System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting data of a professional player; means for training a generative artificial intelligence model based on the collected data of the professional player; means for optimizing the generative artificial intelligence model; means for providing the optimized generative artificial intelligence model to a user terminal; and means for receiving, analyzing, and providing feedback on a result of a match between a user and the generative artificial intelligence model.SELECTED DRAWING: Figure 1

Description

[Technical Field]

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

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

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

[0004] In board games such as Go, Shogi, Chess, and card games, playing against skilled opponents is essential to improving one's skills. However, opportunities to play directly against professional players are extremely limited, hindering the growth of average players. Furthermore, it is difficult to receive direct feedback from professional players. For this reason, there is a need to easily provide an environment where players can play against professionals and learn and improve their skills. [Means for solving the problem]

[0005] The present invention provides a means for collecting data from professional players and training a generative AI model based on that data. It also provides a system that includes a means for optimizing the generative AI model based on the collected data and providing the optimized model to a user terminal. Specifically, the system collects game data from professional players, trains a generative AI model using deep learning technology, and provides the model to the user terminal via a cloud server. It also includes a means for receiving and analyzing the results of users' matches against the generative AI model and providing optimal feedback. In this way, a system is provided that allows users to efficiently improve their own skills while learning from professional techniques.

[0006] A "professional player" is a specialized player with advanced skills and knowledge in a particular board game.

[0007] "Data" refers to information such as the game history of professional players, records of each move made during the game, and strategic patterns.

[0008] A "generative artificial intelligence model" is an AI model that uses machine learning and deep learning technologies to generate data to reproduce the playing style of a professional.

[0009] "Optimization" is the process of evaluating the performance of a generative artificial intelligence model and making adjustments to make the model perform most effectively.

[0010] "User terminal" is a general term for electronic devices used by users, such as smartphones, tablets, and personal computers.

[0011] A "cloud server" is a remote server that provides data and services over the Internet and is accessible from user terminals.

[0012] "Feedback" refers to information that includes an evaluation of the results of a user's match against an AI model and guidance for improvement.

[0013] "Match result" is data showing the overall progress and outcome of a match played by a user against a generated AI model. [Brief explanation of the drawings]

[0014] [Figure 1] 1 is a conceptual diagram showing an example of the configuration of a data processing system according to a first embodiment. [Figure 2] 1 is a conceptual diagram showing an example of main functions of a data processing device and a smart device according to a first embodiment. [Figure 3] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a second embodiment. [Figure 4] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and smart glasses according to a second embodiment. [Figure 5] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a third embodiment. [Figure 6] FIG. 11 is a conceptual diagram showing an example of main functions of a data processing device and a headset-type terminal according to a third embodiment. [Figure 7] FIG. 10 is a conceptual diagram showing an example of the configuration of a data processing system according to a fourth embodiment. [Figure 8] FIG. 10 is a conceptual diagram showing an example of main functions of a data processing device and a robot according to a fourth embodiment. [Figure 9] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 10] 1 shows an emotion map onto which multiple emotions are mapped. [Figure 11] FIG. 3 is a sequence diagram showing a processing flow of the data processing system according to the first embodiment. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1. [Figure 13] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system according to the second embodiment when an emotion engine is combined. [Figure 14]FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0015] An example of an embodiment of a system according to the technology of the present disclosure will be described below with reference to the accompanying drawings.

[0016] First, the terms used in the following description will be explained.

[0017] In the following embodiments, a coded processor (hereinafter simply referred to as a "processor") may be a single arithmetic device or a combination of multiple arithmetic devices. Furthermore, a processor may be a single type of arithmetic device or a combination of multiple types of arithmetic devices. Examples of arithmetic devices include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), and an APU (Accelerated Processing Unit).

[0018] In the following embodiments, a coded RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a working memory by a processor.

[0019] In the following embodiments, the coded storage is one or more non-volatile storage devices that store various programs, various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), and magnetic tapes.

[0020] In the following embodiments, a communication I / F (Interface) with a symbol is an interface including a communication processor, an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), Bluetooth (registered trademark), etc.

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." In other words, "A and / or B" means that it may be only A, only B, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" is also applied when three or more things are expressed connected by "and / or."

[0022] [First embodiment]

[0023] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0024] 1, a data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0025] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0026] The smart device 14 includes a computer 36, a reception device 38, an output device 40, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The reception device 38, the output device 40, and the camera 42 are also connected to the bus 52.

[0027] The reception device 38 includes a touch panel 38A, a microphone 38B, and the like, and receives user input. The touch panel 38A detects contact with an indicator (for example, a pen or a finger) to receive user input by the touch of the indicator. The microphone 38B detects the user's voice to receive user input by voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form of expression that the user 20 can perceive (for example, audio and / or text). The display 40A displays visible information such as text and images in accordance with instructions from the processor 46. The speaker 40B outputs audio in accordance with instructions from the processor 46. The camera 42 is a compact digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 control the exchange of various information between the processor 46 and the processor 28 via the network 54.

[0030] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0031] 2, in the data processing device 12, a specific process is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific process is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0032] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0033] In the smart device 14, the processor 46 performs the reception output process. The storage 50 stores a reception output program 60. The reception output program 60 is used in conjunction with the specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0034] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0035] The present invention is a system that allows users to use an artificial intelligence model created based on collected data from professional players. With this system, users can experience playing against professional players and receive accurate feedback after the match.

[0036] Specific processing of the program

[0037] Professional data collection

[0038] The server collects data on professional players. For example, it retrieves the game history of a professional Go player from an online database or an official website. The data is recorded in coordinate format and used for subsequent processing.

[0039] Data Preprocessing

[0040] The server converts the collected data into a format that can be applied to machine learning. This process extracts features that represent the moves and strategic patterns at each stage. For example, Go game records can be converted into coordinate data, and the overall game progress and the strength of the moves can be quantified.

[0041] AI model generation and training

[0042] The server uses the formatted data to train a generative AI model. Deep learning techniques are used to build a model that reproduces the playing style of a professional. For example, a neural network is built using Keras or TensorFlow, and a model is trained to predict the next move from professional game records.

