system

Abstract

The system according to the present embodiment includes a reception unit, an analysis unit, a proposal unit, and a simulation unit. The reception unit is configured to input a user's situation. The analysis unit is configured to analyze information input by the reception unit. The proposal unit is configured to propose a target issue based on the information analyzed by the analysis unit. The simulation unit is configured to simulate a state after recovery based on the target issue proposed by the proposal unit.

Description

Technical Field

[0001] The technology disclosed herein relates to a system. Background Technology

[0002] Patent Document 1 discloses a personalized chatbot control method executed by at least one processor, the method comprising: receiving user speech; adding the user speech to a prompt containing instructions related to a chatbot role; encoding the prompt; and inputting the encoded prompt into a language model to generate chatbot speech in response to the user speech.

[0003] Patent document 1: Japanese Patent Application Publication No. 2022-180282.

[0004] In existing technologies, there is a problem that patients who are hospitalized for a long time or in rehabilitation are prone to losing their goals and have difficulty finding direction after rehabilitation. Summary of the Invention

[0005] The system described in this embodiment includes a receiving unit, a parsing unit, a proposal unit, and a simulation unit. The receiving unit is used to input the user's status. The parsing unit is used to parse the information input by the receiving unit. The proposal unit proposes a target problem based on the information parsed by the parsing unit. The simulation unit simulates the restored state based on the target problem proposed by the proposal unit. Attached Figure Description

[0006] Figure 1 This is a conceptual diagram illustrating an example of the configuration of a data processing system according to the first embodiment.

[0007] Figure 2 This is a conceptual diagram illustrating an example of the main functions of the data processing apparatus and smart device according to the first embodiment.

[0008] Figure 3 This is a conceptual diagram illustrating an example of the data processing system configuration in the second embodiment.

[0009] Figure 4 This is a conceptual diagram illustrating an example of the main functions of the data processing device and smart glasses according to the second embodiment.

[0010] Figure 5 This is a conceptual diagram illustrating an example of the data processing system configuration in the third embodiment.

[0011] Figure 6 This is a conceptual diagram illustrating an example of the main functions of the data processing device and head-mounted terminal according to the third embodiment.

[0012] Figure 7 This is a conceptual diagram illustrating an example of the data processing system configuration in the fourth embodiment.

[0013] Figure 8 This is a conceptual diagram illustrating an example of the functions of the main parts of the data processing device and robot according to the fourth embodiment.

[0014] Figure 9 It represents an emotion graph that maps multiple emotions.

[0015] Figure 10 It represents an emotion graph that maps multiple emotions.

[0016] Explanation of reference numerals in the attached figures Data processing systems 10, 210, 310, and 410 12 Data processing device 14 Smart devices 214 Smart Glasses 314 Head-mounted terminal 414 Robot. Detailed Implementation

[0017] Hereinafter, an example of an implementation of the system involved in this disclosure will be described with reference to the accompanying drawings.

[0018] First, let's explain the terms used in the following description.

[0019] In the following embodiments, the processor (hereinafter referred to as "processor") can be a single computing device or a combination of multiple computing devices. Furthermore, a processor can be a single computing device or a combination of multiple computing devices. Examples of computing devices include CPU (Central Processing Unit), GPU (Graphics Processing Unit), GPGPU (General-Purpose computing on Graphics Processing Units), APU (Accelerated Processing Unit), or TPU (Tensor Processing Unit), etc.

[0020] In the following implementation, the labeled RAM (Random Access Memory) is a memory that temporarily stores information and is used by the processor as working memory.

[0021] In the following embodiments, the labeled memory is one or more non-volatile storage devices used to store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), disk (e.g., hard disk) or magnetic tape, etc.

[0022] In the following implementation, the labeled Communication I / F (Interface) is an interface that includes a communication processor and an antenna, etc. The Communication I / F is responsible for communication between multiple computers. Examples of communication standards applicable to the Communication I / F include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0023] In the following implementation, "A and / or B" is synonymous with "at least one of A and B". That is, "A and / or B" means that it can be only A, only B, or a combination of A and B. Furthermore, in this specification, when "and / or" connects more than three items, the same approach as "A and / or B" applies.

[0024] [First Implementation] Figure 1 An example of the configuration of the data processing system 10 according to the first embodiment is shown.

[0025] like Figure 1 As shown, the data processing system 10 includes a data processing device 12 and an intelligent 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 includes a processor 28, RAM 30, and memory 32. The processor 28, RAM 30, and memory 32 are connected to a bus 34. Furthermore, 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. An example of the network 54 includes a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0027] The smart device 14 includes a computer 36, a receiver 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, RAM 48, and memory 50. The processor 46, RAM 48, and memory 50 are connected to a bus 52. In addition, the receiver 38, output device 40, and camera 42 are also connected to the bus 52.

[0028] The receiving device 38 includes a touchscreen 38A and a microphone 38B, etc., for receiving user input. The touchscreen 38A receives user input generated by contact with an indicator (e.g., a pen or finger). The microphone 38B receives user input generated by sound by detecting the user's voice. The control unit 46A sends data representing user input received via the touchscreen 38A and microphone 38B to the data processing unit 12. In the data processing unit 12, a specific processing unit 290 (see...) Figure 2 Get the data that represents the user input.

[0029] The output device 40 includes a display 40A and a speaker 40B, etc., and presents data to the user by outputting data in a user-perceptible form (e.g., sound and / or text). The display 40A displays visual information such as text and images according to instructions from the processor 46. The speaker 40B outputs sound according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as 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] Communication I / F 44 is connected to network 54. Communication I / F 44 and 26 are responsible for sending and receiving various information between processor 46 and processor 28 via network 54.

[0031] Figure 2 An example of the main functions of the data processing device 12 and the smart device 14 is shown.

[0032] like Figure 2 As shown, in the data processing apparatus 12, specific processing is performed by the processor 28. A specific processing program 56 is stored in the memory 32. The specific processing program 56 is an example of a "program" as understood in this disclosure. The processor 28 reads the specific processing program 56 from the memory 32 and executes the read specific processing program 56 on the RAM 30. Specific processing is implemented by the processor 28 operating as a specific processing unit 290 based on the specific processing program 56 executed on the RAM 30.

[0033] The memory 32 stores a data generation model 58 and an emotion-specific model 59. The data generation model 58 and the emotion-specific model 59 are used by the specific processing unit 290. The specific processing unit 290 can use the emotion-specific model 59 to infer the user's emotions and perform specific processing based on the user's emotions. The emotion inference function using the emotion-specific model 59 (emotion-specific function) includes various inferences and predictions related to the user's emotions, but is not limited to this example. Furthermore, emotion inference and prediction also include, for example, emotion analysis (analysis).

[0034] In the smart device 14, specific processing is performed by the processor 46. A specific processing program 60 is stored in the memory 50. The specific processing program 60 is used in conjunction with the data processing system 10. The processor 46 reads the specific processing program 60 from the memory 50 and executes the read specific processing program 60 on the RAM 48. Specific processing is implemented by the processor 46 operating as a control unit 46A based on the specific processing program 60 executed on the RAM 48. Furthermore, the smart device 14 may also have the same data generation model and emotion-specific model as the data generation model 58 and the emotion-specific model 59, and use these models to perform the same processing as the specific processing unit 290.

[0035] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device (e.g., a generation server) may have the data generation model 58. In this case, the data processing device 12 obtains the processing results (prediction results, etc.) using the data generation model 58 by communicating with the server device that has the data generation model 58. In addition, the data processing device 12 may be a server device or a user-owned terminal device (e.g., a mobile phone, robot, home appliance, etc.). Next, an example of the processing performed by the data processing system 10 of the first embodiment will be described.

[0036] (Example) The support system described in this invention is designed to support users who are out of work or whose social lives are disrupted due to fractures or injuries. This support system allows users to input their current situation and goals, and a generative AI proposes target tasks before recovery, while also considering their post-recovery stance. For example, users can input information such as their hospitalization due to a fracture, their rehabilitation progress, and their desired occupation and lifestyle after recovery. This information is input into the generative AI, which analyzes it and proposes suitable target tasks. Based on past data and similar cases, the generative AI generates target tasks best suited to the user's situation. For example, it may propose exercise programs corresponding to rehabilitation progress or learning plans for skills needed after recovery. Furthermore, the generative AI considers the user's post-recovery stance. Based on the user's goals and wishes, the generative AI simulates post-recovery occupations and lifestyles and proposes optimal options. For example, it may propose suitable occupational types and ideal lifestyles after recovery. Through this mechanism, users can achieve more effective rehabilitation and a fresh start while maintaining hope and direction. By implementing the target tasks proposed by the generative AI, users can maximize rehabilitation effects and smoothly transition to post-recovery life. Moreover, the generative AI's consideration of the post-recovery stance allows users to recover with peace of mind, having a vision for the future. For example, a long-term hospitalized fracture patient inputs their current condition and goals into the AI ​​generator. The AI ​​then proposes a rehabilitation program based on the user's condition and suggests learning plans for the skills needed after recovery. Simultaneously, based on the user's wishes, the AI ​​simulates their post-recovery career and lifestyle, proposing optimal options. Thus, users can achieve more effective rehabilitation and a fresh start with hope and direction. Through this method, the support system can propose target tasks based on the user's condition and simulate their post-recovery stance. Specifically, this support system can receive information input from the user, including multi-dimensional data such as health status (e.g., fracture location, treatment progress, pain), living environment (e.g., home accessibility, family structure, support system), and psychological state (e.g., anxiety level, motivation, stress level). The system vectorizes this information, for example, as a 128-dimensional numerical vector input to the AI ​​generator. Input examples include "40-year-old male, right leg fracture, 10th day of hospitalization, 30% recovery progress, hopes to work in an office after recovery" and "30-year-old female, left hand fracture, home convalescence, family support, hopes to work part-time after recovery," etc. The AI-generated algorithm employs a large-scale language model or multimodal model based on Transformer to compare the input vector with a database of past cases (such as tens of thousands of rehabilitation records, reinstatement cases, and life reconstruction patterns) to generate target topics most suitable for the user's situation. The output is presented as a structured list of target topics (such as performing rehabilitation exercises A three times a week, learning PC skills after recovery, and a life rhythm adjustment plan) and achievement probability scores for each topic (such as expected achievement rate of 80% and expected achievement rate of 60%).Furthermore, the system simulates post-recovery career adaptability and lifestyle using the user's hopes and goals as parameters. For example, it employs neural networks for career adaptability matching (input: user skills, intentions, constraints; output: suitable career candidate list and adaptability score) and lifestyle simulation (input: living environment, family structure, transportation mode; output: recommended lifestyle and risk assessment). These outputs are visualized in a dashboard format, displaying the advantages and disadvantages of each option and future prediction charts. In terms of subsequent processing, the user selects and executes the proposed tasks, and the progress data is fed back to the system, enabling continuous learning and optimization of the AI ​​model. Technically, this invention differs from traditional human consultation or standardized rehabilitation plans, automatically and quickly proposing high-precision tasks and future simulations tailored to each user's situation. This achieves personalized rehabilitation plans, post-recovery social adaptation support, and data-driven decision support, resulting in significant technical effects such as improved user recovery rates, prevention of relapse, and successful social reintegration. Applicable areas include medical rehabilitation support, career reintegration support, social participation support for people with disabilities, and life reconstruction support for long-term caregivers.

[0037] The support system described in this embodiment includes a receiving unit, an analysis unit, a proposal unit, and a simulation unit. The receiving unit is used to input the user's condition. The user's condition includes, but is not limited to, factors such as health status, living environment, and psychological state. For example, the receiving unit can input information such as the user's hospitalization due to a fracture, rehabilitation progress, and desired occupation and lifestyle after recovery. The analysis unit uses a generative AI to analyze the information input by the receiving unit. Analysis can be based on data analysis methods or algorithms, but is not limited to these. For example, the generative AI can use a machine learning model or data generation algorithm to analyze the user's condition. The proposal unit uses the generative AI to propose target topics based on the information analyzed by the analysis unit. Target topics include, but are not limited to, rehabilitation goals and learning goals. For example, the generative AI generates target topics most suitable for the user's condition based on past data and similar cases. The simulation unit uses the generative AI to simulate the user's post-recovery stance based on the target topics proposed by the proposal unit. Simulation can be based on a simulation model or data, but is not limited to these. For example, the generative AI simulates the user's post-recovery occupation and lifestyle based on the user's goals and wishes, and proposes optimal options. Therefore, the support system of this embodiment can propose target issues based on the user's situation and simulate the post-recovery stance. Specifically, the receiving unit of this support system can receive multi-dimensional data input by the user, such as health status (e.g., fracture location, treatment progress, whether there is pain), living environment (e.g., home accessibility, family structure, support system), and psychological state (e.g., anxiety level, motivation, stress level). The receiving unit converts this information into numerical vectors (e.g., 128-dimensional vectors), performs preprocessing such as standardization or normalization, and then passes it to the parsing unit. Input examples include "40-year-old male, right leg fracture, 10th day of hospitalization, 30% recovery progress, hopes to work in an office after recovery" and "30-year-old female, left hand fracture, home convalescence, family support, hopes to work part-time after recovery," etc. The parsing unit uses a large-scale language model or multimodal model based on Transformer to compare the input vector with a past case database (e.g., tens of thousands of rehabilitation records, reinstatement cases, life reconstruction patterns), and performs optimal feature extraction and clustering. The parsing unit can use a self-attention mechanism to extract important features and perform outlier detection and progress prediction. Based on the analysis results, the proposal department generates a structured list of target topics, including rehabilitation exercise programs (such as exercising A three times a week), learning topics (such as a PC skills learning plan), and lifestyle adjustment plans, and calculates the achievement probability score for each topic (such as an expected achievement rate of 80% or 60%). Furthermore, the proposal department can prioritize and adjust the difficulty of target topics based on the user's physical and psychological constraints and wishes.The simulation department takes the proposed target topic as input and performs neural network-based occupational adaptability matching (input: user skills, intentions, constraints; output: suitable occupation candidate list and adaptability score) and lifestyle simulation (input: living environment, family structure, transportation mode; output: recommended lifestyle and risk assessment). The simulation department can generate multiple scenarios and display the advantages and disadvantages of each option, as well as future prediction charts. In subsequent processing, the user selects and executes the proposed topic, and the progress data is input back into the receiving department to achieve continuous learning and optimization of the AI ​​model. In terms of technical effects, this invention differs from traditional human consultation or standardized rehabilitation plans. It can automatically and quickly propose high-precision topics and future simulations tailored to each user's situation. This achieves personalized rehabilitation plans, post-recovery social adaptation support, and data-driven decision support, resulting in significant technical effects such as improved user recovery rates, prevention of relapse, and successful social reintegration. Applicable areas include medical rehabilitation support, occupational reintegration support, social participation support for people with disabilities, and life reconstruction support for long-term care recipients.

[0038] The support system includes a monitoring unit for monitoring the user's rehabilitation progress. This monitoring unit tracks the user's rehabilitation progress, including, but not limited to, improvements in motor skills and recovery speed. For example, the monitoring unit can measure the user's improvement in motor skills to assess rehabilitation progress. It can also measure the user's recovery speed to assess rehabilitation progress. Thus, by monitoring the user's rehabilitation progress, the monitoring unit provides appropriate support. Specifically, the monitoring unit can acquire the user's body parameters (such as walking distance, muscle strength measurements, range of motion, and balance score) and physiological signals (such as heart rate, blood pressure, and activity level meter data) in real time through sensor devices or wearable terminals. The monitoring unit collects this multidimensional time-series data in tensor form (such as a two-dimensional tensor of sample number × feature number) and performs preprocessing such as noise removal and outlier correction. Input examples include "Daily walking distance: 500 meters, muscle strength measurement: 30 kg, heart rate: 80 bpm" and "Range of motion: 90 degrees, total daily activity recorded by activity level meter: 2000 steps," etc. The monitoring department sends this data to the analysis department or generates AI to calculate rehabilitation progress scores and recovery speed predictions using progress assessment models (such as temporal RNNs or LSTM networks). Output examples include "Rehabilitation Progress: 65%, Recovery Speed: 10% faster than standard" and "Muscle Strength Recovery Prediction: Target expected in two weeks." The monitoring department displays these outputs to users and healthcare professionals through dashboards or notification functions, issuing alerts when progress lags. For subsequent processing, progress data is fed back to the proposal or simulation department for automatic adjustment of target topics or rehabilitation plans. In terms of technical effectiveness, unlike subjective human evaluation or manual recording, the monitoring department achieves objective and high-frequency progress monitoring based on sensor data, enabling optimization of rehabilitation plans, early anomaly detection, and individualized responses for each user. Applicable areas include medical rehabilitation support, home-based care monitoring, sports rehabilitation, and support for people with disabilities.

