgment result that an artificial intelligence model outputs based on input data.

[1763] "Content recommendation" refers to the process of suggesting optimal videos, music, and other media to users based on their biometric data.

[1764] "Changes in emotion" refers to fluctuations in emotional state, such as excitement, joy, or sadness, that a user experiences when viewing content.

[1765] "Providing feedback" refers to suggesting appropriate advice or next actions to the user based on monitored emotional changes.

[1766] This invention is a system that uses a user's biometric data to recommend optimal content in real time and provide appropriate feedback. This system consists of the following main components:

[1767] System Configuration

[1768] 1. Means of acquiring biometric data

[1769] Means: A wearable device for measuring the user's heart rate or galvanic skin response (e.g., a heart rate sensor or galvanic skin response sensor).

[1770] Hardware used: General wearable devices (e.g., smartwatches, fitness bands, etc.)

[1771] 2. Pretreatment Methods

[1772] Method: Denoise and normalize the acquired biometric data

[1773] Software used: Python numerical calculation libraries (e.g., numpy, pandas), data preprocessing libraries (e.g., scikit-learn)

[1774] 3. Feature Extraction Method

[1775] Method: Extract features such as heart rate variability and electrodermal response peaks from the preprocessed data.

[1776] Software used: Python machine learning library (e.g., scikit-learn)

[1777] 4. Artificial Intelligence Models

[1778] Method: Take the extracted features as input and infer the user's real-time emotional state.

[1779] Software used: Deep learning libraries (e.g., TensorFlow, Keras)

[1780] Model: Emotional Inference Model

[1781] 5. Content recommendation method based on inference results

[1782] Method: Recommending optimal content to users based on inference results from an AI model

[1783] Software used: Recommendation systems (e.g., content recommendation algorithms)

[1784] 6. User Notification Methods

[1785] Method: Notify the smartphone of the recommended content so that the user can watch it.

[1786] Hardware used: Smartphone

[1787] Software used: Notification library (e.g., Firebase Cloud Messaging)

[1788] 7. Emotion Monitoring Tools

[1789] Method: While the user is watching the content, heart rate and electrodermal response are collected again to monitor changes in emotions in real time.

[1790] Hardware used: General wearable devices

[1791] Software used: Dedicated SDK (e.g., Apple HealthKit, Fitbit SDK)

[1792] 8. Means of Providing Feedback

[1793] Method: Suggest the next content to watch or specific actions based on changes in emotions

[1794] Software used: Recommendation systems (e.g., feedback generation algorithms)

[1795] Specific examples

[1796] For example, when a user wears a smartwatch and watches a video, the smartwatch measures the user's heart rate and electrodermal response and sends the data to a smartphone. The smartphone then transmits this biometric data to a server at regular intervals. The server then preprocesses the received data and extracts features. An AI model uses this feature data to infer the user's emotional state and recommends the next optimal piece of content based on the inference. The smartphone then notifies the user of the content and provides appropriate feedback by monitoring emotional changes in real time while watching.

[1797] Prompt Sentence Examples

[1798] "Please infer the user's real-time emotional state from their heart rate and electrodermal response based on the content they are currently viewing, and recommend the best content for them to watch next."

[1799] This allows users to enjoy content that is always tailored to their emotional state, resulting in a system that provides an even more entertaining experience.

[1800] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1801] Step 1:

[1802] The user puts on the wearable device.

[1803] The device measures the user's heart rate and electrical skin response in real time, and transmits this data to a smartphone via Bluetooth or other means.

[1804] Input: User's heart rate, galvanic skin response

[1805] Data processing: Acquired from wearable devices

[1806] Output: Biometric data is sent to a smartphone

[1807] Step 2:

[1808] The smartphone receives biometric data from the wearable device and sends it to a server at regular intervals, using Wi-Fi or mobile data for communication.

[1809] Input: Biometric data (heart rate, electrodermal response)

[1810] Data processing: Buffering data at regular intervals

[1811] Output: Biometric data is sent to the server

[1812] Step 3:

[1813] The server performs noise reduction and normalization on the received biometric data, for example, smoothing the heart rate data using a moving average filter.

[1814] Input: Biometric data (heart rate, electrodermal response)

[1815] Data calculation: Noise removal (moving average filter), normalization

[1816] Output: Preprocessed biometric data

[1817] Step 4:

[1818] The server extracts features such as rapid fluctuations in heart rate and peaks in electrodermal response from the preprocessed biometric data.