[0043] Model optimization

[0044] The server uses cross-validation and hyperparameter tuning to optimize the performance of the generated AI model, thereby avoiding overfitting and providing a model that can maintain high accuracy in actual games.

[0045] Deploying and serving the model

[0046] The server uploads the optimized AI model to the cloud server and provides it to the user's device via API. The app installed on the user's device downloads the AI ​​model via this API and stores it locally, allowing users to play against the AI ​​anytime, anywhere.

[0047] Match and Feedback

[0048] The user uses the app to play against the AI ​​model. The device predicts the AI's next move and displays its response. After the match ends, the user sends the match data to the server. The server analyzes the match data, generates feedback on the user's weaknesses and areas for improvement during play, and displays the feedback on the user's device. This feedback helps the user improve their skills.

[0049] Specific examples

[0050] For example, in a Go match, a user opens the smartphone app and taps the start button. The device uses an AI model downloaded from the cloud to suggest the next move. The user continues to make moves in response to the move, progressing the match. After the match ends, the user sends the match log to the server and receives analysis results and feedback in real time. This feedback includes where the wrong move was made, the reasons for the move, and alternatives. By providing an experience that mimics playing against a professional player and specific advice for improving their skills, users can hone their skills efficiently.

[0051] The processing flow will be explained below.

[0052] Step 1:

[0053] The server collects data on professional players, specifically, obtaining their game history and strategy data from online databases and official websites, and storing the data in a database.

[0054] Step 2:

[0055] The server converts the collected data into a format that can be applied to machine learning. For example, Go game record data is converted into coordinate format and the features of each position are extracted. This process maintains the consistency and quality of the data.

[0056] Step 3:

[0057] The server uses the preprocessed data to build a deep learning model, specifically, designing a neural network using frameworks such as TensorFlow and Keras, and training it to reproduce the playing style of a professional.

[0058] Step 4:

[0059] The server evaluates and optimizes the performance of the generated model. It performs cross-validation and hyperparameter tuning to prevent overfitting and improve versatility. The model is then modified based on the evaluation results.

[0060] Step 5:

[0061] The server uploads the optimized model to the cloud server, and an API is built to make it easy for users to access, and the model is provided through this API.

[0062] Step 6:

[0063] The user's device downloads the latest AI model from the cloud server and stores it locally, allowing the user to play against the AI ​​model even offline.

[0064] Step 7:

[0065] The user opens the app and begins playing against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The user then inputs their own next move and continues the match.

[0066] Step 8:

[0067] When a user finishes a match, the results are sent to the server, which analyzes the match data and identifies weaknesses and areas for improvement.

[0068] Step 9:

[0069] The server sends the analysis results and feedback to the user's device. The user's device displays the feedback, helping the user to easily review their game after it has finished. The feedback includes suggestions for mistakes and alternative methods, helping the user improve their skills.

[0070] Example 1

[0071] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0072] Conventional AI-based game systems have struggled to provide users with the same high-level experience as if they were playing against a professional player. Furthermore, they lacked mechanisms for providing detailed feedback to help users improve their skills. Furthermore, AI model optimization and model provisioning were often inefficient.

[0073] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0074] In this invention, the server includes means for collecting data of professional players, means for preprocessing data based on the collected data of professional players, means for training a generative AI model using the preprocessed data, means for optimizing the generated AI model using cross-validation and hyperparameter tuning, means for deploying the optimized generative AI model to a cloud environment and providing it to a user terminal, and means for a user to play against the generative AI model, receive and analyze the results of the game, and provide feedback. This allows users to enjoy an advanced experience similar to playing against professional players, and enables them to improve their skills through detailed feedback.

[0075] "Professional Player Data" means information about an expert or skilled player, including the results of previous matches or games.

[0076] "Data preprocessing" is the process of converting raw data into a form that can be applied to machine learning, including shaping the data, filling in missing values, and treating outliers.

[0077] A "generative artificial intelligence model" is an algorithm or model built using machine learning or deep learning techniques and trained to perform a specific task or prediction.

[0078] "Cross-validation" is a method of evaluating the performance of a machine learning model by dividing the data into multiple parts and checking the model's generalization ability.

[0079] "Hyperparameter tuning" is the process of adjusting parameters related to the model's settings and structure in order to optimize the performance of a machine learning model.

[0080] "Deploying to a cloud environment" means placing optimized models and applications on remote servers available via the Internet.

[0081] "User terminal" refers to a device that is directly operated by a user, and includes smartphones, PCs, tablets, etc.

[0082] "Feedback" refers to advice and comments provided to improve a user's skills based on an analysis of the user's match results.

[0083] The present invention is a system that allows users to use artificial intelligence (AI) models generated based on collected data from professional players. With this system, users can experience playing against professional players and receive accurate feedback after the match.

[0084] System Configuration

[0085] Data collection

[0086] The server collects data on professional players. Specifically, it retrieves the game history of professional Go players from online databases and official websites. This data is recorded in coordinate format and includes the coordinate information of each move, the time of each move, and the result of the game.

[0087] Data Preprocessing

[0088] The server preprocesses the collected data. This process converts raw data into a format that can be applied to machine learning. For example, Go game data is converted into coordinate data, and the overall game progress and the efficiency and impact of each move are quantified.

[0089] AI model generation and learning

[0090] The server uses the preprocessed data to generate and train an AI model. Deep learning techniques are used to build a neural network and create a model that reproduces the playing style of a professional. During this process, libraries such as Keras and TensorFlow are used to define the model, and the model's performance is evaluated using training and validation data.

[0091] Model Optimization

[0092] The server optimizes the model through cross-validation and hyperparameter tuning. This prevents overfitting and provides a model that can maintain high accuracy in actual games. Specifically, it uses methods such as GridSearchCV and RandomSearchCV to find the optimal hyperparameters.

[0093] Deploying and serving the model

[0094] The server deploys the optimized AI model to a cloud environment and provides it to the user's device via an API. Applications installed on the user's device download the model via the API and store it locally. This process allows users to use the latest AI model.