[0039] The support system includes a generation department that uses generative AI to generate appropriate target topics based on past data and similar cases. The generation department utilizes generative AI to generate appropriate target topics based on past data and similar cases. Generative AI includes, for example, machine learning models or data generation algorithms, but is not limited to these. For example, the generation department can generate rehabilitation goals most suitable for the user's condition based on past rehabilitation data. The generation department can also generate learning goals most suitable for the user's condition based on similar cases. Thus, by generating target topics based on past data and similar cases, the generation department can propose the most suitable target topics to the user. Specifically, the generation department refers to a large database of past rehabilitation records, reinstatement cases, and life reconstruction patterns to calculate the similarity (e.g., cosine similarity, Euclidean distance) between the user's input vector (such as a 128-dimensional vector of health, life, and psychological state) and the database. The generation department uses a large-scale language model or multimodal model based on Transformer to match the input vector with the feature space of past data. Input examples include "40-year-old male, right leg fracture, rehabilitation progress 30%" and "30-year-old female, left hand fracture, home recuperation," etc. After extracting similar cases, the Generation Department generates a list of target topics, including rehabilitation goals (such as walking distance of 1km, muscle strength of 30kg) and learning goals (such as PC skill learning, lifestyle adjustment), and calculates the probability of achievement and recommendation priority for each topic. Output examples include "Rehabilitation Goal: Exercise A 3 times a week, estimated achievement rate 80%" and "Learning Goal: PC skill learning, estimated achievement rate 60%." The Generation Department then links these outputs with the Proposal Department or Simulation Department to automatically adjust target topics or branch scenarios based on user conditions and preferences. In subsequent processing, users select and execute topics, and their progress data is fed back to the Generation Department, enabling continuous learning and optimization of the AI ​​model. Technically, unlike traditional unified rehabilitation plans or goal setting relying on human experience, the Generation Department uses data-driven methods and similar case retrieval in a high-dimensional feature space to automatically generate and improve the accuracy of individual optimized target topics. Applicable areas include medical rehabilitation support, vocational reintegration support, social participation support for people with disabilities, and life reconstruction support for long-term care recipients.

[0040] The support system includes a database department that provides a database for generating AI to simulate post-recovery occupations and lifestyles. The database includes, but is not limited to, occupational and lifestyle data. For example, the database department can simulate suitable occupations for users after recovery based on occupational data. It can also simulate suitable lifestyles for users after recovery based on lifestyle data. Thus, the database department utilizes the database to improve the accuracy of simulations when simulating post-recovery occupations and lifestyles. Specifically, the database department stores tens of thousands of occupational data points (such as occupation type, required skills, work environment, and reinstatement cases) and lifestyle data (such as lifestyle patterns, family composition, transportation methods, and life satisfaction) in a structured database. The database department rapidly retrieves and extracts relevant data based on queries from the AI ​​generation or simulation department (such as user skill set, desired occupation type, and living environment conditions). Input examples include "Desired occupation type: office work, skills: PC operation, constraints: inability to commute" and "Living environment: barrier-free housing, family support," etc. The database department provides the search results to the simulation department as input data for neural network occupational adaptability matching or lifestyle simulation. Output examples include "Suitable Career Candidate: Administrative Position, Adaptability 85%" and "Recommended Lifestyle: Working from Home + Weekly Medical Visit, Risk Assessment: Low." In terms of subsequent processing, simulation results are displayed to users in dashboard format, and user choices and feedback are reflected in the continuous updates and optimizations of the database. Technically, unlike traditional static information that provides career and lifestyle advice or relies on human experience, the database department achieves high-speed, high-precision simulations through structured data and AI, enabling individual optimization for each user and improved reliability of future predictions. Applicable areas include medical rehabilitation support, vocational reintegration support, social participation support for people with disabilities, and life reconstruction support for long-term care recipients.

[0041] The proposal department can generate exercise programs corresponding to rehabilitation progress. Utilizing generative AI, the department proposes exercise programs based on rehabilitation progress. These programs include, but are not limited to, exercise type, intensity, and frequency. For example, the department can propose appropriate exercise programs based on the user's rehabilitation progress. It can also adjust the intensity and frequency of the exercise programs based on the user's exercise capacity. Thus, by proposing exercise programs based on rehabilitation progress, the department can maximize the user's rehabilitation effect. Specifically, the department receives user exercise capacity data (such as muscle strength measurements, walking distance, joint range of motion, and activity level) and rehabilitation progress scores from the monitoring or analysis department. The department converts this data into 128-dimensional vectors and inputs it into the generative AI (such as a large language model based on Transformer or a temporal RNN). Input examples include "Muscle strength: 30kg, Walking distance: 500 meters, Progress: 65%" and "Joint range of motion: 90 degrees, Activity level: 2000 steps," etc. The proposal department compares the exercise program with past rehabilitation databases or similar cases to generate exercise programs (e.g., Exercise A 3 times a week, Exercise B 2 times a week, intensity level 2, frequency: every other day). Output examples include "Recommended exercise: 10 squats x 3 sets, 3 times a week, intensity level 2" and "Stretching exercise, daily, intensity level 1," etc. Furthermore, the proposal department can automatically adjust the program's difficulty and frequency based on the user's physical constraints or psychological state (e.g., fatigue, motivation). In terms of follow-up processing, the proposed exercise program is shown to the user or medical staff, and the implementation results and feedback are input back into the system, enabling continuous learning and optimization of the AI ​​model. In terms of technical effectiveness, unlike traditional uniform exercise programs or suggestions relying on human experience, the proposal department achieves individual optimization and automatic adjustment through sensor data and progress tracking, maximizing rehabilitation effects, accelerating recovery, and preventing relapse. Applicable areas include medical rehabilitation support, home-based care support, sports rehabilitation, and support for people with disabilities.

[0042] The simulation department can simulate post-recovery careers and lifestyles based on user preferences. Utilizing generative AI, the simulation department simulates these preferences based on user desires, including, but not limited to, career types and lifestyle choices. For example, the simulation department can simulate suitable post-recovery careers based on user preferences. It can also simulate suitable post-recovery lifestyles. Thus, by simulating post-recovery careers and lifestyles based on user preferences, the simulation department can provide users with optimal options. Specifically, the simulation department vectorizes user-inputted preference information (such as desired career type, work style, pace of life, family structure, mode of transportation, hobbies, etc.) into numerical vectors of 128 dimensions or more, and inputs this data into the generative AI. The simulation department then uses a large-scale language model based on Transformer or a multimodal neural network to compare the input vectors with a database of career and lifestyle data (such as tens of thousands of career cases, lifestyle patterns, and post-recovery satisfaction data). Input examples include "Desired career type: working from home, lifestyle: early riser, family structure: spouse + 1 child" and "Desired career type: light physical labor, lifestyle: 3-day work week, transportation: public transportation." The simulation department calculates the similarity (e.g., cosine similarity, Euclidean distance) between the input vector and each case in the database, extracting the most suitable career and lifestyle candidates. The simulation department also performs neural network future predictions on the extracted candidates (e.g., post-return-to-work satisfaction rating, life risk assessment, health maintenance prediction), and scores each option. Output examples include "Suitable career candidate: administrative job (fitness 85%, satisfaction prediction 80 points)" and "Recommended lifestyle: working from home + weekly medical visit (risk assessment: low)." The above outputs are displayed to the user in the form of a dashboard, showing the advantages and disadvantages of each option and future prediction charts. In terms of subsequent processing, the user selects options, and their selections and feedback are fed back to the simulation department or database department, enabling continuous learning and optimization of the AI ​​model. In terms of technical effectiveness, the simulation department differs from traditional methods that rely on human experience or standardized reinstatement support. It compares multidimensional user intention data with a large case database in a high-dimensional space to automatically generate and rapidly display individually optimized future scenarios. This achieves personalized social adaptation support after recovery, enhanced life reconstruction, and sophisticated decision support, resulting in significant technical effects such as improved user satisfaction, prevention of relapse, and smooth social reintegration. Applicable areas include medical rehabilitation support, vocational reintegration support, social participation support for people with disabilities, life reconstruction support for long-term care recipients, and life planning and re-employment support for the elderly.

[0043] The receiving unit can infer the user's emotions and adjust the display of the input interface accordingly. The receiving unit utilizes generative AI to infer the user's emotions and adjust the display of the input interface based on these inferred emotions. User emotions include, but are not limited to, feelings of unease, relaxation, and anxiety. For example, when the user feels uneasy, the receiving unit provides a calming interface to provide a sense of security. When the user is relaxed, the receiving unit provides a bright interface to make the input process more pleasant. When the user is anxious, the receiving unit provides a simple and highly visible interface for quick input. Thus, by adjusting the display of the input interface according to the user's emotions, the receiving unit makes the user's input process more comfortable. Emotion inference can be achieved through emotion inference functions such as emotion engines or generative AI. Generative AI can be text-based AI (such as LLM) or multimodal AI, but is not limited to these. Specifically, the receiving unit can receive various input data from the user, including text data (such as natural language phrases like "Recent recovery has been very difficult" or "I'm in a good mood today"), speech data (such as speech rate and tone), and facial images (such as facial images captured by a camera). The receiving unit preprocesses this data, including word segmentation and embedding vectorization (e.g., 512-dimensional vectors) for text, spectral transformation and feature extraction (e.g., MFCC features) for speech, and facial feature point extraction (e.g., 68-point markers) for images. Input examples include "speech text of 'I'm feeling uneasy today'", "speech data with a low tone", and "smiling facial images". The receiving unit inputs these multimodal feature vectors into a multimodal Transformer or a large language model with self-attention mechanisms, outputting emotion classifications (e.g., labels such as uneasy, relaxed, anxious, joyful, angry) and emotion scores (e.g., probability distributions such as 0.8 for uneasy and 0.2 for relaxed). Output examples include "emotion label: uneasy, score 0.75" and "emotion label: relaxed, score 0.60". Based on these outputs, the receiving unit automatically adjusts the interface color scheme (e.g., calming blue, bright yellow), layout (e.g., simple button arrangement, detailed description display), and input assistance (e.g., whether to display input guidance, number of input items). In terms of post-processing, continuous data collection on user interface feedback (such as input speed, number of erroneous inputs, and re-input requests) is used for AI model parameter optimization and interface improvement. Technically, unlike traditional static interface design or subjective human judgment, the receiving unit utilizes AI to achieve multi-dimensional emotion inference and real-time interface optimization, significantly reducing user psychological burden, improving input efficiency, decreasing erroneous inputs, and increasing user satisfaction. Applicable areas include medical rehabilitation support systems, input support for people with disabilities, stress management applications, educational interfaces, and customer support systems.

[0044] The receiving unit can analyze a user's past input history and propose appropriate input methods. The receiving unit utilizes generative AI to analyze a user's past input history and propose suitable input methods. Past input history includes, but is not limited to, methods of storing and analyzing input data. For example, the receiving unit can automatically display frequently entered information as candidates. The receiving unit can also prioritize recommending input methods previously used by the user (voice, text, etc.). The receiving unit can also predict and recommend information to be used in specific time periods based on the user's past input history. Thus, by analyzing a user's past input history, the receiving unit can propose the optimal input method. Specifically, the receiving unit retrieves chronologically recorded input history data for each user from the database (such as structured log data including input time, content, method, required time, number of errors, etc.). The receiving unit inputs this historical data into the generative AI as a time-series vector (e.g., a 128-dimensional vector for each historical record, with the most recent 30 records forming an array). Input examples include "2024-06-01 10:00 Text input 'Right leg pain'", "2024-06-02 09:30 Voice input 'Rehabilitation progress 30%'", and "2024-06-03 20:00 Text input 'Good mood'", etc. The receiving unit uses a Transformer-based time series analysis model or an LSTM network to extract and cluster input history patterns, automatically learning user-preferred input methods (e.g., high frequency of voice input, more text input at night) and input tendencies (e.g., repeated input of specific items, input content that changes by weekday). Output examples include "Recommended input method: Voice input (recommendation score 0.85)", "Candidate input items: 'Rehabilitation progress' 'Is there pain'", "Recommended input time period: After 20:00 at night", etc. Based on these output results, the receiving unit implements automatic input candidate completion, default input method switching, and priority display of input items by time period on the input interface. In terms of post-processing, the actual input methods and results selected and used by users are accumulated as feedback data for continuous learning and optimization of the AI ​​model. Regarding technical effectiveness, unlike traditional static input forms or input support relying on human experience, the receiving unit automates and accelerates historical pattern analysis and individual optimized input suggestions through AI, achieving significant technical effects such as improved input efficiency, reduced erroneous input, reduced user burden, and increased system utilization. Applicable areas include medical rehabilitation support systems, input assistance for people with disabilities, daily business reporting systems, educational learning record systems, and customer support reception, among others.

[0045] The receiving unit can customize input items based on the user's current physical and mental state during input. Utilizing generative AI, the receiving unit customizes input items based on the user's current physical and mental state. Physical state includes, but is not limited to, health status and athletic ability. For example, when the user is fatigued, the receiving unit minimizes input items and simplifies the process. When the user is relaxed, the receiving unit provides detailed input items and recommends customizable input methods. When the user is stressed, the receiving unit prioritizes voice input for faster input. Thus, by customizing input items based on the user's current physical and mental state, the receiving unit can reduce the user's burden. Specifically, the receiving unit acquires multi-dimensional numerical vectors (e.g., 128-dimensional) of the user's input health status data (such as walking distance, muscle strength measurements, fatigue scores, and pain levels) and mental state data (such as stress levels, motivation, and mood scores) in real time. The receiving unit inputs this data into generative AI (such as a multimodal large-scale language model) to classify (e.g., fatigue, relaxation, stress) or score (e.g., fatigue level 0.7, stress level 0.8). Input examples include "Muscle strength: 25kg, Fatigue level: 0.8, Stress level: 0.6" and "Mood: Relaxed, no pain." The receiving unit automatically adjusts the number and content of input items based on the AI ​​output. For example, when fatigue is high, only the minimum items such as "Today's recovery progress" and "Is there pain?" are displayed; when relaxed, multiple items such as "Detailed recovery content," "Changes in living environment," and "Details of psychological state" are displayed. When stress is high, voice input or selective input is prioritized to reduce the input burden. Output examples include "Number of items displayed: 2, Input method: Voice priority" and "Number of items displayed: 8, Input method: Text + Selective." In terms of subsequent processing, user input completion rate, required time, and error rate are continuously monitored for AI model parameter optimization and input item design improvement. In terms of technical effectiveness, unlike traditional unified input forms or subjective human judgment, the receiving unit uses AI to achieve multi-dimensional state estimation and real-time input item optimization, significantly reducing the user burden, improving input efficiency and accuracy, and increasing user satisfaction. It is applicable to various fields including medical rehabilitation support systems, input support for people with disabilities, stress management applications, educational interfaces, and daily business reporting systems.

[0046] The receiving unit can infer the user's emotions and determine the input priority order based on the inferred emotions. The receiving unit utilizes generative AI to infer the user's emotions and determine the input priority order based on the inferred emotions. User emotions include, but are not limited to, anxiety, relaxation, and unease. For example, when the user is anxious, the receiving unit prioritizes displaying important input items. When the user is relaxed, the receiving unit provides detailed input items and recommends customizable input methods. When the user is uneasy, the receiving unit prioritizes displaying input items that provide a sense of security. Thus, by determining the input priority order based on the user's emotions, the receiving unit can prioritize displaying important input items. Emotion inference can be achieved through emotion inference functions such as emotion engines or generative AI. Generative AI can be text-based AI (such as LLM) or multimodal AI, but is not limited to these. Specifically, the receiving unit can receive multimodal data such as natural language text input by the user (e.g., "I'm in a hurry today," "I feel uneasy"), speech data (e.g., speech rate, tone), and facial expression images (e.g., facial images acquired by a camera). The receiving unit preprocesses this data, performing text segmentation and vectorization, speech feature extraction, and facial feature point extraction from images, before inputting it into a multimodal AI model (e.g., Transformer-based). The AI ​​model outputs emotion classifications (e.g., anxiety, relaxation, unease) and emotion scores (e.g., anxiety level 0.8, unease level 0.6). Output examples include "Emotion: Anxiety, Score 0.85" and "Emotion: Relaxation, Score 0.70." Based on these outputs, the receiving unit automatically determines the priority order of input items. For example, when anxiety is high, "high-urgency input items (e.g., pain status, recovery progress)" are prioritized; when relaxation is high, "detailed input items (e.g., living environment, psychological state)" are added; and when unease is high, "input items that provide reassurance (e.g., support system, consultation window)" are displayed first. In subsequent processing, user input selections, completion rates, and required time are continuously monitored for AI model parameter optimization and input item design improvement. In terms of technical effectiveness, the receiving unit differs from traditional static input forms or priority sorting based on human experience. It utilizes AI to achieve multi-dimensional emotional inference and real-time priority control of input items, improving input efficiency, reducing user burden, quickly obtaining important information, and increasing user satisfaction. Applicable areas include medical rehabilitation support systems, input support for people with disabilities, stress management applications, educational interfaces, and customer support reception, among others.