[1819] Input: Preprocessed biometric data

[1820] Data calculation: feature extraction (rapid fluctuations, peak detection)

[1821] Output: feature data

[1822] Step 5:

[1823] The server inputs the extracted feature data into an artificial intelligence model to infer the user's emotional state.

[1824] Input: feature data

[1825] Data Computation: Emotional Inference with AI Models

[1826] Output: Inference result (emotional state)

[1827] Step 6:

[1828] The server uses a feedback and recommendation algorithm to recommend the most suitable content to the user based on the inference results.

[1829] Input: Inference result (emotional state)

[1830] Data processing: Content selection by recommendation algorithm

[1831] Output: Recommended content

[1832] Step 7:

[1833] The smartphone notifies the user of the recommended content sent from the server using services such as Firebase Cloud Messaging.

[1834] Input: Recommended content

[1835] Data operation: Notification preparation (message generation)

[1836] Output: Content recommendation notification

[1837] Step 8:

[1838] While the user is watching the content, the smartphone again collects biometric data from the wearable device and monitors emotional changes in real time.

[1839] Input: Biometric data during viewing (heart rate, skin galvanic response)

[1840] Data calculation: Real-time monitoring

[1841] Output: User's emotional change data

[1842] Step 9:

[1843] The server uses the emotion change data to provide the next feedback and recommend content, and generates appropriate feedback using an algorithm.

[1844] Input: Emotion change data

[1845] Data calculation: feedback generation, next content recommendation

[1846] Output: Feedback to the user, next content recommendation

[1847] The above are the specific processing steps for implementing the present invention. This system allows users to enjoy content that is always tailored to their emotional state, improving their entertainment experience.

[1848] 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.

[1849] This invention relates to a system that collects a user's physical information and emotional state in real time and uses an artificial intelligence (AI) model and an emotion engine to make advanced decisions based on that information. In particular, accuracy can be improved in earthquake prediction and investment decision-making by combining physical sensations, intuition, and emotional state. Specific embodiments for implementing the present invention are described below.

[1850] System Configuration

[1851] The system consists of the following main components:

[1852] 1. Sensor device (terminal) - Acquires real-time physical information such as the user's heart rate, galvanic skin response, and hormone levels.

[1853] 2. Data transmission module (terminal) - transmits the acquired biometric data to the server.

[1854] 3. Data preprocessing module (server) - Performs preprocessing such as noise removal and normalization on the received raw data.

[1855] 4. Feature Extraction Module (Server) - Extracts important features from the preprocessed data.

[1856] 5. Emotion Recognition Module (Server) - Uses an emotion engine to infer the user's emotional state from biometric data.

[1857] 6. AI model (server) - receives extracted features and estimated emotional states as input and outputs inference results.

[1858] 7. Result correction module (server) - Corrects the inference results based on the emotional state.

[1859] 8. Result notification module (terminal) - Notifies the user of the inference results.

[1860] Processing flow

[1861] 1. Biometric data acquisition

[1862] The device captures second-by-second physical information such as the user's heart rate, galvanic skin response, and hormone levels using wearable devices and sensors attached to the skin.

[1863] 2. Data transmission

[1864] The device transmits the acquired raw data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[1865] 3. Data preprocessing and feature extraction

[1866] The server performs filtering operations on the received raw data to remove noise, for example, smoothing the heart rate data using a moving average filter, and normalization, scaling the data to a range from 0 to 1.

[1867] Extracting important features from normalized data, for example detecting sudden fluctuations in heart rate or peaks in electrodermal response.

[1868] 4. Emotional State Estimation

[1869] The server uses an emotion recognition module to estimate the user's emotional state from the preprocessed biometric data, for example, by using cosine similarity and emotion models to assess the user's stress state and happiness.

[1870] 5. Input to the AI ​​model

[1871] The server inputs the extracted feature data and estimated emotional state into an AI model, which then uses this information to make predictions for earthquake prediction and investment decision-making.

[1872] 6. Correction of results

[1873] The server adjusts the inference results of the AI ​​model based on the emotional state, enabling more accurate decision-making by adjusting the impact of the emotional state on the results.