[0095] Match and Feedback

[0096] The user uses the app to play against an AI model. During the match, the device calculates and displays the AI's next move. After the match ends, the user sends the match log to the server. The server analyzes the match data and generates feedback on the user's weaknesses and areas for improvement during play, which is displayed on the user's device. This feedback is provided in detail to help the user improve their skills.

[0097] Specific examples

[0098] For example, in a Go match, the user opens the smartphone app and taps the start button. The device uses an AI model downloaded from the cloud to display the next move. The user plays moves in response to the move, progressing the match. After the match ends, the user sends the match log to the server and receives analysis results and feedback in real time. This feedback includes where the wrong move was made, the reasons for the move, and alternatives. By providing an experience that feels like playing against a professional player and specific advice for improving their skills, users can efficiently hone their skills.

[0099] Prompt Sentence Examples

[0100] "Build an AI model that predicts the next move from Go game data. The data used is the game history of professional Go players, and create a model using deep learning. Use Keras or TensorFlow to build a neural network that replicates the playing style of a professional."

[0101] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0102] Step 1: Data collection

[0103] The server collects data from professional players. The input data is collected from online databases and official website APIs. Specifically, the server calls the API to retrieve the game history of the professional players. This data includes the coordinate information of each move, the time of each move, the result of the game, etc. The output data is stored in a local database.

[0104] Step 2: Data Preprocessing

[0105] The data collected by the server is preprocessed. The input is the raw data collected in step 1. Specific operations include shaping the data, filling in missing values, and filtering outliers. For example, Go game record data is converted into coordinate format, and the overall game progress and the efficiency and impact of each move are quantified. The output is preprocessed data that can be applied to machine learning.

[0106] Step 3: AI model generation and training

[0107] The server uses the preprocessed data to generate and train an AI model. The input is the preprocessed data obtained in step 2. Specifically, it uses deep learning techniques to build a neural network and defines the model using Keras or TensorFlow. It evaluates the model's performance using training and validation data and learns optimal weights. The output is an initial AI model.

[0108] Step 4: Model optimization

[0109] The server optimizes the generated AI model. The input is the initial AI model generated in step 3. Specifically, it performs cross-validation and hyperparameter tuning to find optimal hyperparameters using techniques such as GridSearchCV and RandomSearchCV. The output is an optimized AI model.

[0110] Step 5: Deploy and serve the model

[0111] The server deploys the optimized AI model to the cloud environment. The input is the AI ​​model optimized in step 4. Specifically, the model is uploaded to the cloud server and published as an API endpoint. The user device then downloads the AI ​​model via this API and saves it locally. The output is the AI ​​model on the user device.

[0112] Step 6: Match and feedback

[0113] The user uses the app to play against an AI model. The input is the downloaded AI model. Specifically, the user taps the start button on the app, and the device calculates and displays the AI's next move. After the match ends, the user sends the match log to the server. The server analyzes the match data and generates feedback on the user's weaknesses and areas for improvement during play, which is displayed on the user's device. The output is detailed feedback and advice on how to improve the user's skills.

[0114] (Application example 1)

[0115] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0116] In traditional manufacturing, it has been difficult for new employees and employees in training to efficiently learn the skills of professional workers. In particular, there is a need to pass on the skilled techniques and know-how of professional workers and improve production efficiency while maintaining consistent quality. However, current training methods require time and effort and pose a risk of affecting the operating rate of the production line. Therefore, a new system is needed to efficiently and reliably transfer the skills and know-how of professional workers to new employees.

[0117] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0118] In this invention, the server includes means for collecting data of professional workers, means for training a generative AI model based on the collected data of professional workers, means for optimizing the generative AI model, means for providing the optimized generative AI model to a user terminal, and means for receiving and analyzing the results of a user's competition or collaboration with the generative AI model and providing feedback. This enables new employees and employees in training to learn skills effectively and efficiently by competing or collaborating with professional workers and receiving feedback based on real-life experience.

[0119] A "professional worker" refers to an individual who is engaged in manufacturing work with advanced specialized knowledge and skills and who has a certain level of proficiency.

[0120] "Data" refers to information such as operation records, measurement values, and operation logs generated when professional workers perform their work.

[0121] A "generative artificial intelligence model" refers to an artificial intelligence system that is generated by applying machine learning algorithms based on collected data.

[0122] "Learning" refers to the process by which an AI model uses collected data to understand behaviors and patterns and improve its capabilities.

[0123] "Optimization" refers to the process of making adjustments to an artificial intelligence model to maximize its performance and minimize error.

[0124] "User terminal" refers to a device for using an artificial intelligence model, and specifically includes hardware such as smartphones, tablets, PCs, and robots.

[0125] "Competition or collaboration" refers to the process in which a user actually works or collaborates with an artificial intelligence model.

[0126] "Receiving" refers to a user obtaining data generated as a result of a match or collaboration.

[0127] "Analysis" refers to the detailed examination of the results of a competition or collaboration and the processing of data to understand its significance and patterns.

[0128] "Feedback" refers to information that indicates evaluations and areas for improvement based on a user's actions and results.

[0129] The system of the present invention consists of the following main steps: First, collect data from professional workers, generate and optimize a generative AI model based on the data, and then provide the model to the user terminal, analyze the results of the competitive or collaborative play, and provide feedback.

[0130] 1. Professional data collection

[0131] The server automatically collects various data, such as motion records, measurements, and operation logs, from professional workers as they perform their tasks. This data is collected using high-performance sensors, cameras, and other measuring devices. This data is then aggregated in one place and used as a basic dataset for machine learning.

[0132] 2. Data preprocessing and feature extraction

[0133] The server preprocesses the collected data and formats it into a format suitable for machine learning. Specifically, it normalizes the data, fills in missing values, extracts features, etc. This process prepares the data in a way that maximizes the performance of the AI ​​model.