[0047] The receiving unit can consider the user's geographic location information during input, prioritizing the display of highly relevant input items. Utilizing generative AI, the receiving unit considers the user's geographic location information during input, prioritizing the display of highly relevant input items. Geographic location information includes, but is not limited to, GPS data and location information services. For example, when the user is in a hospital, the receiving unit prioritizes displaying medical-related input items. When the user is at home, the receiving unit prioritizes displaying rehabilitation-related input items. When the user is away from home, the receiving unit prioritizes displaying mobile-related input items. Thus, by considering the user's geographic location information and prioritizing the display of highly relevant input items, the receiving unit improves input efficiency. Specifically, the receiving unit can acquire in real-time GPS coordinate data (e.g., latitude 35.6895, longitude 139.6917, accuracy 5 meters), Wi-Fi / Bluetooth location information, facility names and addresses from location information service APIs, etc., from user terminals (e.g., location coordinates + facility category + time information forming a 64-dimensional vector), and vectorize it. The receiving unit inputs these geographic location vectors into the generating AI (such as a multimodal Transformer model), comparing them with historical databases and facility attribute databases (such as labeled data for hospitals, rehabilitation facilities, homes, public transportation, and commercial facilities). Input examples include "Location: Inside the hospital, Time: 10:00 AM," "Location: Home, Time: 8:00 PM," and "Location: Inside the station, Time: 3:00 PM." The generating AI calculates the similarity between the input location vectors and the facility attribute database (such as cosine similarity and distance score), automatically determining the most relevant scenario. Outputs include structured lists such as "Priority Input Items: Treatment content, medication status, next visit plan (when staying in the hospital)," "Priority Input Items: Rehabilitation progress, daily activities, family support status (when staying at home)," and "Priority Input Items: Mode of transportation, distance traveled, purpose of going out (when going out)." In addition, the receiving unit also automatically adjusts the order and display format of input items (such as buttons, selection options, and voice input recommendations) based on the user's past input preferences and time information. In terms of post-processing, feedback data such as actual user input, completion rate, and time required are accumulated and used for continuous learning and optimization of the AI ​​model. Regarding technical effects, unlike traditional static input forms or item prompts relying on human experience, the receiving unit uses AI to achieve real-time geolocation estimation and multi-dimensional data analysis, enabling automatic prompts for input items based on the user's current location and usage scenario. This results in significant technical effects such as improved input efficiency, reduced erroneous input, reduced user burden, and increased system utilization. Applicable areas include medical rehabilitation support systems, home-based care support, outing support applications, input assistance for people with disabilities, daily business reporting systems, and mobile health management applications, among others.

[0048] The receiving unit can analyze a user's social media activity during input and propose relevant input items. Utilizing generative AI, the receiving unit analyzes a user's social media activity during input and proposes relevant input items. Social media activity includes, but is not limited to, content posted and frequency of activity. For example, when a user posts content related to rehabilitation on social media, the receiving unit proposes rehabilitation-related input items. When a user posts content related to career on social media, the receiving unit proposes career-related input items. When a user posts content related to health on social media, the receiving unit proposes health-related input items. Thus, the receiving unit can propose relevant input items by analyzing a user's social media activity. Specifically, within the scope of user permission, the receiving unit collects posting data (such as text, images, videos, posting time, number of likes, number of comments, etc.) from social media APIs, collects it chronologically, and uses a natural language processing engine to segment and vectorize the posted content (e.g., 512-dimensional vectors for each post). Furthermore, to categorize posting frequency and type (e.g., rehabilitation, career, health, interests, family, travel, etc.), it employs a large-scale language model or multimodal model based on Transformer for semantic parsing and sentiment analysis (e.g., positive, negative, neutral). Input examples include "2024-06-01 'Keep going in your recovery'", "2024-06-02 'Challenge a new job'", and "2024-06-03 'Recent health is good'". Based on these analysis results, the receiving department infers the user's areas of interest and current life status, prioritizing the display of highly relevant input items (such as recovery progress, career aspirations, health status, changes in living environment, etc.) on the input interface. Output examples include "Recommended input items: recovery progress, exercise content", "Recommended input items: career aspirations, work format", and "Recommended input items: health status, medication status", etc. In addition, the receiving department also automatically adjusts the timing of input item display and input assistance functions (such as auto-completion and input candidate suggestions) based on the posting frequency and time pattern (such as posting more health content at night). In terms of subsequent processing, the actual user input content, completion rate, time required, error rate, etc., are accumulated as feedback data for continuous learning and optimization of the AI ​​model. In terms of technical effectiveness, the receiving unit differs from traditional static input forms or input support systems that rely on human experience. It utilizes AI to automate and accelerate social media activity analysis and individual optimized input suggestions, resulting in significant improvements in input efficiency, reduced typos, less user burden, and increased system utilization. Applicable areas include medical rehabilitation support systems, input assistance for people with disabilities, daily business reporting systems, educational learning record systems, customer support reception, and personal health management applications.

[0049] The parsing unit can infer a user's emotions and adjust its parsing algorithm accordingly. The parsing unit utilizes generative AI to infer user emotions and adjusts the parsing algorithm based on these inferred emotions. User emotions include, but are not limited to, relaxation, anxiety, and unease. For example, when a user is relaxed, the parsing unit performs detailed analysis to improve accuracy. When a user is anxious, the parsing unit performs rapid analysis to provide quick results. When a user is uneasy, the parsing unit provides reassuring analysis results. Thus, by adjusting the parsing algorithm based on user emotions, the parsing unit improves the accuracy of the parsing results. Emotion inference can be achieved through emotion inference functions such as emotion engines or generative AI. Generative AI can be text-based AI (such as LLM) or multimodal AI, but is not limited to these. Specifically, the parsing unit can receive multimodal data such as natural language text input by the user (e.g., "I'm in a good mood today," "I feel uneasy"), speech data (e.g., speech rate, tone), and facial expression images (e.g., facial images captured by a camera). The parsing unit preprocesses this data, including text segmentation and vectorization (e.g., 512-dimensional vectors), speech MFCC feature extraction, and image facial feature point extraction (e.g., 68-point markers). Input examples include speech text of "'I'm very relaxed today'", "high-pitched speech data", and "smiling face images". The parsing unit inputs these multimodal feature vectors into a multimodal Transformer or a large language model with self-attention mechanisms, outputting sentiment classifications (e.g., labels such as relaxed, anxious, and uneasy) and sentiment scores (e.g., a probability distribution of relaxation level 0.7, anxiety level 0.2, and uneasy level 0.1). Output examples include "sentiment label: relaxed, score 0.75" and "sentiment label: anxious, score 0.60". Based on these outputs, the parsing unit automatically adjusts the parsing algorithm parameters (e.g., parsing depth, number of feature selections, batch size, and threshold settings). For example, when the relaxation level is high, detailed feature extraction and multi-stage clustering are performed to maximize parsing accuracy. When the anxiety level is high, the number of features is simplified, the number of inferences is reduced, and speed is prioritized. When anxiety levels are high, the emphasis is on the interpretability and reassurance of the analysis results, highlighting interpretable features and comparisons with past successful cases. In subsequent processing, the analysis results are linked with the proposal or simulation departments to automatically adjust the target topic or future scenario based on the user's emotional state. Technically, the analysis department differs from traditional unified analysis processes or subjective human judgment by using AI to achieve multi-dimensional emotional inference and real-time analysis algorithm optimization, significantly improving analysis accuracy, optimizing processing speed, reducing user psychological burden, and enhancing interpretability. Applicable areas include medical rehabilitation support systems, analysis support for people with disabilities, stress management applications, educational analysis systems, and customer support analysis, among others.

[0050] The analysis department can improve analysis accuracy by referencing the user's past data during the analysis process. The analysis department utilizes generated AI to improve analysis accuracy by referencing the user's past data. Past data includes, but is not limited to, historical data and log data. For example, the analysis department can refer to the user's past rehabilitation data for optimal analysis. The analysis department can also refer to the user's past health data to improve analysis accuracy. Furthermore, the analysis department can refer to the user's past occupational data for post-recovery occupational-related analysis. Thus, the analysis department improves analysis accuracy by referencing the user's past data. Specifically, the analysis department refers to a structured database of each user's chronologically recorded rehabilitation history data (such as rehabilitation date, exercise content, progress, and achievement rate), health data (such as body temperature, blood pressure, heart rate, and muscle strength measurements), and occupational data (such as past occupational type, job content, work format, and reinstatement history). The analysis department inputs this historical data into the generated AI as a time-series vector (e.g., a 128-dimensional vector for each historical record, with the most recent 30 records forming an array). Input examples include "2024-06-01 Rehabilitation progress 60%", "2024-06-02 Muscle strength measurement 28kg", and "2024-06-03 Occupation: Administrative". The analysis department uses a Transformer-based time series analysis model or an LSTM network to extract and cluster historical data patterns, automatically learning user recovery trends, changes in health status, and changes in occupational adaptability. AI outputs include the selection of the optimal analysis method for the current situation (such as optimization of progress prediction model parameters and automatic adjustment of outlier detection thresholds), a comparison chart with past data, and future predictions (such as predicted muscle strength recovery in two weeks). Output examples include "Progress prediction: Expected achievement rate 80%", "Health status: Stable", and "Occupational adaptability: Office work adaptability 85%". In subsequent processing, the analysis results are linked with the proposal department or simulation department to generate individual optimized target topics or future scenarios based on the user's past data. In terms of technical effectiveness, the analysis department differs from traditional static analysis or judgments relying on human experience. It utilizes AI to automatically select historical pattern analysis and individual optimal analysis methods, improving analysis accuracy, early anomaly detection, personalized responses for each user, and enhanced decision support. Applicable areas include medical rehabilitation support systems, analysis support for people with disabilities, daily business report analysis, education and learning record analysis, and customer support history analysis, among others.

[0051] The analysis department can customize the analysis method based on the user's current physical condition during the analysis process. Utilizing generative AI, the analysis department tailors the analysis method according to the user's current physical condition. Physical condition includes, but is not limited to, health status and athletic ability. For example, the analysis department uses a simplified analysis method when the user is fatigued, a detailed analysis method when the user is relaxed, and a fast analysis method when the user is stressed. Thus, by customizing the analysis method based on the user's current physical condition, the analysis department can provide the optimal analysis results for the user. Specifically, the analysis department acquires multi-dimensional numerical vectors (e.g., 128-dimensional) of user input, including health status data (such as walking distance, muscle strength measurements, fatigue score, and pain status), athletic ability data (such as joint range of motion, activity level, and balance score), and psychological state data (such as stress level, motivation, and mood score). The analysis department inputs this data into the generative AI (e.g., a multimodal large-scale language model) to classify (e.g., fatigue, relaxation, stress) or score the user's state (e.g., fatigue 0.7, stress 0.8). Input examples include "Muscle strength: 25kg, Fatigue level: 0.8, Stress level: 0.6" and "Mood: Relaxed, no pain." The analysis department automatically adjusts the analysis method selection and parameters (such as the number of selected features, analysis depth, batch size, and threshold settings) based on the AI ​​output. For example, when fatigue level is high, only the main features are extracted, using simple clustering or a progression prediction model to reduce computational load. In a relaxed state, multi-stage analysis, detailed outlier detection, and model ensemble analysis are performed to maximize accuracy. When stress level is high, speed is prioritized, selecting a high-real-time inference model. Output examples include "Analysis method: Simple clustering, 5 main features" and "Analysis method: Detailed multi-stage analysis, 20 features." In subsequent processing, the analysis results are linked with the proposal department or simulation department to automatically adjust the target topic or future scenario based on the user's physical condition. In terms of technical effectiveness, the analysis department differs from traditional unified analysis processes or subjective human judgment. Through AI, it achieves multi-dimensional state estimation and real-time analysis method optimization, significantly reducing user burden, improving analysis efficiency and accuracy, and increasing user satisfaction. It is applicable to various fields including medical rehabilitation support systems, disability analysis support, stress management applications, education analysis systems, and daily business report analysis.

[0052] The analysis unit can infer the user's emotions and adjust the display of the analysis results accordingly. The analysis unit utilizes generative AI to infer the user's emotions and adjust the display of the analysis results based on these inferred emotions. User emotions include, but are not limited to, tension, relaxation, and urgency. For example, when the user is tense, the analysis unit provides a concise and highly visual display. When the user is relaxed, the analysis unit provides a display containing detailed information. When the user is in a hurry, the analysis unit provides a display highlighting key points. Thus, by adjusting the display of analysis results based on the user's emotions, the analysis unit achieves a display that is easy for the user to understand. Emotion inference can be achieved through emotion inference functions such as emotion engines or generative AI. Generative AI can be text-based AI (such as LLM) or multimodal AI, but is not limited to these. Specifically, the analysis unit can receive multimodal data such as natural language text input by the user (e.g., "I'm in a hurry today," "I'm very nervous"), speech data (e.g., speech rate, tone), and facial expression images (e.g., facial images captured by a camera). The parsing department preprocesses this data, performing text segmentation and vectorization, speech feature extraction, and facial feature point extraction from images, before inputting it into a multimodal AI model (e.g., Transformer-based). The AI ​​model outputs emotion classifications (e.g., tension, relaxation, urgency) and emotion scores (e.g., tension 0.8, relaxation 0.6, urgency 0.7). Output examples include "Emotion: Tension, Score 0.85" and "Emotion: Relaxation, Score 0.70." Based on these outputs, the parsing department automatically adjusts the display of the parsing results (e.g., color scheme, layout, information content, and whether to display charts). For example, when tension is high, a simple color scheme and large font are used to highlight only key points; when relaxation is high, detailed charts, supplementary explanations, and comparisons with past data are added. When urgency is high, only the most important items are placed at the top, and detailed information is collapsed. In subsequent processing, user display choices, browsing time, and requests for re-display are accumulated and used for continuous learning and optimization of the AI ​​model. In terms of technical effectiveness, the analysis department differs from traditional static display designs or subjective human judgment. It utilizes AI to achieve multi-dimensional emotional inference and real-time display optimization, significantly reducing user psychological burden, improving information comprehension, preventing misunderstandings, and increasing user satisfaction. Applicable areas include medical rehabilitation support systems, disability analysis support, stress management applications, educational analysis systems, and customer support analysis, among others.

[0053] The parsing department can improve parsing accuracy by considering the user's geographical location information during the parsing process. The parsing department utilizes generative AI to improve accuracy by taking the user's geographical location information into account during parsing. Geographical location information includes, but is not limited to, GPS data and location information services. For example, when the user is in a hospital, the parsing department prioritizes parsing medical-related data. When the user is at home, the parsing department prioritizes parsing rehabilitation-related data. When the user is out and about, the parsing department prioritizes parsing data related to movement. Thus, by considering the user's geographical location information, the parsing department improves parsing accuracy. Specifically, the parsing department can acquire GPS coordinate data (e.g., latitude 35.6895, longitude 139.6917, accuracy 5 meters), Wi-Fi / Bluetooth location information, facility names and addresses from location information service APIs, etc., from user terminals (such as smartphones, tablets, and wearable devices) in real time, and vectorize them (e.g., a 64-dimensional vector composed of location coordinates + facility category + time information). The analysis department inputs these geographic location vectors into the generative AI (such as a multimodal Transformer model) and compares them with historical databases and facility attribute databases (such as labeled data for hospitals, rehabilitation facilities, homes, public transportation, and commercial facilities). Input examples include "Location: Inside a hospital, Time: 10:00 AM," "Location: Home, Time: 8:00 PM," and "Location: Inside a station, Time: 3:00 PM." The generative AI calculates the similarity between the input location vectors and the facility attribute database (such as cosine similarity and distance score), automatically determining the most relevant scenario. Based on the determination result, the analysis department automatically switches the priority and analysis method of the data to be analyzed (such as prioritizing medical data, rehabilitation data, and mobile data). Output examples include "Analysis object: Medical data, priority 90%" and "Analysis object: Rehabilitation data, priority 80%." In subsequent processing, the analysis results are linked with the proposal department or simulation department to automatically adjust the target topic or future scenario based on the user's geographic location. In terms of technical effectiveness, the parsing department differs from traditional unified parsing processes or reliance on human experience. It utilizes AI to achieve real-time geolocation estimation and multi-dimensional data analysis, automating and improving the accuracy of parsing based on the user's current location and usage scenario. This results in significant technical benefits such as improved parsing efficiency, reduced misparsing, reduced user burden, and increased system utilization. Applicable areas include medical rehabilitation support systems, home-based care support, mobile support applications, parsing assistance for people with disabilities, daily business report parsing, and mobile health management parsing, among others.

[0054] The parsing department is able to analyze users' social media activities during parsing and use relevant data for analysis. The parsing department utilizes generative AI to analyze users' social media activities during parsing and use relevant data for analysis. Social media activities include, but are not limited to, content posting and activity frequency. For example, when a user posts rehabilitation-related content on social media, the parsing department uses rehabilitation-related data for analysis. When a user posts career-related content on social media, the parsing department uses career-related data for analysis. When a user posts health-related content on social media, the parsing department uses health-related data for analysis. Thus, by analyzing users' social media activities, the parsing department can use relevant data for analysis. Specifically, within the scope of user permission, the parsing department collects posting data (such as text, images, videos, posting time, number of likes, number of comments, etc.) from social media APIs, collects it chronologically, and uses a natural language processing engine to segment and vectorize the posted content (e.g., 512-dimensional vectors for each post). In addition, to categorize posting frequency and type (e.g., rehabilitation, career, health, interests, family, travel), a large-scale language model or multimodal model based on Transformer is used for semantic parsing and sentiment analysis (e.g., positive, negative, neutral). Input examples include "2024-06-01 'Keep going in rehabilitation'", "2024-06-02 'Challenge a new job'", and "2024-06-03 'Recently feeling good'". Based on these analysis results, the parsing department infers the user's areas of interest and current life status, prioritizing the extraction of highly relevant data (e.g., rehabilitation progress, career aspirations, health status, changes in living environment) as parsing objects. AI outputs include "Parsing object: rehabilitation data, priority 90%", "Parsing object: career data, priority 80%", and "Parsing object: health data, priority 85%". Furthermore, the parsing timing and methods (e.g., time series analysis, clustering, sentiment trend analysis) are automatically adjusted based on posting frequency and time patterns (e.g., more health content posted at night). In terms of subsequent processing, the analysis results are linked with the proposal department or simulation department to generate individual optimization target topics or future scenarios based on users' social media activities. In terms of technical effectiveness, the analysis department differs from traditional unified analysis processes or analyses relying on human experience. It uses AI to automate and accelerate social media activity analysis and the extraction of individual optimization analysis objects, achieving significant technical effects such as improved analysis efficiency, reduced misanalysis, reduced user burden, and increased system utilization. Applicable areas include medical rehabilitation support systems, analysis assistance for people with disabilities, daily business report analysis, education and learning record analysis, customer support analysis, and personal health management analysis, among others.