[1874] 7. Output and notification of results

[1875] The server converts the final inference results into a format that is easy for the user to understand, and then sends the results to the device, which then notifies the user in various ways, such as by sound alert, vibration, or display on the screen.

[1876] Specific examples

[1877] Examples of earthquake prediction

[1878] 1. Acquisition of physical information

[1879] The device captures the user's heart rate and electrical skin response every second and sends it to a server.

[1880] 2. Data preprocessing and feature extraction

[1881] The server removes noise from the received data using a moving average filter and normalizes it.

[1882] Detects rapid fluctuations in heart rate and peaks in electrodermal response.

[1883] 3. Emotional State Estimation

[1884] The server uses an emotion engine to infer the user's emotional state (e.g., fear or tension) from the pre-processed data.

[1885] 4. Input to the AI ​​model

[1886] The server inputs the extracted feature data and emotional state into an earthquake prediction AI model to infer the probability of an earthquake occurring.

[1887] 5. Correction of results

[1888] The server will take into account the user's emotional state to correct the inference results, for example by weighting and adjusting the results if the emotional state has a strong influence on the predictions.

[1889] 6. Output and notification of results

[1890] The server converts the probability of an earthquake occurring into "high risk," "medium risk," or "low risk," and sends it to the terminal.

[1891] If the risk is high, the device will notify the user, "There is a high possibility of an earthquake occurring. Please prepare to evacuate."

[1892] Investment Decision Examples

[1893] 1. Acquisition of physical information

[1894] The device measures the user's heart rate, galvanic skin response, and hormone levels every second and transmits them to a server.

[1895] 2. Data preprocessing and feature extraction

[1896] The server processes this data for noise reduction and normalization.

[1897] Detects heart rate variability, peaks in electrodermal response, and surges in hormone levels.

[1898] 3. Emotional State Estimation

[1899] The server uses an emotion engine to estimate the user's emotional state (e.g., anxiety or depression) from the preprocessed data.

[1900] 4. Input to the AI ​​model

[1901] The server inputs these characteristics and emotional states into an investment decision-making AI model to infer decisions on whether to buy, sell, or hold.

[1902] 5. Correction of results

[1903] The server corrects the inference results based on the user's emotional state in order to minimize the impact of the user's emotional state on investment decisions.

[1904] 6. Output and notification of results

[1905] The server receives the inference result, converts it into a format such as "You should buy XX," and sends it to the terminal.

[1906] The terminal notifies the user of the results and prompts them to take specific action.

[1907] The above is a specific embodiment for carrying out the present invention. This system enables advanced decision-making in real time by utilizing the user's physical information and emotional state, thereby increasing the probability of success in earthquake prediction and investment.

[1908] The processing flow will be explained below.

[1909] Step 1:

[1910] Acquiring physical information

[1911] The device collects real-time biometric data, such as the user's heart rate, galvanic skin response, and hormone levels, using wearable and skin-attached sensors.

[1912] Step 2:

[1913] Sending data

[1914] The device transmits the acquired biometric data to a server at regular intervals using wireless communication technologies such as Wi-Fi and Bluetooth.

[1915] Step 3:

[1916] Receiving data

[1917] The server receives the biometric data sent from the device and stores it in chronological order.

[1918] Step 4:

[1919] Noise Reduction

[1920] The server performs filtering on the received raw data using a moving average filter or other method to remove noise, thereby smoothing the heart rate data.

[1921] Step 5:

[1922] Data normalization

[1923] The server performs a normalization process to scale the denoised data to the range of 0 to 1, thereby maintaining data consistency.

[1924] Step 6:

[1925] Feature Extraction

[1926] The server extracts important features from the normalized data, such as detecting sudden fluctuations in heart rate or peaks in electrodermal response.

[1927] Step 7:

[1928] Emotional state estimation

[1929] The server uses an emotion recognition module to estimate the user's emotional state from preprocessed biometric data, and utilizes an emotion engine to assess the user's stress state and happiness.

[1930] Step 8:

[1931] Input of feature data and emotional state

[1932] The server inputs the extracted feature data and estimated emotional state into an AI model, which uses this information to make predictions for earthquake prediction and investment decision-making.

[1933] Step 9:

[1934] AI model inference

[1935] The server inputs the characteristic data and emotional state into the AI ​​model, and infers the probability of an earthquake occurring and investment decisions such as "buy," "sell," or "hold."