[0134] 3. Generating and training AI models

[0135] The server uses the preprocessed data to train a generative artificial intelligence model using deep learning techniques (specifically TensorFlow and Keras), which imitates the actions and patterns of professional workers and develops highly accurate prediction capabilities.

[0136] 4. Optimize and deploy the model

[0137] The server optimizes the generated AI model using cross-validation and hyperparameter tuning. The optimized model is then uploaded to a cloud server and provided to user devices via API. User devices can be a variety of devices, including smartphones, tablets, PCs, and robots.

[0138] 5. Competition and Feedback

[0139] Users compete or collaborate with the AI ​​model through the application. The results of the competition or collaboration are sent to the server in real time. The server analyzes the results and generates detailed feedback based on the user's actions and performance. This feedback is displayed on the user's device, allowing the user to effectively improve their skills and work efficiency.

[0140] Examples:

[0141] For example, when new employees undergo training to operate a robot arm on a manufacturing site, the system uses an AI model generated by a server. This system allows the AI ​​model, which has the same skills as a professional worker, to support the new employee's training. After the training session is over, the system provides analysis results and feedback, specifically indicating which movements the new employee needs to improve. This allows employees to smoothly improve their skills.

[0142] Example prompt sentence:

[0143] "I want to design a training system for new employees. How can I build a system that collects data from professional workers, generates an AI model based on that data, and provides real-time feedback to employees?"

[0144] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0145] Step 1:

[0146] The server collects data from professional workers. Specifically, it automatically acquires data such as operation records, measurements, and operation logs from sensors, cameras, and other measuring devices used by professional workers in the manufacturing process and stores the data in a database.

[0147] Input: Real-time data from sensors and cameras.

[0148] Output: Data of professional workers accumulated in a database.

[0149] Specific operation: Collecting data from various sensors and storing it in a database in real time.

[0150] Step 2:

[0151] The server preprocesses the collected data and formats it into a format suitable for machine learning. Specifically, it normalizes the data, fills in missing values, and extracts necessary features.

[0152] Input: Raw data in the database.

[0153] Output: A dataset formatted for machine learning.

[0154] Specific operations: Normalize the data, impute missing values, and select and extract the features to use.

[0155] Step 3:

[0156] The server uses the preprocessed data to train a generative artificial intelligence model. Specifically, it builds a neural network model using a deep learning framework (e.g., TensorFlow, Keras) and trains it using the collected data.

[0157] Input: The preprocessed dataset.

[0158] Output: A trained artificial intelligence model.

[0159] Specific operations: Set up the neural network architecture and train the model while optimizing the learning parameters.

[0160] Step 4:

[0161] The server optimizes the generated artificial intelligence model, specifically by performing cross-validation and hyperparameter tuning to improve the accuracy of the model.

[0162] Input: A trained artificial intelligence model.

[0163] Output: An optimized artificial intelligence model.

[0164] What it does: It performs cross-validation to find the optimal set of hyperparameters and tune them to maximize the model's accuracy.

[0165] Step 5:

[0166] The server uploads the optimized artificial intelligence model to a cloud server and provides it to the user's device via an API.

[0167] Input: An optimized artificial intelligence model.

[0168] Output: An artificial intelligence model that is downloaded to the user's device.

[0169] Specific operation: Deploying a model to a cloud server and downloading the model via an API call from the user's device.

[0170] Step 6:

[0171] Users use their devices to compete or collaborate with artificial intelligence models. Specifically, users operate the application and work in conjunction with the AI ​​model to carry out manufacturing tasks.

[0172] Input: User operation instructions.

[0173] Output: Work assistance data generated by the AI ​​model.

[0174] Specific behavior: The user performs an operation through the application, and the AI ​​model responds with assistance and instructions.

[0175] Step 7:

[0176] The server analyzes the results of the match or collaboration and provides feedback. Specifically, it analyzes the user's operation log in detail, generates advice on areas for improvement and technical enhancement, and provides it to the user's device.

[0177] Input: User operation log.

[0178] Output: Feedback content.

[0179] Specific operation: Analyzes operation logs using analytical algorithms and generates and returns feedback based on the results.

[0180] Furthermore, an emotion engine that estimates the user's emotion may be combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59 and perform identification processing using the user's emotion.

[0181] This invention is a system that collects data from professional players, generates an artificial intelligence model based on that data, and allows users to use it. It also recognizes the user's emotions in real time and adjusts the content of the match and feedback. Users can experience playing against professional players, receive accurate feedback after the match, and receive instruction that takes their own emotions into consideration.

[0182] Specific processing of the program

[0183] Professional data collection

[0184] The server collects data on professional players. For example, it retrieves the game history of a professional Go player from an online database or an official website and stores the data in a database.

[0185] Data Preprocessing

[0186] The server formats the collected data into a format that can be applied to machine learning. It converts the Go game data into coordinate format and extracts the features of each position. This ensures the consistency and quality of the data.

[0187] AI model generation and training

[0188] The server uses the preprocessed data to build a deep learning model. It designs a neural network using frameworks such as TensorFlow and Keras, and trains it to reproduce the playing style of a professional. The AI ​​model has the ability to predict the next move from the professional game record data.

[0189] Model optimization

[0190] The server evaluates and optimizes the performance of the generated model. It performs cross-validation and hyperparameter tuning to prevent overfitting and improve versatility. The model is then adjusted based on the evaluation results.

[0191] Deploying and serving the model

[0192] The server uploads the optimized model to a cloud server and provides it to the user's device via an API. The app installed on the user's device downloads the AI ​​model via this API and stores it locally, allowing users to play against the AI ​​model even offline.

[0193] Introducing the Emotion Engine

[0194] The user's device uses a camera and microphone to collect the user's facial expressions and voice in real time. The device processes the user's emotional data and uses a specific algorithm to recognize the user's emotional state. This emotional data is reflected in the AI's responses and feedback during the match.

[0195] Battles and Emotion Recognition

[0196] A user opens the app and begins a match against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. For example, if the user is feeling impatient or irritated, the AI ​​can display a message encouraging the user to pause.