[0055] The proposal department can infer users' emotions and adjust the expression of proposals accordingly. The department utilizes generative AI to infer user emotions and adjusts the expression of proposals based on these inferences. User emotions include, but are not limited to, relaxation, anxiety, and unease. For example, when a user is relaxed, the proposal department provides detailed proposals with more options; when a user is anxious, it provides concise proposals with quick options; when a user feels uneasy, it provides reassuring proposals with fewer options. Thus, the proposal department can adjust the expression of proposals based on user emotions, resulting in proposals that are easy for users to understand. Emotion inference can be achieved through emotion engines or generative AI. Generative AI can be text-generating AI (such as LLM) or multimodal generative AI, but is not limited to these. Specifically, the proposal department can receive multimodal data such as natural language text input by users (e.g., "Today is calm," "Very anxious," "Somewhat uneasy"), voice data (e.g., speech rate or tone), and facial images (e.g., facial images obtained through a camera). This proposal preprocesses the data, performing word segmentation and vectorization (e.g., 512-dimensional vectors) on the text, MFCC feature extraction on the speech, and facial feature point extraction (e.g., 68-dot markers) on the images. Input examples include "speech text of 'I'm very relaxed today'", "speech data with a high pitch", and "facial images with a smile". This proposal inputs these multimodal feature vectors into a large language model with a multimodal Transformer or self-attention mechanism, outputting sentiment classification (e.g., labels such as relaxed, anxious, and uneasy) or sentiment scores (e.g., probability distributions of relaxation 0.7, anxiety 0.2, and uneasy 0.1). Output examples include "sentiment label: relaxed, score 0.75" and "sentiment label: anxious, score 0.60". Based on these outputs, this proposal automatically adjusts the expression of the proposal (e.g., level of detail, number of options, length of explanatory text, color scheme, and layout). For example, when the user is highly relaxed, detailed explanations and multiple options (e.g., rehabilitation exercises A-C, lifestyle improvement plans 1-3) are provided to facilitate comparison and consideration. When the user is highly anxious, a concise proposal containing only the key points is displayed with a large button (e.g., only one most important topic, two options) to support quick decision-making. When the user is highly uneasy, a reassuring color scheme (e.g., blue tones) is used, emphasizing the support system and past success stories, and the number of options is reduced (e.g., 1-2 items). In terms of follow-up processing, data on user reactions when selecting and executing proposals (e.g., selection speed, re-proposal requests, satisfaction feedback) is continuously collected for AI model parameter optimization and proposal expression improvement. In terms of technical effectiveness, this proposal department differs from traditional static proposal displays and subjective human judgment. Through AI, it achieves multi-dimensional emotional inference and real-time proposal expression optimization, significantly reducing user psychological burden, improving decision-making efficiency, preventing misunderstandings, and increasing user satisfaction.It is applicable to various fields including medical rehabilitation support systems, decision support for people with disabilities, stress management applications, educational proposal systems, and customer support proposals.

[0056] The proposal department can improve the accuracy of proposals by referencing users' past data. The proposal department utilizes generated AI to improve the accuracy of proposals by referencing users' past data. Past data includes, but is not limited to, historical data and log data. For example, the proposal department can refer to users' past rehabilitation data to make proposals most suitable for their current situation; it can also refer to users' past health data to improve the accuracy of proposals; and it can also refer to users' past occupational data to make career proposals regarding their return to work. Thus, by referring to users' past data, the proposal department can improve the accuracy of proposals. Specifically, this proposal department refers to a structured database of rehabilitation history data (e.g., rehabilitation implementation date, exercise content, progress, achievement rate), health data (e.g., body temperature, blood pressure, heart rate, muscle strength measurements), and occupational data (e.g., past occupations, job content, work style, return-to-work history) recorded for each user in a time series format. This historical data is input into the generated AI as a time series vector (e.g., each historical record is vectorized into 128 dimensions, with the most recent 30 records forming an array). Input examples include "2024-06-01 Rehabilitation progress 60%", "2024-06-02 Muscle strength measurement 28kg", and "2024-06-03 Occupation: Administrative". This proposal department utilizes a Transformer-based time series analysis model or LSTM network to automatically extract patterns from historical data and perform clustering, learning the user's recovery trend, changes in health status, and changes in occupational adaptability. AI outputs include selecting the most suitable proposal method for the current situation (e.g., optimization of progress prediction model parameters, automatic adjustment of outlier detection thresholds), comparison charts with past data, and future predictions (e.g., predicted muscle strength recovery in two weeks). Output examples include "Recommended proposal: Exercise A 3 times a week, achieving 80% of the expected result", "Health management proposal: Increase blood pressure measurement frequency", and "Occupational return-to-work proposal: Office-related adaptability 85%". In subsequent processing, the proposal results will be linked with the user or simulation department to generate individual optimal target topics or future scenarios based on the user's past data. In terms of technical effectiveness, this proposal department differs from traditional static proposals or those relying on human experience rules. Through AI, it automatically selects the optimal proposal method based on historical pattern analysis, significantly improving proposal accuracy, accelerating anomaly detection, and enabling highly personalized user responses and decision support. Applicable areas include medical rehabilitation support systems, proposal support for people with disabilities, daily business report proposals, educational learning record proposals, and historical customer support proposals, among others.

[0057] The proposal department can customize proposal content based on the user's current physical condition during the proposal process. Utilizing generative AI, the department tailors proposal content based on the user's current physical condition. Physical condition includes, but is not limited to, health status and athletic ability. For example, the department can provide simplified proposals when the user is fatigued, detailed proposals when the user is relaxed, and quick proposals when the user is stressed. Thus, the department can customize proposal content based on the user's current physical condition, providing the optimal proposal. Specifically, the department acquires multi-dimensional numerical vectors (e.g., 128-dimensional) of user input, including health status data (e.g., walking distance, muscle strength measurements, fatigue score, pain status), athletic ability data (e.g., joint range of motion, activity level, balance score), and psychological state data (e.g., stress level, motivation, mood score). This data is then input into the generative AI (e.g., a multimodal large-scale language model) to classify (e.g., fatigue, relaxation, stress) or rate (e.g., fatigue 0.7, stress 0.8). Input examples include "Muscle strength: 25kg, Fatigue level: 0.8, Stress level: 0.6" and "Mood: Relaxed, no pain." Based on the AI ​​output, this proposal department automatically adjusts the number and detail of proposal content, the length of explanatory text, and the types of options. For example, when fatigue is high, only minimal content such as "Today's rehabilitation progress proposal" or "Proposal only regarding pain" is displayed; when relaxed, multiple contents such as "Detailed rehabilitation content," "Changes in living environment," and "Details of psychological state" are displayed. When stress is high, to support rapid decision-making, the range of options is narrowed and key points are emphasized. Output examples include "Number of proposal items: 2, Content: Simple rehabilitation" and "Number of proposal items: 8, Content: Detailed rehabilitation + life improvement." In terms of follow-up processing, the user's proposal selection rate, implementation rate, required time, and satisfaction are continuously monitored for AI model parameter optimization and proposal content design improvement. In terms of technical effectiveness, this proposal department differs from traditional unified proposals or subjective human judgment. By using AI to achieve multi-dimensional state estimation and real-time proposal content optimization, it can significantly reduce the user's burden, improve proposal efficiency and accuracy, and increase user satisfaction. It is applicable to various fields, including medical rehabilitation support systems, proposal support for people with disabilities, stress management applications, educational proposal systems, and daily business proposals.

[0058] The proposal department can infer users' emotions and determine the priority of proposals based on these inferred emotions. The department utilizes generative AI to infer user emotions and prioritizes proposals accordingly. User emotions include, but are not limited to, anxiety, relaxation, and unease. For example, the department can prioritize important proposals when a user is anxious, provide detailed proposals with more options when the user is relaxed, and prioritize reassuring proposals when the user is uneasy. Thus, the department can prioritize proposals based on user emotions, thereby displaying important proposals first. Emotion inference can be achieved through emotion engines or generative AI, among other emotion inference functions. Generative AI can be text-generated AI (such as LLM) or multimodal generative AI, but is not limited to these. Specifically, the proposal department can receive multimodal data such as natural language text input by the user (e.g., "I'm in a hurry today," "I'm a little uneasy"), voice data (e.g., speech rate or tone), and facial images (e.g., facial images obtained through a camera). This proposal department preprocesses the data, performing word segmentation and vectorization on the text, feature extraction on the speech, and facial feature point extraction on the images, before inputting them into a multimodal AI model (e.g., Transformer-based). The AI ​​model outputs an emotion classification (e.g., anxiety, relaxation, unease) or an emotion score (e.g., anxiety level 0.8, unease level 0.6). Output examples include "Emotion: Anxiety, Score 0.85" and "Emotion: Relaxation, Score 0.70." Based on these outputs, the proposal department automatically determines the priority of proposal items. For example, when the anxiety level is high, "proposals with high urgency (e.g., whether there is pain, recovery progress)" are placed at the top; when the relaxation level is high, "detailed proposals (e.g., living environment, psychological state)" are added; and when the unease level is high, "proposals that bring a sense of security (e.g., support system, consultation window)" are prioritized. In subsequent processing, user proposal selection, implementation rate, required time, and satisfaction are continuously monitored for AI model parameter optimization and proposal item design improvement. In terms of technical effectiveness, this proposal department differs from traditional static proposals or human-based priority settings. It utilizes AI to achieve multi-dimensional emotional inference and real-time priority control of proposal projects, thereby improving proposal efficiency, reducing user burden, quickly obtaining important information, and increasing user satisfaction. Applicable areas include medical rehabilitation support systems, proposal support for people with disabilities, stress management applications, educational proposal systems, and customer support proposals, among others.

[0059] The proposal department can consider users' geographical location information when submitting proposals, prioritizing proposals with high relevance. The proposal department utilizes generative AI to consider users' geographical location information during the proposal process, prioritizing proposals with high relevance. Geographical location information includes, but is not limited to, GPS data and location information services. For example, the proposal department prioritizes medical-related proposals when the user is in a hospital, rehabilitation-related proposals when the user is at home, and mobility-related proposals when the user is out and about. Thus, the proposal department can prioritize proposals with high relevance by considering users' geographical location information. Specifically, this proposal department acquires in real-time GPS coordinate data (e.g., latitude 35.6895, longitude 139.6917, accuracy 5m) collected from user terminals (e.g., smartphones, tablets, wearable devices), Wi-Fi / Bluetooth-based location information, and facility names and address information obtained from location information service APIs, and inputs this data as a multi-dimensional vector (e.g., location coordinates + facility category + time information, 64 dimensions). This proposal department inputs these geographic location vectors into an AI (e.g., a multimodal Transformer model) and compares them with historical databases or facility attribute databases (e.g., labeled data for hospitals, rehabilitation facilities, homes, public transportation, and commercial facilities). Input examples include "Location: Inside a hospital, Time: 10:00 AM," "Location: At home, Time: 8:00 PM," and "Location: Inside a station, Time: 3:00 PM." The AI ​​calculates the similarity between the input location vectors and the facility attribute database (e.g., cosine similarity, distance score) and automatically determines the most relevant scenario. Based on the determination results, this proposal department automatically switches the priority and display format of the proposal content (e.g., medical proposals first, rehabilitation proposals first, and mobility proposals first). Output examples include "Priority proposal: Treatment content, medication status, next visit plan (when staying in the hospital)," "Priority proposal: Rehabilitation progress, daily living activities, family support status (when staying at home)," and "Priority proposal: Mode of travel, travel distance, purpose of going out (when going out)," etc. In terms of follow-up processing, feedback data such as the actual content of user-selected and implemented proposals, completion rates, and required time are accumulated for continuous learning and optimization of the AI ​​model. Regarding technical effectiveness, this proposal department differs from traditional static proposals or project suggestions based on human experience. Through AI-powered real-time geolocation estimation and multi-dimensional data analysis, it can automatically suggest proposal content most suitable for the user's current location and usage scenario, thereby improving proposal efficiency, reducing erroneous proposals, alleviating user burden, and increasing system utilization. Applicable areas include medical rehabilitation support systems, home-based care support, outing support applications, proposal assistance for people with disabilities, daily business report proposals, and mobile health management applications, among others.

[0060] The proposal department can analyze users' social media activities during the proposal process to generate relevant proposals. The department utilizes generative AI to analyze users' social media activities during the proposal process, and these activities include, but are not limited to, content posting and activity frequency. For example, the proposal department can generate rehabilitation-related proposals when users post rehabilitation-related content on social media; it can also generate career-related proposals when users post career-related content; and it can also generate health-related proposals when users post health-related content. Thus, the proposal department can generate relevant proposals by analyzing users' social media activities. Specifically, within the scope of user permission, the proposal department collects posting data from social media APIs in time series (e.g., text, images, videos, posting time, number of likes, number of comments, etc.), and uses a natural language processing engine to segment and vectorize the posted content (e.g., 512-dimensional vectorization for each post). Furthermore, to categorize posting frequency and type (e.g., rehabilitation, career, health, interests, family, travel, etc.), it also utilizes large-scale Transformer-based language models or multimodal models for content semantic analysis and sentiment analysis (e.g., positive, negative, neutral). Input examples include "2024-06-01 'Striving for recovery'", "2024-06-02 'Taking on a new job'", and "2024-06-03 'Recently feeling very healthy'". Based on these analysis results, the proposal department infers the user's areas of interest and current life status, prioritizing the display of highly relevant proposals (such as: recovery progress, career aspirations, health status, changes in living environment, etc.) on the input interface. Output examples include "Recommended proposals: recovery progress, exercise content", "Recommended proposals: career aspirations, work style", and "Recommended proposals: health status, medication status", etc. In addition, the proposal department also considers the posting frequency and time period pattern (such as: more health-related posts at night), automatically adjusting the display timing of proposal content and auxiliary functions (such as: auto-completion, proposal candidate suggestions). In terms of follow-up processing, feedback data such as the actual content of proposals selected and implemented by users, completion rate, time required, and satisfaction are accumulated for the continuous learning and optimization of the AI ​​model. In terms of technical effectiveness, this proposal department differs from traditional static proposals or proposal support based on human experience. It utilizes AI to automate and expedite social media activity analysis and the creation of optimal individual proposals, thereby improving proposal efficiency, reducing erroneous proposals, alleviating user burden, and increasing system utilization. Applicable areas include medical rehabilitation support systems, proposal assistance for people with disabilities, daily business report proposals, educational learning record proposals, customer support proposals, and personal health management applications, among others.

[0061] The simulation department can improve simulation accuracy by referencing users' past data during simulation. The simulation department utilizes generated AI to improve simulation accuracy by referencing users' past data during simulation. Past data includes, but is not limited to, historical data and log data. For example, the simulation department can refer to users' past rehabilitation data to create simulations best suited to the current situation; it can also refer to users' past health data to improve simulation accuracy; and it can refer to users' past occupational data to create simulations about their post-reinstatement career. Thus, by referring to users' past data, the simulation department can improve simulation accuracy. Specifically, this simulation department refers to a structured database of rehabilitation history data (e.g., rehabilitation implementation date, exercise content, progress, achievement rate), health data (e.g., body temperature, blood pressure, heart rate, muscle strength measurements), and occupational data (e.g., past occupations, job content, work style, reinstatement history) recorded for each user in a time series. This simulation department inputs this historical data as time series vectors (e.g., each historical record is vectorized into 128 dimensions, with the most recent 30 records forming an array) to generate AI. Input examples include "2024-06-01 Rehabilitation progress 60%", "2024-06-02 Muscle strength measurement 28kg", and "2024-06-03 Occupation: Administrative". This simulation department utilizes a Transformer-based time series analysis model or LSTM network to automatically extract patterns from historical data and perform clustering, learning the user's recovery trend, changes in health status, and changes in occupational adaptability. AI output includes future scenario predictions (e.g., predicted muscle strength recovery in two weeks, return-to-work adaptability score), comparison charts with past data, and selection of the optimal simulation method (e.g., optimization of progress prediction model parameters, automatic adjustment of outlier detection threshold). Output examples include "Simulation result: Expected recovery 80%" and "Return-to-work adaptability: Office-related adaptability 85%". Based on these outputs, this simulation department displays the expected achievement of future occupational, life scenarios, or rehabilitation plans to the user in dashboard format. In subsequent processing, the user's selected scenarios or feedback are fed back to the simulation department or database department, and the AI ​​model continues to learn and optimize. In terms of technical effectiveness, this simulation department differs from traditional static simulations or those relying on human experience and rules of thumb. Through AI, it achieves historical pattern analysis and automatic selection of individual optimal simulation methods, significantly improving simulation accuracy, accelerating anomaly detection, and enabling highly personalized user responses and decision support. Applicable areas include medical rehabilitation support systems, simulation support for people with disabilities, business reinstatement scenario generation, vocational simulation for education, and customer support reinstatement support, among others.