[1936] Step 10:

[1937] Receiving inference results

[1938] The server receives the inference results from the AI ​​model, which are then further corrected rather than used as is.

[1939] Step 11:

[1940] Correcting the results

[1941] The server adjusts the inference results of the AI ​​model based on the emotional state, enabling more accurate decision-making by adjusting the impact of the emotional state on the results.

[1942] Step 12:

[1943] Sending the results

[1944] The server then sends the corrected inference results to the terminal, where they are converted into a format that is easy for the user to understand.

[1945] Step 13:

[1946] User Notification

[1947] The device notifies the user of the corrected inference results by means of voice alerts, vibrations, screen displays, etc., and prompts the user to take specific action.

[1948] The above is a specific flow of the program processing of the present invention, which enables advanced decision-making in real time by utilizing the user's physical information and emotional state.

[1949] Example 2

[1950] 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."

[1951] Conventional earthquake prediction and investment decision-making systems have the problem of low prediction accuracy because they do not take into account the user's emotional state or physical information. Furthermore, it is difficult to collect data and make decisions in real time, making it difficult to prompt users to take immediate action. Furthermore, data preprocessing and feature extraction are insufficient, making it difficult to obtain reliable results.

[1952] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1953] In this invention, the server includes means for acquiring a user's physical information in real time, means for transmitting the acquired physical information to the server, means for removing noise from the physical information received by the server and normalizing the data, means for extracting features from the preprocessed data, means for estimating an emotional state using an emotion recognition module, means for inputting the feature data and the emotional state into an artificial intelligence model to obtain an inference result, means for correcting the inference result based on the emotional state, and means for notifying the user of the corrected inference result. This enables highly accurate predictions that reflect the user's physical information and emotional state in real time, and allows the user to be quickly notified of specific actions.

[1954] "Physical information" refers to biometric data such as a user's heart rate, galvanic skin response, and hormone levels.

[1955] "Real-time" refers to data being collected and processed almost immediately.

[1956] "Server" refers to the computer system that receives, pre-processes, extracts features, estimates emotional states, and performs inference on biometric data.

[1957] "Terminal" refers to a device for collecting and transmitting a user's physical information to a server.

[1958] "Noise reduction" refers to the process of removing unwanted signals and disturbances from raw data.

[1959] "Normalization" refers to the process of scaling data to a particular range.

[1960] "Feature extraction" refers to the process of finding important indicators and patterns from preprocessed data.

[1961] "Emotion Recognition Module" refers to software or algorithms for inferring a user's emotional state based on biometric data.

[1962] An "artificial intelligence model" refers to a model that uses machine learning algorithms to make inferences.

[1963] "Inference results" refer to the predictions and judgments that an artificial intelligence model outputs based on input data.

[1964] "Correction" refers to the process of adjusting the initial inference result by taking into account factors such as emotional state.

[1965] "Notification" refers to the system informing the user of inference results and actions.

[1966] This invention is a system that collects a user's physical information and emotional state in real time, and uses an artificial intelligence (AI) model and emotion engine to make advanced decisions based on that information. In particular, the accuracy of earthquake prediction and investment decision-making can be improved by combining physical sensations, intuition, and emotional state.

[1967] Hardware and software used

[1968] The devices used are wearable or skin-attached sensor devices that can capture biometric data such as heart rate, galvanic skin response, and hormone levels within seconds.

[1969] Specifically, smartwatches and biosensors are used as terminals, and these devices transmit data to a server via wireless communication technologies such as Bluetooth and Wi-Fi.

[1970] On the server, Python's pandas library and sklearn's StandardScaler are used to remove noise and normalize the biometric data. OpenAI's sentiment analysis API is used to estimate emotional states. Furthermore, an AI model built with TensorFlow is used to infer the probability of earthquake occurrence and investment decisions.

[1971] Data processing and calculation

[1972] 1. Biometric data acquisition

[1973] The device captures the user's heart rate, galvanic skin response, and hormone levels every second. Specifically, the smartwatch measures heart rate and galvanic skin response, while sensors under the skin periodically measure hormone levels.

[1974] 2. Data transmission

[1975] The device transmits the acquired biometric data to a smartphone via Bluetooth, and the smartphone then transmits the data to a server using Wi-Fi.