[0197] Sending and analyzing match results

[0198] Once a user has finished a match, the results are sent to the server. The server analyzes the match data and identifies weaknesses and areas for improvement in the user's play. At the same time, feedback based on emotional data is also generated. This allows the user to receive guidance not only on technical but also psychological aspects.

[0199] Providing Feedback

[0200] The server sends the analysis results and feedback to the user's device, which displays the feedback to help the user review the game afterward. The feedback includes advice on mistakes, alternatives, and even advice based on the user's emotional state.

[0201] Specific examples

[0202] For example, in a Go match, a user opens a smartphone app and taps the start button. The device uses an AI model downloaded from the cloud to suggest the next move. During the match, the user's face is captured on camera, and their facial expression data is analyzed by the emotion engine. The engine determines in real time whether the user is concentrating on the match or feeling anxious or irritated, and suggests an appropriate response. After the match ends, the user sends the match log and emotional data to the server, and receives the analysis results along with technical and psychological advice in real time. In this way, users can learn professional techniques while also mastering how to control their own mental state.

[0203] The processing flow will be explained below.

[0204] Step 1:

[0205] The server collects data on professional players, specifically, obtaining their game history and strategy data from online databases and official websites, and storing the data in a database.

[0206] Step 2:

[0207] The server converts the collected data into a format that can be applied to machine learning. For example, it converts Go game record data into coordinate format and extracts the features of each position. This ensures the consistency and quality of the data.

[0208] Step 3:

[0209] The server uses the preprocessed data to build a deep learning model. It designs a neural network using frameworks such as TensorFlow and Keras, and trains it to reproduce the playing style of a professional. The AI ​​model has the ability to predict the next move from the professional game record data.

[0210] Step 4:

[0211] The server evaluates and optimizes the performance of the generated model. It performs cross-validation and hyperparameter tuning to prevent overfitting and improve versatility. The model is then adjusted based on the evaluation results.

[0212] Step 5:

[0213] The server uploads the optimized model to the cloud server, and an API is built to make it easy for users to access, and the model is provided through this API.

[0214] Step 6:

[0215] The user's device downloads the latest AI model from the cloud server and stores it locally, allowing the user to play against the AI ​​model even offline.

[0216] Step 7:

[0217] The user device uses a camera and microphone to collect the user's facial expressions and voice in real time, and the collected data is sent to an emotion engine to analyze the user's emotional state.

[0218] Step 8:

[0219] The user opens the app and begins a match against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. For example, if the user becomes impatient, the AI ​​displays a message suggesting a pause.

[0220] Step 9:

[0221] The user's device records the user's emotional data and battle data during the match, allowing the user to track emotional changes along with the match history.

[0222] Step 10:

[0223] After a user finishes a match, the match results and emotional data are sent to the server, which analyzes the match data and identifies weaknesses and areas for improvement.

[0224] Step 11:

[0225] The server sends the analysis results and feedback to the user's device. The feedback includes error detection, alternative solutions, and advice based on the user's emotional state. This provides not only technical guidance but also emotional and mental support.

[0226] In this way, users can learn professional techniques while also mastering how to control their own mental state, thereby efficiently improving their skills.

[0227] Example 2

[0228] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0229] Conventional technology not only provides users with the experience of playing against professional players, but also fails to provide feedback that reflects their emotional state during the match in real time. As a result, while users can learn technical skills, it is difficult for them to improve their mental control or emotional state. To solve this problem, a system is needed that collects users' emotional data and adjusts the content of the match and feedback based on that data.

[0230] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for collecting data of professional players, a means for training an AI model to be generated based on the collected data of professional players, and a means for optimizing the generated AI model. This makes it possible to faithfully reproduce the playing style of a professional and provide it to the user. Furthermore, the user terminal includes a means for collecting and analyzing user emotional data using a camera and microphone, a means for adjusting the content of the match and feedback based on the user emotional data in real time, and a means for receiving, analyzing, and providing feedback on the results of the user's match against the generated AI model. This allows the user to receive guidance on both technical and mental aspects.

[0231] "Professional Player Data" means data including the match history, playing style, and related statistical information of players who are recognized as professionals in the game.

[0232] An "artificial intelligence model" is a mathematical structure trained using machine learning techniques, and refers to a system that makes predictions and judgments based on input data.

[0233] A "user terminal" is a device, such as a personal computer, smartphone, or tablet, that is directly operated by a user to send and receive information.

[0234] "Emotion data" is data that indicates the psychological state of the user, obtained from facial expressions, voice, etc., and is quantified or classified using an emotion analysis engine.

[0235] "Feedback" refers to guidance and advice provided based on the results of a match or the user's performance, and includes advice on technical improvements and mental aspects.

[0236] "Collection means" refers to software and hardware configurations for obtaining data via the Internet.

[0237] A "training tool" is a process and tool for analyzing data and training a model using machine learning algorithms.

[0238] "Optimization means" refers to a method of adjusting and evaluating the generated artificial intelligence model to improve its performance, including hyperparameter adjustment and cross-validation.

[0239] "Means of providing" refers to a deployment method for making the optimized artificial intelligence model available to user devices, including cloud APIs and download links.

[0240] "Means of analysis" refers to the processes and tools used to analyze collected data statistically or through machine learning to extract meaningful information.

[0241] The present invention is a system that collects data from professional players and allows users to use an artificial intelligence model generated based on this data. This system also recognizes the user's emotions in real time and adjusts the match content and feedback, allowing users to experience playing against professional players. In addition, users receive accurate feedback after the match and can receive instruction that takes their own emotions into consideration.

[0242] First, the server collects data on professional players. For example, it retrieves the game history of a professional Go player from an online database or official website and stores that data in a database. Specific technologies used include web scraping using Python's BeautifulSoup and Scrapy, and storing the retrieved data in AWS RDS.

[0243] Next, the server converts the collected data into a format that can be applied to machine learning. For example, Go game data is converted into coordinate format, and the features of each position are extracted. At this stage, data processing libraries such as pandas and NumPy are used.