[0062] The simulation department can customize simulation methods based on the user's current physical condition during simulation. Utilizing a generative AI, the simulation department tailors simulation methods according to the user's current physical condition. Physical condition includes, but is not limited to, health status and motor ability. For example, the simulation department can use a simplified simulation method when the user is fatigued, a detailed simulation method when the user is relaxed, and a rapid simulation method when the user is stressed. Thus, the simulation department can customize simulation methods based on the user's current physical condition, providing the user with optimal simulation results. Specifically, this simulation department acquires multi-dimensional numerical vectors (e.g., 128-dimensional) of user input, including health status data (e.g., walking distance, muscle strength measurement, fatigue score, pain status), motor ability data (e.g., joint range of motion, activity level, balance score), and psychological status data (e.g., stress level, motivation, mood score). This data is then input into a generative AI (e.g., a multimodal large-scale language model) to classify (e.g., fatigue, relaxation, stress) or rate (e.g., fatigue 0.7, stress 0.8). Input examples include "Muscle strength: 25kg, Fatigue level: 0.8, Stress level: 0.6" and "Emotion: Relaxed, no pain." Based on the AI ​​output, this simulation department automatically adjusts the selection of simulation methods and parameters (such as the number of selected features, the number of scenario branches, the batch size, and threshold settings). For example, when fatigue is high, only the main features are extracted, and a simplified scenario branching or progress prediction model is used to reduce computational load; in a relaxed state, multi-stage simulation, detailed anomaly detection, and multi-model integrated analysis are performed to maximize accuracy; when stress is high, speed is prioritized, and a high-real-time inference model is selected. Output examples include "Simulation method: simplified branching, 5 main features" and "Simulation method: detailed multi-stage analysis, 20 features." In subsequent processing, the simulation results are linked with the user or proposal department to automatically adjust the target topic or future scenario based on the user's physical condition. In terms of technical effectiveness, this simulation department differs from traditional unified simulation processes or subjective human judgment. By using AI to achieve multi-dimensional state estimation and real-time simulation method optimization, it can significantly reduce the user's burden, improve simulation efficiency and accuracy, and increase user satisfaction. It is applicable to various fields, including medical rehabilitation support systems, simulation support for people with disabilities, stress management applications, educational scenario generation, and business resumption simulation.

[0063] The simulation department can infer the user's emotions and determine the priority of simulations based on the inferred emotions. The simulation department uses generative AI to infer the user's emotions and determines the priority of simulations based on the inferred emotions. User emotions include, but are not limited to, anxiety, relaxation, and unease. For example, the simulation department can prioritize important simulations when the user is anxious, provide detailed simulations with more options when the user is relaxed, and prioritize simulations that bring a sense of security when the user is uneasy. Thus, the simulation department can determine the priority of simulations based on the user's emotions, thereby prioritizing important simulations. Emotion inference can be achieved through emotion inference functions such as emotion engines or generative AI. Generative AI can be text-generated AI (such as LLM) or multimodal generative AI, but is not limited to these. Specifically, this simulation department can receive multimodal data such as natural language text input by the user (e.g., "I'm in a hurry today," "I'm a little uneasy"), speech data (e.g., speech rate or tone), and facial expression images (e.g., facial images obtained through a camera). This simulation department preprocesses the data, performing word segmentation and vectorization on text, feature extraction on speech, and facial feature point extraction on images, before inputting them into a multimodal AI model (e.g., Transformer-based). The AI ​​model outputs emotion classifications (e.g., anxiety, relaxation, unease) or emotion scores (e.g., anxiety level 0.8, unease level 0.6). Output examples include "Emotion: Anxiety, Score 0.85" and "Emotion: Relaxation, Score 0.70." Based on these outputs, the simulation department automatically determines the priority of simulation projects. For example, when anxiety is high, simulations with high urgency (e.g., pain, recovery progress) are placed at the top; when relaxation is high, detailed simulations (e.g., living environment, psychological state) are added; and when unease is high, simulations that bring a sense of security (e.g., support system, consultation window) are prioritized. In subsequent processing, user simulation selection, implementation rate, required time, and satisfaction are continuously monitored for AI model parameter optimization and simulation project design improvement. In terms of technical effectiveness, this simulation department differs from traditional static simulations or human-based rule-of-fact prioritization. It utilizes AI to achieve multi-dimensional emotional inference and real-time project priority control, thereby improving simulation efficiency, reducing user burden, quickly acquiring crucial information, and enhancing user satisfaction. Applicable areas include medical rehabilitation support systems, simulation support for people with disabilities, stress management applications, educational scenario generation, and customer support simulation, among others.

[0064] The simulation department can consider the user's geographical location information during simulation, prioritizing simulations with high relevance. The simulation department utilizes generative AI to consider the user's geographical location information during simulation, prioritizing simulations with high relevance. Geographical location information includes, but is not limited to, GPS data and location information services. For example, the simulation department prioritizes medical-related simulations when the user is in a hospital, rehabilitation-related simulations when the user is at home, and mobility-related simulations when the user is out and about. Thus, the simulation department can prioritize simulations with high relevance by considering the user's geographical location information. Specifically, this simulation department acquires in real-time GPS coordinate data (e.g., latitude 35.6895, longitude 139.6917, accuracy 5m) collected by the user's terminal (e.g., smartphone, tablet, wearable device), Wi-Fi / Bluetooth-based location information, facility names and address information obtained from location information service APIs, etc., and inputs them as a multi-dimensional vector (e.g., location coordinates + facility category + time information, 64 dimensions). This simulation department inputs these geographic location vectors into an AI (e.g., a multimodal Transformer model) and compares them with historical databases or facility attribute databases (e.g., labeled data from hospitals, rehabilitation facilities, homes, public transportation, and commercial facilities). Input examples include "Location: Inside the hospital, Time: 10:00 AM," "Location: At home, Time: 8:00 PM," and "Location: Inside the station, Time: 3:00 PM." The generated AI calculates the similarity between the input location vectors and the facility attribute database (e.g., cosine similarity, distance score), automatically determining the most relevant scene. Based on the determination results, this simulation department automatically switches the priority order and display format of the simulation content (e.g., medical simulation priority, rehabilitation simulation priority, and mobility simulation priority). Output examples include "Priority simulation: Treatment content, medication status, next visit plan (when staying in the hospital)," "Priority simulation: Rehabilitation progress, daily living activities, family support status (when staying at home)," and "Priority simulation: Mode of travel, travel distance, purpose of going out (when going out)," etc. In terms of follow-up processing, feedback data such as the actual simulation content selected and implemented by users, completion rate, and required time are accumulated for continuous learning and optimization of the AI ​​model. In terms of technical effectiveness, this simulation department differs from traditional static simulations or project prompts based on human experience rules. Through AI-powered real-time geolocation estimation and multi-dimensional data analysis, it can automatically suggest simulation content most suitable for the user's current location and usage scenario, thereby improving simulation efficiency, reducing erroneous simulations, alleviating user burden, and increasing system utilization. Applicable areas include medical rehabilitation support systems, home-based care support, outing support applications, simulation assistance for people with disabilities, business resumption simulations, and mobile health management applications, among others.

[0065] The simulation department can analyze users' social media activities during simulation and perform relevant simulations. The simulation department utilizes generative AI to analyze users' social media activities during simulation and perform relevant simulations. Social media activities include, but are not limited to, content posting and activity frequency. For example, the simulation department can perform rehabilitation-related simulations when users post rehabilitation-related content on social media; it can also perform career-related simulations when users post career-related content; and it can also perform health-related simulations when users post health-related content. Thus, the simulation department can perform relevant simulations by analyzing users' social media activities. Specifically, within the scope of user permission, this simulation department collects posting data from social media APIs in time series (e.g., text, images, videos, posting time, number of likes, number of comments, etc.), and uses a natural language processing engine to segment and vectorize the posted content (e.g., 512-dimensional vectorization for each post). Furthermore, to classify posting frequency and categories (e.g., rehabilitation, career, health, interests, family, travel, etc.), it also utilizes large-scale language models or multimodal models based on Transformer for content semantic analysis and sentiment analysis (e.g., positive, negative, neutral). Input examples include "2024-06-01 'Striving for recovery'", "2024-06-02 'Taking on a new job'", and "2024-06-03 'Recently feeling very healthy'". Based on these analysis results, the simulation department infers the user's areas of interest and current life status, prioritizing simulations with high relevance (such as: recovery progress, career intentions, health status, changes in living environment, etc.). AI outputs include "Recommended simulations: recovery progress prediction, exercise content scenarios", "Recommended simulations: career resumption scenarios, work form simulation", and "Recommended simulations: health status prediction, medication management scenarios", etc. In addition, it also considers the frequency and time period of publication (such as: more health-related publications at night), automatically adjusting the display timing of simulation content and auxiliary functions (such as: auto-completion, scenario candidate prompts). In terms of subsequent processing, feedback data such as the actual simulation content selected and implemented by the user, completion rate, time required, and satisfaction are accumulated for the continuous learning and optimization of the AI ​​model. In terms of technical effectiveness, this simulation department differs from traditional static simulations or scenario prompts based on human experience rules. It utilizes AI to automate and accelerate the analysis of social media activities and the creation of optimal simulation content, thereby improving simulation efficiency, reducing erroneous simulations, alleviating user burden, and increasing system utilization. Applicable areas include medical rehabilitation support systems, simulation assistance for people with disabilities, business resumption scenario generation, educational learning record simulation, customer support scenario generation, and personal health management applications, among others.

[0066] The monitoring department can infer users' emotions and adjust the monitoring frequency based on the inferred emotions. The monitoring department uses generative AI to infer users' emotions and adjusts the monitoring frequency accordingly. User emotions include, but are not limited to, anxiety, relaxation, and restlessness. For example, when a user feels anxious, the monitoring department monitors frequently to provide reassurance; when a user is relaxed, the monitoring frequency can be reduced to alleviate stress; and when a user is anxious, monitoring can be conducted quickly and results provided rapidly. Thus, the monitoring department can adjust the monitoring frequency based on the user's emotions, thereby providing reassurance. Emotion inference can be achieved through emotion inference functions such as emotion engines or generative AI. Generative AI can be text-generated AI (such as LLM) or multimodal generative AI, but is not limited to these. Specifically, this monitoring department can receive multimodal data such as natural language text input by the user (e.g., "I've been feeling uneasy lately," "I'm calm today"), voice data (e.g., speech rate or tone), and facial expression images (e.g., facial images obtained through a camera). This monitoring department preprocesses the data, performing word segmentation and vectorization (e.g., 512-dimensional vectors) on the text, MFCC feature extraction on the speech, and facial feature point extraction (e.g., 68-point markers) on the images. Input examples include "speech text of 'I'm very uneasy today'", "speech data with a low tone", and "facial image with a smile". This monitoring department inputs these multimodal feature vectors into a large language model equipped with a multimodal Transformer or self-attention mechanism, outputting sentiment classification (e.g., labels such as uneasy, relaxed, anxious) or sentiment scores (e.g., probability distribution of 0.8 for uneasy and 0.2 for relaxed). Output examples include "sentiment label: uneasy, score 0.75" and "sentiment label: relaxed, score 0.60". Based on these output results, this monitoring department automatically adjusts the monitoring frequency. For example, when the uneasy level is high, "monitor health status every 5 minutes" and "monitor vital signs in real time" are used; when the relaxed level is high, "simple monitoring once a day" and "report only once a week" are used; and when the anxious level is high, "instant monitoring" and "instant result notification" are used. In terms of follow-up processing, monitoring results and user feedback (such as reassurance feedback, re-monitoring requests, and changes in stress scores) are continuously collected for AI model parameter optimization and monitoring design improvement. Regarding technical effectiveness, this monitoring department differs from traditional unified monitoring plans or subjective human judgment by using AI to achieve multi-dimensional emotional inference and real-time monitoring frequency optimization. This significantly reduces user psychological burden, enhances reassurance, improves monitoring efficiency, and reduces false positives. Applicable areas include medical rehabilitation support systems, health monitoring for people with disabilities, stress management applications, educational monitoring systems, and customer support monitoring, among others.

[0067] The monitoring department can improve monitoring accuracy by referencing users' past data during monitoring. The monitoring department utilizes generated AI to improve monitoring accuracy by referencing users' past data. Past data includes, but is not limited to, historical data and log data. For example, the monitoring department can refer to users' past rehabilitation data to make monitoring decisions most suitable for their current situation; it can also refer to users' past health data to improve monitoring accuracy; and it can also refer to users' past occupational data to make occupational monitoring decisions regarding their return to work. Thus, by referring to users' past data, the monitoring department can improve monitoring accuracy. Specifically, this monitoring department refers to a structured database of rehabilitation history data (e.g., rehabilitation implementation date, exercise content, progress, achievement rate), health data (e.g., body temperature, blood pressure, heart rate, muscle strength measurements), and occupational data (e.g., past occupation, job content, work style, return to work history) recorded for each user in a time series format. This monitoring department inputs this historical data as a time series vector (e.g., each historical record is vectorized into 128 dimensions, with the most recent 30 records forming an array) into the generated AI. Input examples include "2024-06-01 Rehabilitation progress 60%", "2024-06-02 Muscle strength measurement 28kg", "2024-06-03 Occupation: Administrative". This monitoring department utilizes a Transformer-based time series analysis model or LSTM network to automatically extract patterns from historical data and perform clustering, learning the user's recovery trend, changes in health status, and changes in occupational adaptability. AI outputs include selecting the most suitable monitoring method for the current situation (e.g., optimization of progress prediction model parameters, automatic adjustment of outlier detection thresholds), comparison charts with past data, and future predictions (e.g., predicted muscle strength recovery in two weeks). Output examples include "Progress prediction: Expected to reach 80%", "Health status: Stable", "Occupational adaptability: Office-related adaptability 85%". Based on these outputs, this monitoring department automatically adjusts monitoring frequency, monitoring items, and alarm thresholds. For subsequent processing, monitoring results and user feedback (e.g., responses to anomaly detection, feedback, and re-monitoring requests) are continuously collected for AI model parameter optimization and monitoring design improvement. In terms of technical effectiveness, this monitoring department differs from traditional static monitoring or judgments relying on human experience. Through AI, it achieves historical pattern analysis and automatic selection of individual optimal monitoring methods, significantly improving monitoring accuracy, accelerating anomaly detection, enabling personalized user responses, and enhancing peace of mind. Applicable areas include medical rehabilitation support systems, disability monitoring support, daily business monitoring, educational learning record monitoring, and customer support historical monitoring, among others.

[0068] The monitoring department can infer users' emotions and determine the priority of monitoring based on the inferred emotions. The monitoring department uses generative AI to infer users' emotions and determines the priority of monitoring based on the inferred emotions. User emotions include, but are not limited to, anxiety, relaxation, and unease. For example, the monitoring department can prioritize important monitoring when users are anxious, provide detailed monitoring and additional options when users are relaxed, and prioritize reassuring monitoring when users feel uneasy. Thus, the monitoring department can determine the priority of monitoring based on users' emotions, thereby prioritizing important monitoring. Emotion inference can be achieved through emotion inference functions such as emotion engines or generative AI. Generative AI can be text-generated AI (such as LLM) or multimodal generative AI, but is not limited to these. Specifically, this monitoring department can receive multimodal data such as natural language text input by users (e.g., "I'm in a hurry today," "I'm a little uneasy"), voice data (e.g., speech rate or tone), and facial expression images (e.g., facial images obtained through a camera). This monitoring department preprocesses the data, performing word segmentation and vectorization on text, feature extraction on speech, and facial feature point extraction on images, before inputting them into a multimodal AI model (e.g., Transformer-based). The AI ​​model outputs an emotion classification (e.g., anxiety, relaxation, unease) or an emotion score (e.g., anxiety level 0.8, unease level 0.6). Output examples include "Emotion: Anxiety, Score 0.85" and "Emotion: Relaxation, Score 0.70." Based on these outputs, the monitoring department automatically determines the priority of monitoring items. For example, when anxiety is high, "high-urgency monitoring (e.g., vital signs, pain)" is placed at the top; when relaxation is high, "detailed monitoring (e.g., living environment, psychological state)" is added; and when unease is high, "monitoring that brings reassurance (e.g., support system, consultation window)" is prioritized. In subsequent processing, the user's monitoring selections, completion rates, required time, and level of reassurance are continuously monitored for AI model parameter optimization and monitoring item design improvement. In terms of technical effectiveness, this monitoring department differs from traditional static monitoring or human-based priority setting. It utilizes AI to achieve multi-dimensional emotional inference and real-time priority control of monitoring items, improving monitoring efficiency, reducing user burden, quickly obtaining important information, and enhancing peace of mind. Applicable areas include medical rehabilitation support systems, disability monitoring and support, stress management applications, educational monitoring systems, and customer support monitoring, among others.