[1976] 3. Data preprocessing and feature extraction

[1977] The server performs noise removal and data normalization on the received raw data. For example, it uses Python's pandas library to smooth the data with a moving average filter and sklearn's StandardScaler to normalize the data to the range of 0 to 1. It then extracts important features, such as rapid fluctuations in heart rate and peaks in electrodermal responses.

[1978] 4. Emotional State Estimation

[1979] The server estimates the user's emotional state from the preprocessed data using an emotion recognition module. For example, the server uses OpenAI's emotion analysis API to evaluate the user's stress level and happiness from the normalized data.

[1980] 5. Input to the AI ​​model

[1981] The server inputs the extracted feature data and estimated emotional state into an AI model to perform inference. For example, the server inputs the data into an earthquake prediction model built using TensorFlow to predict the probability of an earthquake occurring.

[1982] 6. Correction of results

[1983] The server corrects the inference results of the AI ​​model based on the emotional state, adjusting the influence of emotional states such as high stress if they have a strong impact on the results.

[1984] 7. Output and notification of results

[1985] The server converts the final inference results into a user-friendly format and sends them to the device, which notifies the user of the results by means of an alert sound, vibration, or on-screen display.

[1986] Examples and prompts

[1987] Specific examples

[1988] Examples of earthquake prediction:

[1989] The device captures the user's heart rate and galvanic skin response every second and sends them to a server. The server smooths and normalizes the data, detecting sudden fluctuations in heart rate and peaks in galvanic skin response. An emotion engine estimates fear and tension, and inputs the data into an earthquake prediction AI model. The server then classifies the risk of an earthquake occurrence as "high risk" or similar, and notifies the device of the results.

[1990] Examples of investment decisions:

[1991] The device acquires the user's heart rate, galvanic response, and hormone levels, and sends them to a server. The server preprocesses this data and detects heart rate fluctuations, peaks in galvanic response, and sudden spikes in hormone levels. An emotion engine estimates feelings of anxiety and depression, and inputs the data into an investment decision-making AI model. The server then makes a decision to "buy," "sell," or "hold," and notifies the user of the result.

[1992] Prompt Sentence Examples

[1993] Earthquake prediction prompt:

[1994] "Get the user's heart rate and electrodermal response data every second to estimate their emotional state, such as fear or tension. Then, based on this, assess the risk of an earthquake occurring and notify them if the risk is high."

[1995] Investment Decision Prompt:

[1996] "Collect data on the user's heart rate, galvanic skin response, and hormone levels to estimate their emotional state, such as anxiety or depression. Based on that, make investment decisions to buy, sell, or hold, and notify the user of the results."

[1997] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1998] Step 1:

[1999] Biometric data acquisition

[2000] The device collects real-time physical information every second, such as the user's heart rate, galvanic skin response, and hormone levels. Specifically, this data is collected by a smartwatch or sub-skin sensors.

[2001] Input: User's physical information

[2002] Output: Acquired biometric data (heart rate, galvanic skin response, hormone levels)

[2003] Step 2:

[2004] Sending data

[2005] The device transmits the acquired biometric data to a server at regular intervals. Specifically, the data is transferred to a smartphone via Bluetooth, and the smartphone then transmits the data to the server using Wi-Fi.

[2006] Input: Biometric data (heart rate, galvanic skin response, hormone levels)

[2007] Output: Biometric data sent to the server

[2008] Step 3:

[2009] Data Preprocessing

[2010] The server performs noise removal and normalization on the received biometric data. Specifically, it smooths the data with a moving average filter using Python's pandas library and normalizes the data to the range of 0 to 1 using sklearn's StandardScaler.

[2011] Input: Received biometric data

[2012] Output: Denoised normalized data

[2013] Step 4:

[2014] Feature Extraction

[2015] The server extracts important features from the denoised and normalised data, specifically detecting rapid fluctuations in heart rate and peaks in electrodermal response.

[2016] Input: Denoised normalized data

[2017] Output: Extracted feature data (heart rate spikes, electrodermal response peaks)

[2018] Step 5:

[2019] Emotional state estimation

[2020] The server uses an emotion recognition module to estimate the user's emotional state from the preprocessed data. For example, it uses OpenAI's emotion analysis API to evaluate the user's stress level or happiness from the data.