[0244] The server then uses the preprocessed data to build a deep learning model. Using deep learning frameworks such as TensorFlow and Keras, it designs a neural network and trains it to replicate the playing style of a professional. This gives the AI ​​model the ability to predict the next move from professional match data.

[0245] Cross-validation and hyperparameter tuning are required to evaluate and optimize the performance of the constructed model. The server performs cross-validation using scikit-learn and finds optimal hyperparameters using methods such as grid search and random search.

[0246] The optimized model is uploaded to a cloud server by the server and provided to user devices via an API. The server is built using Heroku or AWS Lambda, and a RESTful API is configured. User devices can download the AI ​​model via this API and save it locally. This allows users to play against the AI ​​model even offline.

[0247] The user's device uses a camera and microphone to collect the user's facial expressions and voice in real time. The facial expression data is analyzed using OpenCV or Microsoft Azure's Face API, and a specific algorithm is used to recognize the user's emotional state. This emotional data is reflected in the AI's responses and feedback during the match.

[0248] The user opens the app and begins a match against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. If the user becomes impatient or irritated, the device can display a message encouraging the user to pause.

[0249] After a match, the user sends the match results to the server. The server analyzes the match data and identifies the user's weaknesses and areas for improvement. At the same time, feedback based on emotional data is generated, allowing the user to receive guidance not only on technical but also psychological aspects.

[0250] This system not only allows users to experience playing against professional players, but also provides instruction on both technical and mental aspects, allowing users to learn professional techniques while also mastering their own mental control.

[0251] Prompt Sentence Examples

[0252] For example, "Please tell me the specific steps to develop a Go game application with emotion recognition capabilities using Python. Please explain how to use Azure's Face API to recognize the user's emotions and adjust the AI's response based on that data."

[0253] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0254] Step 1: Professional data collection

[0255] The server collects data on professional players. Specifically, it uses Python's BeautifulSoup and Scrapy to scrape the game history of professional players from online databases and official websites. Website URLs are used as input, and game data in JSON format is output. This data is then uploaded to AWS's RDS for efficient management and storage.

[0256] Step 2: Preprocessing the data

[0257] The data collected by the server is formatted into a format that can be applied to machine learning. Specifically, the Python pandas library is used to convert the JSON data into a data frame. The input is the JSON-formatted game data obtained in step 1, and the output is a structured data frame with missing values ​​imputed. Next, the Go game record data is converted into coordinate format, and features of each position are extracted.

[0258] Step 3: Generate and train an AI model

[0259] The server uses the preprocessed data to build a deep learning model. It designs a neural network using frameworks such as TensorFlow or Keras. The input is the data frame shaped in step 2, and the output is a trained AI model. Specifically, it supplies the training data to the model and optimizes the model parameters using backpropagation.

[0260] Step 4: Optimize the model

[0261] The server evaluates and optimizes the generated model. Specifically, it uses scikit-learn to perform cross-validation and adjusts hyperparameters using grid search and random search. The trained AI model obtained in step 3 is used as input, and an optimized AI model is obtained as output. This prevents the model from overfitting and increases its versatility.

[0262] Step 5: Deploy and serve the model

[0263] The server uploads the optimized model to a cloud server and provides it to the user device via an API. Specifically, the model is deployed to Heroku or AWS Lambda, and a RESTful API is built using Flask or FastAPI. The optimized model obtained in step 4 is used as input, and the model provided to the user device via the API is obtained as output. The user device downloads the AI ​​model via this API and stores it locally.

[0264] Step 6: Implementing the Emotion Engine

[0265] The user's device uses a camera and microphone to collect the user's facial expressions and voice in real time. Specifically, the facial expression data is analyzed using OpenCV and Microsoft Azure's Face API. The input is real-time media data obtained from the camera and microphone, and the output is the user's emotional data. This emotional data is reflected in the AI's responses and feedback during the match.

[0266] Step 7: Match and Emotion Recognition

[0267] The user opens the app and begins playing against a professional copy AI. Specifically, the device uses an AI model downloaded from the cloud to predict the next move and presents it to the user. The input is the next move information entered by the user into the app and real-time emotional data, and the output is the optimal move and a response message based on the user's emotion. If the user becomes impatient or irritated, the device displays a message encouraging them to pause.

[0268] Step 8: Submitting and analyzing match results

[0269] When a user finishes a match, the match results are sent to the server. Specifically, the device converts the match results and emotional data into JSON format and sends it to the server via an API. The input is the user's match results and emotional data, and the output is the analysis results. Based on this, the server analyzes the user's weaknesses and areas for improvement during play and generates technical and mental feedback.

[0270] Step 9: Provide feedback

[0271] The server sends the analysis results and feedback to the user device. Specifically, the analysis results are generated in JSON format and sent to the user device via API. The analysis results from step 8 are used as input, and feedback is obtained as output, which is displayed on the user device. The user device displays the received feedback in a format that is easy for the user to understand, and provides technical advice and advice based on the user's emotional state.

[0272] (Application example 2)

[0273] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart device 14 will be referred to as a "terminal."

[0274] Conventional AI models based on data from professional players can provide users with technical feedback, but they have the problem of being unable to provide feedback or adjust game content that takes into account the user's emotional state. This can potentially reduce users' learning efficiency and satisfaction. The present invention aims to provide a system that recognizes a user's emotional state in real time and provides appropriate feedback and adjusts game content based on that emotion.

[0275] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for collecting data of professional players, a means for training a generative AI model based on the collected data of professional players, and a means for optimizing the generative AI model. This makes it possible to experience a realistic battle based on the skills of professional players.

[0276] The server further includes means for providing the optimized generative AI model to the user terminal, means for receiving and analyzing the results of the user's match against the generative AI model and providing feedback, and means for recognizing the user's emotional state in real time and adjusting the content of the match and feedback based on the user's emotions. This allows the user to receive feedback that takes into account their own emotional state, improving learning efficiency and satisfaction.

[0277] "Professional player data" refers to information that indicates the match records and playing styles of professional players in sports such as Go, Shogi, Chess, or card games.