[0069] The monitoring department can consider users' geographical location information during monitoring, prioritizing highly relevant monitoring. The monitoring department utilizes generative AI to prioritize highly relevant monitoring based on users' geographical location information. Geographical location information includes, but is not limited to, GPS data and location information services. For example, the monitoring department prioritizes medical-related monitoring when the user is in a hospital, rehabilitation-related monitoring when the user is at home, and mobility-related monitoring when the user is out and about. Thus, the monitoring department can prioritize highly relevant monitoring by considering the user's geographical location information. Specifically, this monitoring department acquires in real-time GPS coordinate data (e.g., latitude 35.6895, longitude 139.6917, accuracy 5m) collected by user terminals (e.g., smartphones, tablets, wearable devices), Wi-Fi / Bluetooth-based location information, facility names and address information obtained from location information service APIs, etc., and inputs them as a multi-dimensional vector (e.g., location coordinates + facility category + time information, 64 dimensions). This monitoring department inputs these geographic location vectors into an AI (e.g., a multimodal Transformer model) and compares them with historical databases or facility attribute databases (e.g., labeled data from hospitals, rehabilitation facilities, homes, public transportation, and commercial facilities). Input examples include "Location: Inside a hospital, Time: 10:00 AM," "Location: At home, Time: 8:00 PM," and "Location: Inside a station, Time: 3:00 PM." The AI ​​calculates the similarity between the input location vectors and the facility attribute database (e.g., cosine similarity, distance score), automatically determining the most relevant monitoring scenario. Based on the determination result, this monitoring department automatically switches the priority order and monitoring method for the monitored data (e.g., prioritizing medical data, rehabilitation data, and mobile data). Output examples include "Monitoring object: Medical data, Priority 90%" and "Monitoring object: Rehabilitation data, Priority 80%." In subsequent processing, the monitoring results are linked with the user or the analysis department to automatically optimize individualized monitoring design and alarm generation based on the user's geographic location. In terms of technical effectiveness, this monitoring department differs from traditional static monitoring or judgment based on human experience. It utilizes AI to achieve real-time geolocation estimation and multi-dimensional data analysis, automating and improving monitoring accuracy based on the user's current location and usage scenario. This results in increased monitoring efficiency, reduced false detections, reduced user burden, and higher system utilization. Applicable areas include medical rehabilitation support systems, home-based care support, outing support applications, disability monitoring assistance, daily business monitoring, and mobile health management applications, among others.

[0070] The generation unit can infer the user's emotions and adjust the generation algorithm based on the inferred emotions. The generation unit utilizes generative AI to infer the user's emotions and adjusts the generation algorithm accordingly. User emotions include, but are not limited to, relaxation, anxiety, and unease. For example, the generation unit can perform detailed generation to improve accuracy when the user is relaxed, perform rapid generation and provide results quickly when the user is anxious, and provide reassuring results when the user feels uneasy. Thus, the generation unit can adjust the generation algorithm based on the user's emotions, thereby improving the accuracy of the generated results. Emotion inference can be achieved through emotion inference functions such as emotion engines or generative AI. Generative AI can be text generation AI (such as LLM) or multimodal generation AI, but is not limited to these. Specifically, this generation unit can receive multimodal data such as natural language text input by the user (e.g., "Today is peaceful," "Somewhat uneasy"), speech data (e.g., speech rate or tone), and facial expression images (e.g., facial images acquired through a camera). This generation unit preprocesses the data, performing word segmentation and vectorization (e.g., 512-dimensional vectors) on the text, MFCC feature extraction on the speech, and facial feature point extraction (e.g., 68-dot markers) on the images. Input examples include speech text such as "'I'm very relaxed today'", "speech data with a high pitch", and "facial images with a smile". This generation unit inputs these multimodal feature vectors into a large language model with a multimodal Transformer or self-attention mechanism, outputting sentiment classifications (e.g., labels such as relaxed, anxious, and uneasy) or sentiment scores (e.g., probability distributions such as relaxation 0.7, anxiety 0.2, and uneasy 0.1). Output examples include "sentiment label: relaxed, score 0.75" and "sentiment label: anxious, score 0.60". Based on these output results, this generation unit automatically adjusts the parameters of the generation algorithm (e.g., generation depth, number of feature selections, batch size, and threshold setting). For example, when the user is highly relaxed, detailed feature extraction and multi-stage generation are performed to maximize generation accuracy; when the user is highly anxious, the number of features is reduced and the number of inferences is decreased to prioritize speed; when the user is highly uneasy, the interpretability and reassurance of the generated results are emphasized, highlighting interpretable features and comparisons with past successful cases. In terms of subsequent processing, the generated results are linked with the proposal department or simulation department to automatically adjust the target topic or future scenario based on the user's emotional state. In terms of technical effectiveness, this generation department differs from traditional unified generation processes or subjective human judgment. It uses AI to achieve multi-dimensional emotional inference and real-time generation algorithm optimization, which can improve generation accuracy, optimize processing speed, reduce the user's psychological burden, and enhance interpretability. Applicable areas include medical rehabilitation support systems, generation support for people with disabilities, stress management applications, educational generation systems, and customer support generation, among others.

[0071] The generation department can improve generation accuracy by referencing users' past data during the generation process. The generation department utilizes generation AI to improve accuracy by referencing users' past data during generation. Past data includes, but is not limited to, historical data and log data. For example, the generation department can refer to users' past rehabilitation data to generate data most suitable for their current situation; it can also refer to users' past health data to improve generation accuracy; and it can refer to users' past occupational data to generate information about their post-reinstatement career. Thus, by referring to users' past data, the generation department can improve generation accuracy. Specifically, this generation department refers to a structured database of each user's time-series recorded rehabilitation history data (e.g., rehabilitation implementation date, exercise content, progress, achievement rate), health data (e.g., body temperature, blood pressure, heart rate, muscle strength measurements), and occupational data (e.g., past occupations, job content, work style, reinstatement history), etc. This generation department inputs this historical data as a time-series vector (e.g., each historical record is vectorized into 128 dimensions, with the most recent 30 records forming an array) into the generation AI. Input examples include "2024-06-01 Rehabilitation progress 60%", "2024-06-02 Muscle strength measurement 28kg", "2024-06-03 Occupation: Administrative position", etc. This generation department utilizes a Transformer-based time series analysis model or LSTM network to automatically extract patterns from historical data and perform clustering, learning the user's recovery trend, changes in health status, and changes in occupational adaptability. AI output includes selecting the most suitable generation method for the current situation (e.g., optimization of progress prediction model parameters, automatic adjustment of outlier detection thresholds), a comparison chart with past data, and future predictions (e.g., predicted muscle strength recovery in two weeks). Output examples include "Generated result: Exercise A 3 times a week, achieving 80% of the expected result", "Generated result: Health management proposal, increase blood pressure measurement frequency", "Generated result: Occupational return-to-work scenario, office-related adaptability 85%", etc. In subsequent processing, the generated results will be linked with the proposal department or simulation department to generate individual optimal target topics or future scenarios based on the user's past data. In terms of technical effectiveness, this generation system differs from traditional static generation or reliance on human experience and rules of thumb. By utilizing AI to analyze historical patterns and automatically select the optimal generation method for each individual case, it significantly improves generation accuracy, accelerates anomaly detection, and enables highly personalized user responses and decision support. Applicable areas include medical rehabilitation support systems, disability support systems, daily business report generation, educational learning record generation, and customer support history generation, among others.

[0072] The generation unit can infer the user's emotions and determine the priority of generation based on the inferred emotions. The generation unit uses generative AI to infer the user's emotions and determines the priority of generation based on the inferred emotions. User emotions include, but are not limited to, anxiety, relaxation, and unease. For example, the generation unit prioritizes important generation when the user is anxious, provides detailed generation with more options when the user is relaxed, and prioritizes generation that brings a sense of security when the user is uneasy. Thus, the generation unit can determine the priority of generation based on the user's emotions, thereby prioritizing important generation. Emotion inference can be achieved through emotion inference functions such as emotion engines or generative AI. Generative AI can be text generation AI (such as LLM) or multimodal generation AI, but is not limited to these. Specifically, this generation unit can receive multimodal data such as natural language text input by the user (e.g., "I'm in a hurry today," "I'm a little uneasy"), voice data (e.g., speech rate or tone), and facial expression images (e.g., facial images obtained through a camera). This generation department preprocesses the data, performing word segmentation and vectorization on text, feature extraction on speech, and facial feature point extraction on images, before inputting them into a multimodal AI model (e.g., Transformer-based). The AI ​​model outputs an emotion classification (e.g., anxiety, relaxation, unease) or an emotion score (e.g., anxiety level 0.8, unease level 0.6). Output examples include "Emotion: Anxiety, Score 0.85" and "Emotion: Relaxation, Score 0.70." Based on these outputs, this generation department automatically determines the priority of generated items. For example, when anxiety is high, "high-urgency generation (e.g., whether there is pain, recovery progress)" is placed at the top; when relaxation is high, "detailed generation (e.g., living environment, psychological state)" is added; and when unease is high, "reassuring generation (e.g., support system, consultation window)" is prioritized. In subsequent processing, user generation choices, implementation rate, required time, and satisfaction are continuously monitored for AI model parameter optimization and generation item design improvement. In terms of technical effectiveness, this generation system differs from traditional static generation or human-based priority setting. It utilizes AI to achieve multi-dimensional emotional inference and real-time project priority control, improving generation efficiency, reducing user burden, quickly acquiring important information, and enhancing user satisfaction. Applicable areas include medical rehabilitation support systems, generation support for people with disabilities, stress management applications, educational generation systems, and customer support generation, among others.

[0073] The generation unit can consider the user's geographical location information during generation, prioritizing highly relevant generation. The generation unit utilizes generative AI to consider the user's geographical location information during generation, prioritizing highly relevant generation. Geographical location information includes, but is not limited to, GPS data and location information services. For example, the generation unit prioritizes medical-related generation when the user is in a hospital, rehabilitation-related generation when the user is at home, and mobility-related generation when the user is out and about. Thus, the generation unit can prioritize highly relevant generation by considering the user's geographical location information. Specifically, this generation unit acquires in real-time GPS coordinate data (e.g., latitude 35.6895, longitude 139.6917, accuracy 5m) collected by the user's terminal (e.g., smartphone, tablet, wearable device), Wi-Fi / Bluetooth-based location information, facility names and address information obtained from location information service APIs, etc., and inputs them as a multi-dimensional vector (e.g., location coordinates + facility category + time information, 64 dimensions). This generation unit inputs these geographic location vectors into the generation AI (e.g., a multimodal Transformer model) and compares them with historical databases or facility attribute databases (e.g., labeled data from hospitals, rehabilitation facilities, homes, public transportation, and commercial facilities). Input examples include "Location: Inside the hospital, Time: 10:00 AM," "Location: At home, Time: 8:00 PM," and "Location: Inside the station, Time: 3:00 PM." The generation AI calculates the similarity between the input location vector and the facility attribute database (e.g., cosine similarity, distance score) and automatically determines the most relevant scene. Based on the determination result, this generation unit automatically switches the priority order and display format of the generated content (e.g., medical generation priority, rehabilitation generation priority, and mobility generation priority). Output examples include "Prioritized generation: Treatment content, medication status, next visit plan (when staying in the hospital)," "Prioritized generation: Rehabilitation progress, daily living activities, family support status (when staying at home)," and "Prioritized generation: Mode of travel, travel distance, purpose of going out (when going out)," etc. In terms of post-processing, feedback data such as the actual content selected and implemented by users, completion rate, and time required are accumulated for continuous learning and optimization of the AI ​​model. Regarding technical effectiveness, this generation department differs from traditional static generation or manual rule-of-fact project suggestions. Through AI-powered real-time geolocation estimation and multi-dimensional data analysis, it can automatically suggest the most suitable content for the user's current location and usage scenario, thereby improving generation efficiency, reducing errors, alleviating user burden, and increasing system utilization. Applicable areas include medical rehabilitation support systems, home-based care support, outing support applications, disability assistance generation, daily business report generation, and mobile health management applications, among others.

[0074] The database department can infer users' emotions and adjust the database's search results based on these inferred emotions. The database department utilizes generative AI to infer users' emotions and adjusts the database's search results accordingly. User emotions include, but are not limited to, feelings of unease, relaxation, and anxiety. For example, when a user feels uneasy, the database department prioritizes displaying search results that evoke a sense of reassurance; when a user is relaxed, it can provide detailed search results with more options; and when a user is anxious, it can prioritize displaying important search results. Thus, the database department can adjust the database's search results based on users' emotions, thereby providing users with the optimal search results. Emotion inference can be achieved through emotion inference functions such as emotion engines or generative AI. Generative AI can be text generation AI (such as LLM) or multimodal generation AI, but is not limited to these.

[0075] The database department can improve update accuracy by referencing users' past data during database updates. This is achieved using generative AI. Past data includes, but is not limited to, historical data and log data. For example, the database department can refer to a user's past recovery data to update data most suitable for their current condition. Furthermore, it can refer to a user's past health data to improve update accuracy. Even further, it can refer to a user's past occupational data to update data related to their post-recovery occupation. Thus, by referencing users' past data, the database department can improve the accuracy of database updates.

[0076] The database department can infer users' emotions and determine the database search priority order based on the inferred user emotions. The database department utilizes generative AI to infer users' emotions and determine the database search priority order based on the inferred user emotions. User emotions include, but are not limited to, anxiety, relaxation, and unease. For example, when a user is anxious, the database department prioritizes displaying important search results. Furthermore, when a user is relaxed, the database department can provide detailed search results, increasing the number of options. Further, when a user is uneasy, the database department can also prioritize displaying search results that bring a sense of security. Thus, the database department can determine the database search priority order based on users' emotions, thereby prioritizing the display of important search results. Emotion inference can be achieved through emotion inference functions such as emotion engines or generative AI. Generative AI can be text generation AI (e.g., LLM) or multimodal generation AI, but is not limited to these.

[0077] When updating the database, the database department can consider the user's geographic location information and prioritize updating highly relevant data. The database department utilizes generative AI to prioritize updating data with high relevance based on the user's geographic location information during database updates. Geographic location information includes, but is not limited to, GPS data or location information services. For example, when a user is in a hospital, the database department prioritizes updating medical-related data. Furthermore, when a user is at home, the database department can prioritize updating rehabilitation-related data. Further, when a user is away from home, the database department can prioritize updating mobility-related data. Thus, by considering the user's geographic location information, the database department can prioritize updating highly relevant data.

[0078] The system involved in this embodiment is not limited to the example described above. For example, various modifications can be made as described below.

[0079] The receiving unit can infer the user's emotions and adjust the display of the input interface accordingly. For example, when the user is anxious, a calming color scheme can be provided to create a sense of security. Conversely, when the user is relaxed, a bright color scheme can be provided to make input more enjoyable. Furthermore, when the user is anxious, a simple and highly visible interface can be provided to allow for quick input. Thus, the receiving unit can adjust the display of the input interface based on the user's emotions, making the user's input experience more comfortable.

[0080] The monitoring department can not only monitor the user's recovery progress but also infer the user's emotions and adjust the monitoring frequency accordingly. For example, when the user feels anxious, frequent monitoring can provide reassurance. Conversely, when the user is relaxed, the monitoring frequency can be reduced to alleviate stress. Furthermore, when the user feels anxious, monitoring can be conducted quickly to provide rapid results. Thus, the monitoring department can adjust the monitoring frequency based on the user's emotions, thereby providing reassurance.

[0081] The generation unit can infer the user's emotions and adjust the generation algorithm accordingly. For example, when the user is relaxed, detailed generation is performed to improve accuracy. Furthermore, when the user is anxious, rapid generation can be performed to quickly provide results. Moreover, when the user is uneasy, reassuring results can be provided. Thus, the generation unit can adjust the generation algorithm based on the user's emotions, thereby improving the accuracy of the generated results.

[0082] The proposal department can infer users' emotions and adjust the way proposals are expressed accordingly. For example, when users are relaxed, detailed proposals can be made, offering more options. Conversely, when users are anxious, concise proposals can be made, quickly providing options. Furthermore, when users are uneasy, reassuring proposals can be made, reducing the number of options. Thus, the proposal department can adjust the way proposals are expressed based on users' emotions, resulting in proposals that are easy for users to understand.

[0083] The database department can infer a user's emotions and adjust the database search results accordingly. For example, when a user feels uneasy, search results that evoke a sense of reassurance are prioritized. Furthermore, when a user is relaxed, more detailed search results can be provided, increasing the number of options. Moreover, when a user feels anxious, important search results can be prioritized. Thus, the database department can adjust the database search results based on the user's emotions, thereby providing the user with the optimal search results.

[0084] The receiving unit can analyze a user's past input history and suggest appropriate input methods. For example, it can automatically display frequently entered information as candidate options. Furthermore, it can prioritize input methods previously used by the user (voice, text, etc.). Moreover, it can predict information used in specific time periods and propose suggestions based on the user's past input history. Thus, the receiving unit can suggest the optimal input method by analyzing the user's past input history.

[0085] During analysis, the analysis department can refer to the user's past data to improve accuracy. For example, it can refer to the user's past recovery data to perform analysis most suitable for the current situation. Furthermore, it can refer to the user's past health data to improve accuracy. Even further, it can refer to the user's past occupational data to perform analysis related to their post-recovery career. Therefore, by referring to the user's past data, the analysis department can improve the accuracy of its analysis.

[0086] When submitting proposals, the proposal department can refer to users' past data to improve the accuracy of the proposals. For example, by referring to users' past recovery data, they can create proposals most suitable for their current condition. Furthermore, they can refer to users' past health data to improve the accuracy of the proposals. Even further, they can refer to users' past occupational data to create proposals related to their post-recovery careers. Therefore, by referring to users' past data, the proposal department can improve the accuracy of its proposals.

[0087] During simulations, the simulation department can refer to the user's past data to improve simulation accuracy. For example, it can refer to the user's past rehabilitation data to conduct simulations best suited to the current situation. Furthermore, it can refer to the user's past health data to improve simulation accuracy. Even further, it can refer to the user's past occupational data to conduct simulations related to their post-recovery occupation. Thus, by referencing the user's past data, the simulation department can improve the accuracy of its simulations.

[0088] The monitoring department can improve the accuracy of monitoring by referring to the user's past data. For example, by referring to the user's past recovery data, monitoring can be tailored to the user's current condition. Furthermore, the accuracy can be improved by referring to the user's past health data. Even further, the accuracy can be improved by referring to the user's past occupational data to conduct monitoring related to their post-recovery career. Therefore, by referring to the user's past data, the monitoring department can improve the accuracy of its monitoring.