[2021] Input: Extracted feature data

[2022] Output: Estimated emotional state (stress, happiness)

[2023] Step 6:

[2024] Input to the AI ​​model

[2025] The server inputs the extracted feature data and estimated emotional state into a generative AI model to perform inference for earthquake prediction and investment decision-making. Specifically, the server inputs the data into an AI model using TensorFlow.

[2026] Input: extracted feature data, estimated emotional state

[2027] Output: Initial inference results (probability of earthquake occurrence, investment decision)

[2028] Step 7:

[2029] Correcting the results

[2030] The server corrects the initial inference results of the AI ​​model based on the emotional state, adjusting its influence if the emotional state has a strong influence on the results.

[2031] Input: Initial inference result, estimated emotional state

[2032] Output: Corrected inference results

[2033] Step 8:

[2034] Notification of results

[2035] The server converts the corrected inference results into a user-friendly format and sends them to the device, which notifies the user of the results by sound alerts, vibrations, or a display on the screen.

[2036] Input: Corrected inference results

[2037] Output: The result reported to the user

[2038] (Application example 2)

[2039] 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."

[2040] Conventional smart home security systems primarily detect anomalies based on environmental data, without taking into account the user's physical condition or emotional state. As a result, they sometimes fail to detect actual anomalies, or sometimes generate frequent false alarms. This invention aims to improve the accuracy of anomaly detection by combining the user's biometric information and emotional state, providing a more reliable security experience.

[2041] 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.

[2042] In this invention, the server includes means for acquiring biometric data, means for preprocessing the acquired biometric data, means for extracting features from the preprocessed biometric data, means for inputting the extracted features into an artificial intelligence model to obtain an inference result, means for outputting the obtained inference result and notifying the user, means for estimating the emotional state of the user and detecting an abnormality, and means for notifying the user of a specific countermeasure based on the abnormality detection result. This enables an abnormality detection that reflects the user's physical condition and emotional state.

[2043] "Biometric data" is data indicative of a user's physiological state, such as a user's heart rate, galvanic skin response, or hormone levels.

[2044] "Preprocessing" is a process that removes noise from biometric data and prepares the data in a form that is easier to analyze.

[2045] "Feature extraction" is the process of extracting important patterns and trends from pre-processed data.

[2046] An "artificial intelligence model" is an algorithm or system that learns from large amounts of data and makes appropriate inferences about unknown data.

[2047] "Inference results" are predictions or judgments obtained from data processed by an artificial intelligence model.

[2048] "Notification means" refers to a means for notifying the user of inference results or anomaly detection results, particularly via a smartphone or other device.

[2049] "Emotional state" refers to the psychological state such as anxiety, tension, or stress felt by the user.

[2050] "Anomaly detection" is the process of recognizing safety risks and anomalies based on a person's biometric data and emotional state.

[2051] "Countermeasures" are specific actions or procedures that a user should take in response to a detected abnormality.

[2052] This invention is a security service-related system that collects a user's physical information and emotional state in real time and makes advanced decisions based on that data using an artificial intelligence model and emotion engine. In particular, it includes technology to improve the accuracy of anomaly detection in smart home security systems.

[2053] System Configuration

[2054] The system consists of the following main components:

[2055] 1. Wearable devices: collect real-time biometric data such as the user's heart rate, galvanic skin response, and hormone levels.

[2056] 2. Smartphone: A device for transmitting biometric data acquired from wearable devices to a server.

[2057] 3. Server: Includes the following software modules:

[2058] Data preprocessing module: Removes noise and normalizes raw data (software used: Azure Data Factory).

[2059] Feature Extraction Module: Extracts important features from the preprocessed data.

[2060] Emotion recognition module: Estimates emotional state from biometric data (software used: Google Cloud AI).

[2061] AI model: Detects anomalies based on extracted features and emotional states (software used: TensorFlow).

[2062] Result notification module: Notifies the user of the results of anomaly detection via their smartphone (software used: Firebase Cloud Messaging).

[2063] Processing flow

[2064] The server first receives biometric data acquired from the wearable device, denoises and normalizes it using Azure Data Factory, and then extracts features from the preprocessed data, such as sudden fluctuations in heart rate and peaks in electrodermal responses.

[2065] The emotion recognition module uses Google Cloud AI to infer the user's emotional state from pre-processed data and identify states such as anxiety or tension. These features and emotional state are then fed into a TensorFlow-powered AI model for anomaly detection.