[0278] A "generative AI model" is an AI model built using machine learning based on data from professional players, and has the ability to predict the next move during a match.

[0279] "Optimization" refers to the process of evaluating the performance of a generated artificial intelligence model and using machine learning techniques to improve the model's accuracy.

[0280] "User terminal" refers to any device used by a user, such as a smartphone, smart glasses, head-mounted display, or robot.

[0281] "Emotional state" refers to the psychological state that a user exhibits during a match, such as anxiety, irritation, or concentration.

[0282] "Feedback" is information on improvements and advice for the user's playing style that is provided based on the results of matches between the AI ​​model and the user.

[0283] "Deep learning technology" is a type of machine learning technology used to train artificial intelligence models, and is a method of learning complex patterns and features using multi-layer neural networks.

[0284] An "algorithm" refers to a procedure or computational method for solving a specific problem, and in this case refers to a series of processing steps used to analyze a user's emotional state.

[0285] A "cloud server" refers to a server that provides large-scale computer resources that can be remotely accessed via the Internet.

[0286] This invention is a system that collects data from professional players and generates a generative AI model based on that data. In this system, the user terminal recognizes the user's emotional state in real time and provides appropriate feedback corresponding to that emotion. The specific configuration and operating procedure are shown below.

[0287] System Configuration

[0288] 1. Data collection methods:

[0289] The server collects data on professional players of games such as Go, Shogi, Chess, and card games, including game records and playing styles. The data is obtained from online databases or official websites and stored in a database.

[0290] 2. Means of generating artificial intelligence models:

[0291] The server trains an artificial intelligence model based on the collected data. Deep learning technology is used to design a neural network and train it to reproduce the playing style of a professional. Frameworks such as TensorFlow and Keras are used.

[0292] 3. Optimization methods for artificial intelligence models:

[0293] The server evaluates the performance of the generated AI model and performs optimization, improving the model's accuracy through cross-validation and hyperparameter tuning.

[0294] 4. Means of provision to user terminal:

[0295] The server uploads the optimized AI model to a cloud server and provides it to the user's device via an API, allowing the user to play against the AI ​​model even offline.

[0296] 5. Emotion recognition means:

[0297] The user device uses a camera and microphone to collect the user's facial expressions and voice in real time, and uses algorithms to recognize the user's emotional state. This data is reflected in AI responses and feedback during the match.

[0298] 6. Competition and feedback channels:

[0299] The user opens the app and begins a match against a professional copy AI. During the match, the user's device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. After the match ends, the match results and emotion data are sent to the server, which provides analysis results and feedback.

[0300] Hardware and software used

[0301] Hardware: Smartphones, smart glasses, head-mounted displays, robots, webcams

[0302] Software: TensorFlow, Keras, facial_expression_recognition library

[0303] Examples of concrete examples and prompts

[0304] For example, in a Go match, a user opens a smartphone app and taps the "Start Match" button. During the match, the camera captures the user's face and analyzes their facial expressions. If the user shows signs of anxiety or irritation, the app displays a message saying, "Would you like to take a break?" After the match ends, the results and emotional data are sent to a server, and technical and psychological feedback is received.

[0305] Example prompt sentence:

[0306] "If the user's emotions show signs of anxiety or irritation during a Go match, implement a system that displays an appropriate message to encourage the user to take a break."

[0307] With the above system, users can learn the techniques of professional players while also learning efficiently while controlling their own emotions.

[0308] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0309] Step 1:

[0310] The server collects data from professional players. Specifically, it retrieves match records for Go, Shogi, chess, and card games from online databases and official websites, and stores the data in an internal database. The input is the match record data of professional players, and the output is formatted match record data. To ensure data consistency and quality, data processing such as format conversion and deletion of duplicate data is performed.

[0311] Step 2:

[0312] The server trains an AI model based on the collected data. Specifically, the collected match records are input into a neural network using deep learning frameworks such as TensorFlow and Keras, and training is performed to reproduce the playing style of professional players. The input is formatted match record data, and the output is a trained AI model. Here, data calculations such as model construction, preprocessing of learning data, and training are performed.

[0313] Step 3:

[0314] The server evaluates and optimizes the performance of the generated AI model. Specifically, it evaluates the predictive accuracy of the trained model and improves the model's accuracy using cross-validation and hyperparameter tuning. The input is the trained AI model and evaluation data, and the output is an optimized AI model. Parameters are adjusted to prevent overfitting and increase versatility.

[0315] Step 4:

[0316] The server uploads the optimized AI model to the cloud server and provides it to the user's device via API. The input is the optimized AI model, and the output is the model distribution to the user's device. The model is downloaded to the user's device so that the user can play offline.

[0317] Step 5:

[0318] The user device uses a camera and microphone to collect the user's facial expressions and voice in real time and recognize their emotional state. Specifically, the images captured by the camera are input into an emotion recognition algorithm to classify the user's emotional state. The input is the user's facial images and voice data captured in real time, and the output is the user's emotional state data. Here, facial expression analysis and voice analysis are performed.

[0319] Step 6:

[0320] A user starts a match against an artificial intelligence model, and the user's device predicts the AI's next move during the match. The input is the user's movements during the match (such as the board state), and the output is the AI's next move. During the match, the emotion engine monitors the user's emotional state in real time and adjusts the AI's response and feedback content as necessary. This is where real-time predictions and emotional response adjustments take place.

[0321] Step 7:

[0322] After a user finishes a match, the match results and the user's emotional state data are sent to the server. The server analyzes this data and generates feedback including the user's weaknesses and areas for improvement during play. The input is the match results and emotional state data, and the output is the analysis results and feedback. Data analysis and feedback generation are performed.

[0323] Step 8:

[0324] The server sends the generated feedback to the user's device, which then displays it to the user. The input is the analysis results and feedback, and the output is the feedback display to the user. The feedback includes not only technical advice but also mental support based on the user's emotional state. This allows the user to improve both their playing skills and emotional management.

[0325] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0326] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0327] In the above embodiment, an example in which the specific process is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific process may be performed by the smart device 14.