[0089] The following is a brief description of the processing flow of the implementation method.

[0090] Step 1: The receiving unit inputs the user's information. This includes factors such as health status, living environment, and mental state. The receiving unit inputs information about the user's hospitalization due to a fracture, rehabilitation progress, and post-recovery career and lifestyle goals.

[0091] Step 2: The parsing unit uses the generating AI to analyze the information input from the receiving unit. The parsing is based on data analysis techniques or algorithms. For example, the generating AI uses machine learning models or data generation algorithms to analyze the user's situation.

[0092] Step 3: The proposal department utilizes AI to generate target topics based on the information analyzed by the analysis department. Target topics may include, for example, rehabilitation goals and learning goals. The AI ​​generates target topics most suitable for the user's situation based on past data and similar cases.

[0093] Step 4: The Simulation Department utilizes generative AI to simulate the post-recovery stance based on the target issues proposed by the Proposal Department. The simulation is based on a simulation model or the data used. The generative AI, based on the user's goals and desires, simulates the post-recovery career and lifestyle, and proposes the optimal choices.

[0094] The specific processing unit 290 sends 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 voice representing the user's input to the result of the specific processing. The control unit 46A sends the voice data representing the user's 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 voice data.

[0095] Data generation model 58 is what is known as generative AI (Artificial Intelligence). An example of data generation model 58 includes ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> Generative AI, such as data generation model 58, is obtained by deep learning through a neural network. The data generation model 58 is input with a prompt containing instructions, and with inference data such as speech data representing speech, text data representing text, and image data representing images (e.g., data of still images or data of moving images). The data generation model 58 infers the inference data based on the instructions shown in the prompt and outputs the inference result in one or more data forms such as speech data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the above-described specific processing while using the data generation model 58. The data generation model 58 can also be a fine-tuned model to output inference results from prompts without instructions; in this case, the data generation model 58 can output inference results from prompts without instructions. The data processing device 12, etc., includes multiple data generation models 58, including AI other than generative AI. AI other than generative AI includes, but is not limited to, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or Naive Bayes. Furthermore, AI can also act as an AI agent. Additionally, when the processing described above is performed by AI, this processing may be partially or entirely performed by AI, but is not limited to this example. Moreover, processing performed by AI, including generative AI, can be replaced by rule-based processing, and vice versa.

[0096] Furthermore, the processing performed by the aforementioned data processing system 10 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart device 14, but it can also be executed jointly by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart device 14. Additionally, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart device 14 or external devices, and the smart device 14 acquires or collects information required for processing from the data processing device 12 or external devices.

[0097] Each of the aforementioned elements, such as the receiving unit, analysis unit, proposal unit, simulation unit, monitoring unit, generation unit, and database unit, can be implemented, for example, in at least one of the smart device 14 and the data processing device 12. For example, the receiving unit is implemented by the control unit 46A of the smart device 14, used to input the user's condition. The analysis unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, using the generating AI to analyze the input information. The proposal unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, proposing target topics based on the analyzed information. The simulation unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, simulating the post-recovery stance based on the proposed target topics. The monitoring unit is implemented, for example, by the control unit 46A of the smart device 14, used to monitor the user's recovery progress. The generation unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, generating appropriate target topics based on past data and similar cases. The database unit is implemented, for example, by the database 24 of the data processing device 12, providing a database to the generating AI when simulating the post-recovery occupation and lifestyle. The correspondence between the various parts and the devices or control units is not limited to the examples above and can be modified in various ways.

[0098] [Second Implementation] Figure 3 An example of the configuration of the data processing system 210 according to the second embodiment is shown.

[0099] like Figure 3 As shown, the data processing system 210 includes a data processing device 12 and smart glasses 214. One example of the data processing device 12 is a server.

[0100] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, RAM 30, and memory 32. The processor 28, RAM 30, and memory 32 are connected to a bus 34. Furthermore, 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. An example of the network 54 includes a WAN and / or LAN, etc.

[0101] 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, RAM 48, and memory 50. The processor 46, RAM 48, and memory 50 are connected to a bus 52. Furthermore, the microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0102] Microphone 238 receives user commands by receiving the user's voice. Microphone 238 captures the user's voice, converts the captured sound into speech data, and outputs it to processor 46. Speaker 240 outputs sound according to commands from processor 46.

[0103] Camera 42 is a small digital camera equipped with an optical system such as 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, used to capture the user's surroundings (e.g., the shooting range defined by an angle of view equivalent to the field of vision of an average healthy person).

[0104] Communication I / F 44 is connected to network 54. Communication I / F 44 and 26 are responsible for the transmission and reception of various information between processor 46 and processor 28 via network 54. The transmission and reception of various information between processor 46 and processor 28 using communication I / F 44 and 26 is performed in a secure state.

[0105] Figure 4 An example of the main functions of the data processing device 12 and the smart glasses 214 is shown. Figure 4 As shown, in the data processing device 12, specific processing is performed by the processor 28. The memory 32 stores a specific processing program 56.

[0106] The processor 28 reads a specific processing program 56 from the memory 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is implemented by the processor 28 operating as a specific processing unit 290 based on the specific processing program 56 executed on the RAM 30.

[0107] The memory 32 stores a data generation model 58 and an emotion-specific model 59. The data generation model 58 and the emotion-specific model 59 are used by the specific processing unit 290. The specific processing unit 290 can use the emotion-specific model 59 to infer the user's emotions and perform specific processing based on the user's emotions. The emotion inference function using the emotion-specific model 59 (emotion-specific function) includes various inferences and predictions related to the user's emotions, but is not limited to this example. Furthermore, emotion inference and prediction also include, for example, emotion analysis (analysis).

[0108] In the smart glasses 214, specific processing is performed by the processor 46. A specific processing program 60 is stored in the memory 50. The processor 46 reads the specific processing program 60 from the memory 50 and executes the read specific processing program 60 on the RAM 48. Specific processing is implemented by the processor 46 operating as a control unit 46A based on the specific processing program 60 executed on the RAM 48. Furthermore, the smart glasses 214 may also have the same data generation model and emotion-specific model as the data generation model 58 and the emotion-specific model 59, and use these models to perform the same processing as the specific processing unit 290.

[0109] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains the processing results (prediction results, etc.) using the data generation model 58 by communicating with the server device that has the data generation model 58. In addition, the data processing device 12 may be a server device or a user-owned terminal device (e.g., a mobile phone, robot, home appliance, etc.).

[0110] The specific processing unit 290 sends 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 voice input representing the user's input to the specific processing result. The control unit 46A sends the voice data representing the user's 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.

[0111] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 includes generative AIs such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with a prompt containing instructions, and also with inference data such as speech data representing speech, text data representing text, and image data representing images (e.g., data of still images or data of moving images). The data generation model 58 infers the inference data based on the instructions shown in the prompt and outputs the inference result in one or more data forms such as speech data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the aforementioned specific processing while using the data generation model 58. The data generation model 58 can also be a fine-tuned model to output inference results from prompts that do not contain instructions; in this case, the data generation model 58 is able to output inference results from prompts that do not contain instructions. The data processing device 12, etc., includes various data generation models 58, which include AI other than generative AI. These AIs include, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or Naive Bayes, and are capable of various processing methods, but are not limited to these examples. Furthermore, AI can also be an AI agent. Additionally, when the processing described above is performed by AI, this processing may be partially or entirely performed by AI, but is not limited to these examples. Moreover, processing performed by AI including generative AI can be replaced by rule-based processing, and vice versa.

[0112] The data processing system 210 of the second embodiment performs the same processing as the data processing system 10 of the first embodiment. The processing performed by the data processing system 210 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the smart glasses 214, but it can also be executed jointly by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the smart glasses 214. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the smart glasses 214 or external devices, and the smart glasses 214 acquires or collects information required for processing from the data processing device 12 or external devices.

[0113] Each of the aforementioned elements, such as the receiving unit, analysis unit, proposal unit, simulation unit, monitoring unit, generation unit, and database unit, can be implemented, for example, in at least one of the smart glasses 214 and the data processing device 12. For example, the receiving unit is implemented by the control unit 46A of the smart glasses 214, used to input the user's condition. The analysis unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, using the generating AI to analyze the input information. The proposal unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, proposing target topics based on the analyzed information. The simulation unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, simulating the post-recovery stance based on the proposed target topics. The monitoring unit is implemented, for example, by the control unit 46A of the smart glasses 214, used to monitor the user's recovery progress. The generation unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, generating appropriate target topics based on past data and similar cases. The database unit is implemented, for example, by the database 24 of the data processing device 12, providing a database to the generating AI when simulating the post-recovery occupation and lifestyle. The correspondence between the various parts and the devices or control units is not limited to the examples above and can be modified in various ways.

[0114] [Third Implementation] Figure 5 An example of the configuration of the data processing system 310 according to the third embodiment is shown.

[0115] like Figure 5 As shown, the data processing system 310 includes a data processing device 12 and a head-mounted terminal 314. An example of the data processing device 12 is a server.

[0116] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, RAM 30, and memory 32. The processor 28, RAM 30, and memory 32 are connected to a bus 34. Furthermore, 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. An example of the network 54 includes a WAN and / or LAN, etc.

[0117] The head-mounted 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, RAM 48, and memory 50. The processor 46, RAM 48, and memory 50 are connected to a bus 52. Furthermore, the microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0118] Microphone 238 receives user commands by receiving the user's voice. Microphone 238 captures the user's voice, converts the captured sound into speech data, and outputs it to processor 46. Speaker 240 outputs sound according to commands from processor 46.

[0119] Camera 42 is a small digital camera equipped with an optical system such as 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, used to capture the user's surroundings (e.g., the shooting range defined by an angle of view equivalent to the field of vision of an average healthy person).

[0120] Communication I / F 44 is connected to network 54. Communication I / F 44 and 26 are responsible for the transmission and reception of various information between processor 46 and processor 28 via network 54. The transmission and reception of various information between processor 46 and processor 28 using communication I / F 44 and 26 is performed in a secure state.

[0121] Figure 6 An example of the main functions of the data processing device 12 and the head-mounted terminal 314 is shown. Figure 6 As shown, in the data processing device 12, specific processing is performed by the processor 28. The memory 32 stores a specific processing program 56.

[0122] The processor 28 reads a specific processing program 56 from the memory 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is implemented by the processor 28 operating as a specific processing unit 290 based on the specific processing program 56 executed on the RAM 30.

[0123] The memory 32 stores a data generation model 58 and an emotion-specific model 59. The data generation model 58 and the emotion-specific model 59 are used by the specific processing unit 290. The specific processing unit 290 can use the emotion-specific model 59 to infer the user's emotions and perform specific processing based on the user's emotions. The emotion inference function using the emotion-specific model 59 (emotion-specific function) includes various inferences and predictions related to the user's emotions, but is not limited to this example. Furthermore, emotion inference and prediction also include, for example, emotion analysis (analysis).

[0124] In the head-mounted terminal 314, specific processing is performed by the processor 46. A specific program 60 is stored in the memory 50. The processor 46 reads the specific program 60 from the memory 50 and executes the read specific program 60 on the RAM 48. Specific processing is implemented by the processor 46 operating as a control unit 46A based on the specific program 60 executed on the RAM 48. Furthermore, the head-mounted terminal 314 may also have the same data generation model and emotion-specific model as the data generation model 58 and the emotion-specific model 59, and use these models to perform the same processing as the specific processing unit 290.

[0125] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains the processing results (prediction results, etc.) using the data generation model 58 by communicating with the server device that has the data generation model 58. In addition, the data processing device 12 may be a server device or a user-owned terminal device (e.g., a mobile phone, robot, home appliance, etc.).

[0126] The specific processing unit 290 sends the result of the specific processing to the head-mounted terminal 314. In the head-mounted terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires voice representing the user's input to the specific processing result. The control unit 46A sends the voice data representing the user's 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.

[0127] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 includes generative AIs such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with a prompt containing instructions, and also with inference data such as speech data representing speech, text data representing text, and image data representing images (e.g., data of still images or data of moving images). The data generation model 58 infers the inference data based on the instructions shown in the prompt and outputs the inference result in one or more data forms such as speech data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the aforementioned specific processing while using the data generation model 58. The data generation model 58 can also be a fine-tuned model to output inference results from prompts that do not contain instructions; in this case, the data generation model 58 is able to output inference results from prompts that do not contain instructions. The data processing device 12, etc., includes various data generation models 58, which include AI other than generative AI. These AIs include, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or Naive Bayes, and are capable of various processing methods, but are not limited to these examples. Furthermore, AI can also be an AI agent. Additionally, when the processing described above is performed by AI, this processing may be partially or entirely performed by AI, but is not limited to these examples. Moreover, processing performed by AI including generative AI can be replaced by rule-based processing, and vice versa.

[0128] The data processing system 310 of the third embodiment performs the same processing as the data processing system 10 of the first embodiment. The processing performed by the data processing system 310 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the head-mounted terminal 314, but it can also be executed jointly by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the head-mounted terminal 314. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the head-mounted terminal 314 or external devices, and the head-mounted terminal 314 acquires or collects information required for processing from the data processing device 12 or external devices.

[0129] Each of the aforementioned elements, such as the receiving unit, analysis unit, proposal unit, simulation unit, monitoring unit, generation unit, and database unit, can be implemented, for example, in at least one of the head-mounted terminal 314 and the data processing device 12. For example, the receiving unit is implemented by the control unit 46A of the head-mounted terminal 314, used to input the user's condition. The analysis unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, using the generating AI to analyze the input information. The proposal unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, proposing target topics based on the analyzed information. The simulation unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, simulating the post-recovery stance based on the proposed target topics. The monitoring unit is implemented, for example, by the control unit 46A of the head-mounted terminal 314, used to monitor the user's recovery progress. The generation unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, generating appropriate target topics based on past data and similar cases. The database unit is implemented, for example, by the database 24 of the data processing device 12, providing a database to the generating AI when simulating the post-recovery occupation and lifestyle. The correspondence between the various parts and the devices or control units is not limited to the examples above and can be modified in various ways.

[0130] [Fourth Implementation] Figure 7 An example of the configuration of the data processing system 410 according to the fourth embodiment is shown.

[0131] like Figure 7 As shown, the 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.

[0132] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 includes a processor 28, RAM 30, and memory 32. The processor 28, RAM 30, and memory 32 are connected to a bus 34. Furthermore, 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. An example of the network 54 includes a WAN and / or LAN, etc.

[0133] Robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control object 443. The computer 36 includes a processor 46, RAM 48, and memory 50. The processor 46, RAM 48, and memory 50 are connected to a bus 52. Furthermore, the microphone 238, speaker 240, camera 42, and control object 443 are also connected to the bus 52.

[0134] Microphone 238 receives user commands by receiving the user's voice. Microphone 238 captures the user's voice, converts the captured sound into speech data, and outputs it to processor 46. Speaker 240 outputs sound according to commands from processor 46.

[0135] The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture and shutter, and an imaging element such as a CMOS image sensor or a CCD image sensor, used to photograph the user's surroundings (e.g., the shooting range defined by an angle of view equivalent to the field of vision of an average healthy person).

[0136] Communication I / F 44 is connected to network 54. Communication I / F 44 and 26 are responsible for the transmission and reception of various information between processor 46 and processor 28 via network 54. The transmission and reception of various information between processor 46 and processor 28 using communication I / F 44 and 26 is performed in a secure state.

[0137] The controlled object 443 includes a display device, LEDs for the eyes, and motors for driving the arms, hands, and feet. The posture and movements of the robot 414 are controlled by controlling the motors for the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. In addition, facial expressions of the robot 414 can also be expressed by controlling the illumination state of the LEDs for the robot 414's eyes.

[0138] Figure 8 An example of the main functions of the data processing device 12 and the robot 414 is shown. For example... Figure 8 As shown, in the data processing device 12, specific processing is performed by the processor 28. The memory 32 stores a specific processing program 56.

[0139] The processor 28 reads a specific processing program 56 from the memory 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is implemented by the processor 28 operating as a specific processing unit 290 based on the specific processing program 56 executed on the RAM 30.

[0140] The memory 32 stores a data generation model 58 and an emotion-specific model 59. The data generation model 58 and the emotion-specific model 59 are used by the specific processing unit 290. The specific processing unit 290 can use the emotion-specific model 59 to infer the user's emotions and perform specific processing based on the user's emotions. The emotion inference function using the emotion-specific model 59 (emotion-specific function) includes various inferences and predictions related to the user's emotions, but is not limited to this example. Furthermore, emotion inference and prediction also include, for example, emotion analysis (analysis).

[0141] In robot 414, specific processing is performed by processor 46. A specific program 60 is stored in memory 50. Processor 46 reads the specific program 60 from memory 50 and executes the read specific program 60 on RAM 48. Specific processing is achieved by processor 46 acting as control unit 46A based on the specific program 60 executed on RAM 48. Furthermore, robot 414 may also have the same data generation model and emotion-specific model as data generation model 58 and emotion-specific model 59, and use these models to perform the same processing as specific processing unit 290.

[0142] Furthermore, other devices besides the data processing device 12 may also have the data generation model 58. For example, a server device may have the data generation model 58. In this case, the data processing device 12 obtains the processing results (prediction results, etc.) using the data generation model 58 by communicating with the server device that has the data generation model 58. In addition, the data processing device 12 may be a server device or a user-owned terminal device (e.g., a mobile phone, robot, home appliance, etc.).