[2066] As a result, if an abnormality is detected, a notification will be sent to the user's smartphone via Firebase Cloud Messaging with instructions on how to deal with the problem. The user can then take appropriate action based on the notification they receive.

[2067] Specific examples

[2068] For example, in a "smart home security system equipped with emotion sensors," if a user experiences a sudden rise in heart rate and a peak in their electrodermal response, the emotion recognition module will determine that the user is in a state of high anxiety, and this state will be registered as an abnormality in the system. The AI ​​model will also analyze environmental data (for example, whether the door is open or closed, and the sound level in the room), and if it finally detects an abnormality, it will send a notification to the user's smartphone saying, "The door is not closed. Please check."

[2069] Prompt Sentence Examples

[2070] An example of a prompt is "Design a smart home security system that monitors heart rate and galvanic skin response in real time, and detects and notifies users if anxiety levels are high."

[2071] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[2072] Step 1:

[2073] Biometric data acquisition

[2074] The terminal (wearable device) collects real-time data on the user's heart rate, skin galvanic response, and hormone levels. This data is collected from sensors and received by a smartphone. The input is the user's biometric data, and the output is the raw data received by the smartphone. Specifically, data is recorded every second and sent to the smartphone via Bluetooth or other means.

[2075] Step 2:

[2076] Sending data

[2077] The device (smartphone) sends the received biometric data to a server using Wi-Fi or Bluetooth. The server receives this data and supplies it to the next processing step. The input is the acquired raw data, and the output is the data sent to the server. Specifically, the smartphone application packets the data at intervals and sends it.

[2078] Step 3:

[2079] Data Preprocessing

[2080] The raw data sent to the server is denoised and normalised using Azure Data Factory. This purifies the data and makes it easier to analyse. The input is the raw data sent to the server, and the output is the denoised and normalised data. Specific operations include smoothing the heart rate data using a moving average filter and scaling the data from 0 to 1.

[2081] Step 4:

[2082] Feature Extraction

[2083] Important features such as sudden changes in heart rate and peaks in electrodermal responses are extracted from data that has been preprocessed on the server. The input is the preprocessed data, and the output is the extracted features. Specifically, an algorithm is used to analyze the temporal fluctuations of the data and identify sudden changes and peak values.

[2084] Step 5:

[2085] Emotional state estimation

[2086] The extracted feature data is input into Google Cloud AI's emotion recognition module, which estimates the user's emotional state. The input is the extracted feature data, and the output is the estimated emotional state. Specifically, stress and anxiety states are visualized using techniques such as cosine similarity.

[2087] Step 6:

[2088] Input to AI model and anomaly detection

[2089] The estimated emotional state and feature data are input into an AI model (TensorFlow) to perform anomaly detection. The input is the feature data and emotional state, and the output is the anomaly detection result. Specifically, the trained AI model analyzes this data and determines whether there is an anomaly.

[2090] Step 7:

[2091] Result notification

[2092] If an abnormality is detected, the server uses Firebase Cloud Messaging to send a notification to the user's smartphone. The input is the abnormality detection result, and the output is the notification that is displayed on the user's smartphone. Specifically, a message such as "The door is not closed. Please check" is sent.

[2093] Step 8:

[2094] User Action

[2095] The user receives the notification and takes appropriate action based on the specific countermeasures shown. The input is the notification displayed on the smartphone, and the output is the user's action. Specifically, the user checks the door and takes action such as closing it if necessary.

[2096] This enables highly accurate anomaly detection and notification based on the user's biometric information and emotional state.

[2097] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[2098] 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.

[2099] 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 robot 414.

[2100] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[2101] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[2102] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[2103] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[2104] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[2105] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[2106] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[2107] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[2108] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[2109] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[2110] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[2111] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[2112] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[2113] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[2114] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[2115] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[2116] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[2117] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[2118] The following is further disclosed regarding the above embodiment.

[2119] (Claim 1)

[2120] a means for acquiring biometric data;

[2121] means for pre-processing the acquired biometric data;

[2122] means for extracting features from the preprocessed biometric data;

[2123] A means for inputting the extracted features into an artificial intelligence model to obtain an inference result;

[2124] a means for outputting the obtained inference result and notifying the user;

[2125] A system including:

[2126] (Claim 2)

[2127] 10. The system of claim 1, further comprising means for denoising and normalising the biometric data.