[0143] The specific processing unit 290 sends the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires voice representing the user's input regarding the result of the specific processing. The control unit 46A sends the voice data representing 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.

[0144] The data generation model 58 is a so-called generative AI. An example of the data generation model 58 includes generative AIs such as ChatGPT. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with a prompt containing instructions, and also with inference data such as speech data representing speech, text data representing text, and image data representing images (e.g., data of still images or data of moving images). The data generation model 58 infers the inference data based on the instructions shown in the prompt and outputs the inference result in one or more data forms such as speech data, text data, and image data. The data generation model 58 includes, for example, text generation AI, image generation AI, and multimodal generation AI. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization. The specific processing unit 290 performs the aforementioned specific processing while using the data generation model 58. The data generation model 58 can also be a fine-tuned model to output inference results from prompts that do not contain instructions; in this case, the data generation model 58 is able to output inference results from prompts that do not contain instructions. The data processing device 12, etc., includes various data generation models 58, which include AI other than generative AI. These AIs include, for example, linear regression, logistic regression, decision trees, random forests, support vector machines (SVM), k-means clustering, convolutional neural networks (CNN), recurrent neural networks (RNN), generative adversarial networks (GAN), or Naive Bayes, and are capable of various processing methods, but are not limited to these examples. Furthermore, AI can also be an AI agent. Additionally, when the processing described above is performed by AI, this processing may be partially or entirely performed by AI, but is not limited to these examples. Moreover, processing performed by AI including generative AI can be replaced by rule-based processing, and vice versa.

[0145] The data processing system 410 of the fourth embodiment performs the same processing as the data processing system 10 of the first embodiment. The processing performed by the data processing system 410 is executed by the specific processing unit 290 of the data processing device 12 or the control unit 46A of the robot 414, but it can also be executed jointly by the specific processing unit 290 of the data processing device 12 and the control unit 46A of the robot 414. Furthermore, the specific processing unit 290 of the data processing device 12 acquires or collects information required for processing from the robot 414 or external devices, and the robot 414 acquires or collects information required for processing from the data processing device 12 or external devices.

[0146] Each of the aforementioned elements, such as the receiving unit, analysis unit, proposal unit, simulation unit, monitoring unit, generation unit, and database unit, can be implemented, for example, in at least one of the robot 414 and the data processing device 12. For example, the receiving unit is implemented by the control unit 46A of the robot 414, used to input the user's condition. The analysis unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, using the generating AI to analyze the input information. The proposal unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, proposing target topics based on the analyzed information. The simulation unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, simulating the post-recovery stance based on the proposed target topics. The monitoring unit is implemented, for example, by the control unit 46A of the robot 414, used to monitor the user's recovery progress. The generation unit is implemented, for example, by the specific processing unit 290 of the data processing device 12, generating appropriate target topics based on past data and similar cases. The database unit is implemented, for example, by the database 24 of the data processing device 12, providing a database to the generating AI when simulating the post-recovery occupation and lifestyle. The correspondence between the various parts and the devices or control units is not limited to the examples above and can be modified in various ways.

[0147] Furthermore, the emotion-specific model 59, serving as an emotion engine, can determine the user's emotion based on a specific mapping. Specifically, the emotion-specific model 59 can determine the user's emotion based on an emotion graph that serves as a specific mapping (see...). Figure 9 The system determines the user's emotions. In addition, the emotion-specific model 59 can also determine the robot's emotions in the same way, and the specific processing unit 290 can also perform specific processing using the robot's emotions.

[0148] Figure 9 This is a diagram representing an emotion map 400 that maps various emotions. In the emotion map 400, emotions are arranged radially from the center in concentric circles. The closer to the center of the concentric circles, the more primitive the emotion is. Further out on the concentric circles, emotions are arranged representing states or actions arising from mood. Emotion is a concept that includes both feelings and mental states. To the left of the concentric circles, emotions generated by reactions occurring in the brain are arranged roughly. To the right of the concentric circles, emotions guided by situational judgments are arranged roughly. Above and below the concentric circles, emotions generated by reactions occurring in the brain and guided by situational judgments are arranged roughly. Furthermore, the emotion of "pleasure" is arranged above the concentric circles, and the emotion of "unpleasantness" is arranged below. Thus, in the emotion map 400, various emotions are mapped according to the structure of emotion generation, while easily generated emotions are mapped nearby.

[0149] These emotions are distributed at the 3 o'clock position on the Emotion Chart 400, and usually fluctuate between peace and unease. In the right half of the Emotion Chart 400, because situational awareness is more dominant than internal feelings, it gives a sense of calm.

[0150] The inner side of the emotion diagram 400 represents the mind, and the outer side of the emotion diagram 400 represents actions. Therefore, the further you go to the outer side of the emotion diagram 400, the more the emotion can be seen (manifested in actions).

[0151] Here, human emotions are based on a balance of various factors such as posture and blood sugar levels. When these balances deviate from the ideal, it indicates unhappiness; when they approach the ideal, it indicates pleasure. In robots, cars, and motorcycles, emotions can also be created based on a balance of factors such as posture and remaining battery power. When these balances deviate from the ideal, it indicates unhappiness; when they approach the ideal, it indicates pleasure. Emotion maps can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Speech Emotion Recognition and Brain Physiological Signal Analysis Systems for Emotion, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). In the left half of the emotion map, emotions belonging to the "Reaction" domain, where sensation is dominant, are arranged. Furthermore, in the right half of the emotion map, emotions belonging to the "Situation" domain, where situational cognition is dominant, are arranged.

[0152] The emotion map defines two types of emotions that promote learning. One is a negative emotion located near the middle of "repentance" or "reflection" on the situation side. That is, when the robot experiences negative emotions such as "I never want to feel this way again" or "I never want to be scolded again." The other is a positive emotion located near "desire" on the response side. That is, when the robot experiences positive feelings such as "wanting more" or "wanting to know more."

[0153] The emotion-specific model 59 feeds user input into a pre-learned neural network to obtain emotion values ​​representing each emotion shown in the emotion graph 400, and determines the user's emotion. This neural network is pre-learned based on multiple learning data sets that combine user input with emotion values ​​representing each emotion shown in the emotion graph 400. Furthermore, this neural network is learned to... Figure 10 As shown in sentiment graph 900, sentiment values ​​in nearby configurations are similar to each other. Figure 10 Examples show how emotions such as "peace of mind," "stability," and "reassurance" can result in similar emotional values.

[0154] In the above embodiments, a specific processing is described by a single computer 22, but the technology disclosed herein is not limited to this, and distributed processing by multiple computers, including computer 22, is also possible.

[0155] In the above embodiments, an example of storing a specific processing program 56 in memory 32 is illustrated, but the technology disclosed herein is not limited thereto. For example, the specific processing program 56 may also 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 performs specific processing according to the specific processing program 56.

[0156] Alternatively, the specific processing program 56 can be stored in a storage device such as a server connected to the data processing device 12 via a network 54, and the specific processing program 56 can be downloaded and installed into the computer 22 upon request from the data processing device 12.

[0157] Furthermore, it is not necessary to store the entire specific process 56 in a storage device such as a server connected to the data processing device 12 via the network 54, nor is it necessary to store the entire specific process 56 in the memory 32; a portion of the specific process 56 may also be stored.

[0158] As a hardware resource for performing specific processing, various processors can be used. For example, a CPU is a general-purpose processor that functions as a hardware resource for performing specific processing by executing software, i.e., programs. Additionally, a dedicated circuit can be listed as a processor; it is a processor with a circuit structure specifically designed for performing specific processing, such as a FPGA (Field-Programmable Gate Array), PLD (Programmable Logic Device), or ASIC (Application Specific Integrated Circuit). Every processor has built-in or connected memory, and every processor executes specific processing by using that memory.

[0159] The hardware resources for performing a specific process can consist of one of these various processors, or a combination of two or more processors of the same or different types (e.g., a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Furthermore, the hardware resources for performing a specific process can also be a single processor.

[0160] As an example of a single processor, the first type consists of a combination of one or more CPUs and software, which functions as a hardware resource to perform specific processing. The second type uses a processor, such as a System-on-a-chip (SoC), which implements the entire system functionality, including multiple hardware resources performing specific processing, using a single IC chip. In this case, the specific processing is implemented using one or more of the aforementioned processors that serve as hardware resources.

[0161] Furthermore, as the hardware architecture of these various processors, more specifically, circuits combining semiconductor elements and other circuit components can be used. Moreover, the specific process described above is merely an example. Therefore, it goes without saying that, without departing from the main point, unnecessary steps can be removed, new steps can be added, or the processing order can be changed.

[0162] Furthermore, although the above examples have been described in terms of first to fourth embodiments, some or all of these embodiments can also be combined. Additionally, the smart device 14, smart glasses 214, head-mounted terminal 314, and robot 414 are just examples; they can be combined separately or are other devices.

[0163] The foregoing descriptions and illustrations are detailed explanations of the parts covered by this disclosure and are merely one example of this disclosure. For instance, the descriptions of the above-described structure, function, role, and effect are just one example of the structure, function, role, and effect of the parts covered by this disclosure. Therefore, it goes without saying that, without departing from the spirit of this disclosure, unnecessary parts can be deleted, new elements can be added, or replacements can be made to the foregoing descriptions and illustrations. Furthermore, to avoid confusion and facilitate understanding of the parts covered by this disclosure, explanations of technical common sense that does not require special explanation for implementing this disclosure have been omitted from the foregoing descriptions and illustrations.

[0164] All documents, patent applications and technical standards described in this specification are incorporated herein by reference as if they were specifically and individually described as incorporated by reference.

[0165] (Note 1) A system comprising: The receiving unit is used to input the user's status. The parsing unit is used to parse the information input by the receiving unit; The proposal department proposes target topics based on the information analyzed by the analysis department. The simulation department simulates the restored state based on the target issue proposed by the proposal department.

[0166] (Note 2) The system as described in Appendix 1 is characterized by including a monitoring unit for monitoring the user's recovery progress.

[0167] (Note 3) The system as described in Appendix 1 is characterized by including a generation unit, which generates appropriate target topics based on past data and similar cases through a generation AI.

[0168] (Note 4) The system as described in Appendix 1 is characterized by including a database unit for generating AI in the simulation of restored occupations and lifestyles.

[0169] (Note 5) The system as described in Appendix 1 is characterized in that the proposal department proposes an exercise program corresponding to the progress of rehabilitation.

[0170] (Note 6) The system as described in Appendix 1 is characterized in that the simulation unit simulates the restored occupation and lifestyle according to the user's wishes.

[0171] (Note 7) The system as described in Appendix 1 is characterized in that the receiving unit estimates the user's emotion and adjusts the display mode of the input interface according to the estimated user emotion.

[0172] (Note 8) The system as described in Appendix 1 is characterized in that the receiving unit analyzes the user's past input history and proposes an appropriate input method.

[0173] (Note 9) The system as described in Appendix 1 is characterized in that, when inputting data, the receiving unit customizes the input items according to the user's current physical condition and psychological state.

[0174] (Postscript 10) The system as described in Appendix 1 is characterized in that the receiving unit estimates the user's emotions and determines the input priority order based on the estimated user emotions.

[0175] (Postscript 11) The system as described in Appendix 1 is characterized in that, when inputting data, the receiving unit considers the user's geographical location information and prioritizes displaying highly relevant input items.

[0176] (Postscript 12) The system as described in Appendix 1 is characterized in that, when inputting, the receiving unit analyzes the user's social media activities and proposes relevant input items.

[0177] (Postscript 13) The system as described in Appendix 1 is characterized in that the parsing unit infers the user's emotions and adjusts the parsing algorithm based on the inferred user emotions.

[0178] (Postscript 14) The system as described in Appendix 1 is characterized in that, during the parsing process, the parsing unit refers to the user's past data to improve the accuracy of the parsing.

[0179] (Postscript 15) The system as described in Appendix 1 is characterized in that the parsing unit customizes the parsing method according to the user's current physical condition during parsing.

[0180] (Postscript 16) The system as described in Appendix 1 is characterized in that the parsing unit infers the user's emotions and adjusts the display method of the parsing results according to the inferred user emotions.

[0181] (Postscript 17) The system as described in Appendix 1 is characterized in that the parsing unit considers the user's geographical location information during parsing to improve the accuracy of the parsing.

[0182] (Postscript 18) The system as described in Appendix 1 is characterized in that, during parsing, the parsing unit analyzes the user's social media activities and uses the relevant data for parsing.

[0183] (Postscript 19) The system as described in Appendix 1 is characterized in that the proposal department presupposes the user's emotions and adjusts the expression of the proposal according to the presumed user emotions.

[0184] (Postscript 20) The system as described in Appendix 1 is characterized in that the proposal department, when making a proposal, refers to the user's past data to improve the accuracy of the proposal.

[0185] (Postscript 21) The system as described in Appendix 1 is characterized in that the proposal department customizes the proposal content based on the user's current physical condition when making a proposal.

[0186] (Postscript 22) The system as described in Appendix 1 is characterized in that the proposal department presupposes the user's emotions and determines the priority of proposals based on the presumed user emotions.

[0187] (Postscript 23) The system as described in Appendix 1 is characterized in that, when making a proposal, the proposal department considers the user's geographical location information and prioritizes proposals with high relevance.

[0188] (Postscript 24) The system as described in Appendix 1 is characterized in that, when making a proposal, the proposal department analyzes the user's social media activities and makes relevant proposals.

[0189] (Postscript 25) The system as described in Appendix 1 is characterized in that the simulation unit improves the accuracy of the simulation by referring to the user's past data during the simulation.

[0190] (Postscript 26) The system as described in Appendix 1 is characterized in that the simulation unit customizes the simulation method according to the user's current physical condition during simulation.

[0191] (Postscript 27) The system as described in Appendix 1 is characterized in that the simulation unit estimates the user's emotions and determines the priority of the simulation based on the estimated user emotions.

[0192] (Postscript 28) The system as described in Appendix 1 is characterized in that, during simulation, the simulation unit considers the user's geographical location information and prioritizes simulations with high relevance.

[0193] (Postscript 29) The system as described in Appendix 1 is characterized in that, during simulation, the simulation unit analyzes the user's social media activities and performs related simulations.

[0194] (Note 30) The system as described in Appendix 2 is characterized in that the monitoring unit estimates the user's emotions and adjusts the monitoring frequency according to the estimated user emotions.

[0195] (Postscript 31) The system as described in Appendix 2 is characterized in that the monitoring unit improves the accuracy of monitoring by referring to the user's past data during monitoring.

[0196] (Note 32) The system as described in Appendix 2 is characterized in that the monitoring unit presupposes the user's emotions and determines the monitoring priority based on the presumed user emotions.

[0197] (Postscript 33) The system as described in Appendix 2 is characterized in that, when monitoring, the monitoring unit considers the user's geographical location information and prioritizes monitoring that is highly relevant.

[0198] (Postscript 34) The system as described in Appendix 3 is characterized in that the generation unit estimates the user's emotions and adjusts the generation algorithm based on the estimated user emotions.

[0199] (Postscript 35) The system as described in Appendix 3 is characterized in that the generation unit improves the accuracy of generation by referring to the user's past data during the generation process.

[0200] (Postscript 36) The system as described in Appendix 3 is characterized in that the generation unit presupposes the user's emotions and determines the generation priority order based on the presupposed user emotions.

[0201] (Postscript 37) The system as described in Appendix 3 is characterized in that, during generation, the generation unit considers the user's geographical location information and prioritizes generating data with high relevance.

[0202] (Postscript 38) The system as described in Appendix 4 is characterized in that the database department presupposes the user's sentiment and adjusts the database retrieval results based on the presumed user sentiment.

[0203] (Postscript 39) The system described in Appendix 4 is characterized in that, when updating the database, the database unit refers to the user's past data to improve the accuracy of the update.

[0204] (Postscript 40) The system as described in Appendix 4 is characterized in that the database department presupposes the user's sentiment and determines the database retrieval priority based on the presumed user sentiment.

[0205] (Postscript 41) The system described in Appendix 4 is characterized in that, when updating the database, the database department considers the user's geographical location information and prioritizes updating data with high relevance.

Claims

1. A system, characterized in that, include: The receiving unit is used to input the user's status. The parsing unit is used to parse the information input by the receiving unit; The proposal department proposes target topics based on the information analyzed by the analysis department. The simulation department simulates the restored state based on the target issue proposed by the proposal department.

2. The system as described in claim 1, characterized in that, This includes a monitoring department used to monitor the progress of users' recovery.

3. The system as described in claim 1, characterized in that, It includes a generation unit that generates appropriate target topics based on past data and similar cases using AI.

4. The system as described in claim 1, characterized in that, It includes a database section, which is used to generate AI for simulated restoration of occupations and lifestyles.

5. The system as described in claim 1, characterized in that, The proposal department suggested exercise programs corresponding to the progress of rehabilitation.

6. The system as described in claim 1, characterized in that, The simulation department simulates the restored profession and lifestyle according to the user's wishes.

7. The system as described in claim 1, characterized in that, The receiving unit estimates the user's emotions and adjusts the display mode of the input interface according to the estimated user emotions.

8. The system as described in claim 1, characterized in that, The receiving unit analyzes the user's past input history and proposes an appropriate input method.

9. The system as described in claim 1, characterized in that, When inputting data, the receiving unit customizes the input items based on the user's current physical and mental state.

Images