[2128] (Claim 3)

[2129] 10. The system of claim 1, further comprising means for extracting heart rate rapid fluctuations and electrodermal response peaks from the preprocessed biometric data.

[2130] "Example 1"

[2131] (Claim 1)

[2132] a means for acquiring biometric data;

[2133] means for pre-processing the acquired biometric data;

[2134] means for extracting features from the preprocessed biometric data;

[2135] A means for inputting the extracted features into an artificial intelligence model to obtain an inference result;

[2136] a means for outputting the obtained inference result and notifying the user;

[2137] Using wearable devices or sensors attached to the skin,

[2138] A means for transmitting data using wireless communication technologies such as Bluetooth and Wi-Fi;

[2139] means for smoothing the data using a moving average filter;

[2140] a means of normalizing the data to the range 0 to 1;

[2141] A means for converting the inference results into "high risk," "medium risk," and "low risk";

[2142] A system including:

[2143] (Claim 2)

[2144] 10. The system of claim 1, further comprising means for denoising and normalising the biometric data.

[2145] (Claim 3)

[2146] 10. The system of claim 1, further comprising means for extracting heart rate rapid fluctuations and electrodermal response peaks from the preprocessed biometric data.

[2147] "Application Example 1"

[2148] (Claim 1)

[2149] a means for acquiring biometric data;

[2150] means for pre-processing the acquired biometric data;

[2151] means for extracting features from the preprocessed biometric data;

[2152] A means for inputting the extracted features into an artificial intelligence model to obtain an inference result;

[2153] A means for recommending optimal content to users based on the inference results obtained;

[2154] means for notifying the user of the recommended content and monitoring the user's emotional changes in real time as the user watches the content;

[2155] a means of providing feedback based on monitored changes in emotion;

[2156] A system including:

[2157] (Claim 2)

[2158] 10. The system of claim 1, further comprising means for denoising and normalising the biometric data.

[2159] (Claim 3)

[2160] 10. The system of claim 1, further comprising means for extracting heart rate rapid fluctuations and electrodermal response peaks from the preprocessed biometric data.

[2161] "Example 2: Combining Emotion Engines"

[2162] (Claim 1)

[2163] A means for acquiring physical information of a user in real time;

[2164] means for transmitting the acquired physical information to a server;

[2165] a means for denoising the received body information and normalising the data at the server;

[2166] means for extracting features from the preprocessed data;

[2167] means for estimating an emotional state using an emotion recognition module;

[2168] A means for inputting the feature data and the emotional state into an artificial intelligence model and obtaining an inference result;

[2169] means for correcting the inference result based on the emotional state;

[2170] a means for notifying a user of the corrected inference result;

[2171] A system including:

[2172] (Claim 2)

[2173] 10. The system of claim 1, further comprising means for extracting abrupt fluctuations in biometric data or peaks in electrodermal responses.

[2174] (Claim 3)

[2175] 10. The system of claim 1, further comprising means for transmitting data from the terminal to the server using wireless communication technology.

[2176] "Application example 2 when combining emotion engines"

[2177] (Claim 1)

[2178] a means for acquiring biometric data;

[2179] means for pre-processing the acquired biometric data;

[2180] means for extracting features from the preprocessed biometric data;

[2181] A means for inputting the extracted features into an artificial intelligence model to obtain an inference result;

[2182] a means for outputting the obtained inference result and notifying the user;

[2183] means for estimating a user's emotional state and detecting anomalies;

[2184] A means for notifying a user of a specific countermeasure based on the anomaly detection result;

[2185] A system including:

[2186] (Claim 2)

[2187] 10. The system of claim 1, further comprising means for denoising and normalising the biometric data.

[2188] (Claim 3)

[2189] 10. The system of claim 1, further comprising means for extracting heart rate rapid fluctuations and electrodermal response peaks from the preprocessed biometric data. [Explanation of symbols]

[2190] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. a means for acquiring biometric data; means for pre-processing the acquired biometric data; means for extracting features from the preprocessed biometric data; A means for inputting the extracted features into an artificial intelligence model to obtain an inference result; a means for outputting the obtained inference result and notifying the user; A system including:

2. 10. The system of claim 1, further comprising means for denoising and normalising the biometric data.

3. 10. The system of claim 1, further comprising means for extracting heart rate rapid fluctuations and electrodermal response peaks from the preprocessed biometric data.

Images