[0328] [Second embodiment]

[0329] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0330] 3, the data processing system 210 includes the data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0331] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0332] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication I / F 44. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, and the camera 42 are also connected to the bus 52.

[0333] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0334] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0335] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0336] Fig. 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Fig. 4, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0337] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0338] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0339] In the smart glasses 214, the reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0340] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal."

[0341] The present invention is a system that allows users to use an artificial intelligence model created based on collected data from professional players. With this system, users can experience playing against professional players and receive accurate feedback after the match.

[0342] Specific processing of the program

[0343] Professional data collection

[0344] The server collects data on professional players. For example, it retrieves the game history of a professional Go player from an online database or an official website. The data is recorded in coordinate format and used for subsequent processing.

[0345] Data Preprocessing

[0346] The server converts the collected data into a format that can be applied to machine learning. This process extracts features that represent the moves and strategic patterns at each stage. For example, Go game records can be converted into coordinate data, and the overall game progress and the strength of the moves can be quantified.

[0347] AI model generation and training

[0348] The server uses the formatted data to train a generative AI model. Deep learning techniques are used to build a model that reproduces the playing style of a professional. For example, a neural network is built using Keras or TensorFlow, and a model is trained to predict the next move from professional game records.

[0349] Model optimization

[0350] The server uses cross-validation and hyperparameter tuning to optimize the performance of the generated AI model, thereby avoiding overfitting and providing a model that can maintain high accuracy in actual games.

[0351] Deploying and serving the model

[0352] The server uploads the optimized AI model to the cloud server and provides it to the user's device via API. The app installed on the user's device downloads the AI ​​model via this API and stores it locally, allowing users to play against the AI ​​anytime, anywhere.

[0353] Match and Feedback

[0354] The user uses the app to play against the AI ​​model. The device predicts the AI's next move and displays its response. After the match ends, the user sends the match data to the server. The server analyzes the match data, generates feedback on the user's weaknesses and areas for improvement during play, and displays the feedback on the user's device. This feedback helps the user improve their skills.

[0355] Specific examples

[0356] For example, in a Go match, a user opens the smartphone app and taps the start button. The device uses an AI model downloaded from the cloud to suggest the next move. The user continues to make moves in response to the move, progressing the match. After the match ends, the user sends the match log to the server and receives analysis results and feedback in real time. This feedback includes where the wrong move was made, the reasons for the move, and alternatives. By providing an experience that mimics playing against a professional player and specific advice for improving their skills, users can hone their skills efficiently.

[0357] The processing flow will be explained below.

[0358] Step 1:

[0359] The server collects data on professional players, specifically, obtaining their game history and strategy data from online databases and official websites, and storing the data in a database.

[0360] Step 2:

[0361] The server converts the collected data into a format that can be applied to machine learning. For example, Go game record data is converted into coordinate format and the features of each position are extracted. This process maintains the consistency and quality of the data.

[0362] Step 3:

[0363] The server uses the preprocessed data to build a deep learning model, specifically, designing a neural network using frameworks such as TensorFlow and Keras, and training it to reproduce the playing style of a professional.

[0364] Step 4:

[0365] The server evaluates and optimizes the performance of the generated model. It performs cross-validation and hyperparameter tuning to prevent overfitting and improve versatility. The model is then modified based on the evaluation results.

[0366] Step 5:

[0367] The server uploads the optimized model to the cloud server, and an API is built to make it easy for users to access, and the model is provided through this API.

[0368] Step 6:

[0369] The user's device downloads the latest AI model from the cloud server and stores it locally, allowing the user to play against the AI ​​model even offline.

[0370] Step 7:

[0371] The user opens the app and begins playing against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The user then inputs their own next move and continues the match.

[0372] Step 8:

[0373] When a user finishes a match, the results are sent to the server, which analyzes the match data and identifies weaknesses and areas for improvement.

[0374] Step 9:

[0375] The server sends the analysis results and feedback to the user's device. The user's device displays the feedback, helping the user to easily review their game after it has finished. The feedback includes suggestions for mistakes and alternative methods, helping the user improve their skills.

[0376] Example 1

[0377] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0378] Conventional AI-based game systems have struggled to provide users with the same high-level experience as if they were playing against a professional player. Furthermore, they lacked mechanisms for providing detailed feedback to help users improve their skills. Furthermore, AI model optimization and model provisioning were often inefficient.

[0379] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0380] In this invention, the server includes means for collecting data of professional players, means for preprocessing data based on the collected data of professional players, means for training a generative AI model using the preprocessed data, means for optimizing the generated AI model using cross-validation and hyperparameter tuning, means for deploying the optimized generative AI model to a cloud environment and providing it to a user terminal, and means for a user to play against the generative AI model, receive and analyze the results of the game, and provide feedback. This allows users to enjoy an advanced experience similar to playing against professional players, and enables them to improve their skills through detailed feedback.

[0381] "Professional Player Data" means information about an expert or skilled player, including the results of previous matches or games.

[0382] "Data preprocessing" is the process of converting raw data into a form that can be applied to machine learning, including shaping the data, filling in missing values, and treating outliers.

[0383] A "generative artificial intelligence model" is an algorithm or model built using machine learning or deep learning techniques and trained to perform a specific task or prediction.

[0384] "Cross-validation" is a method of evaluating the performance of a machine learning model by dividing the data into multiple parts and checking the model's generalization ability.

[0385] "Hyperparameter tuning" is the process of adjusting parameters related to the model's settings and structure in order to optimize the performance of a machine learning model.

[0386] "Deploying to a cloud environment" means placing optimized models and applications on remote servers available via the Internet.

[0387] "User terminal" refers to a device that is directly operated by a user, and includes smartphones, PCs, tablets, etc.

[0388] "Feedback" refers to advice and comments provided to improve a user's skills based on an analysis of the user's match results.

[0389] The present invention is a system that allows users to use artificial intelligence (AI) models generated based on collected data from professional players. With this system, users can experience playing against professional players and receive accurate feedback after the match.

[0390] System Configuration

[0391] Data collection

[0392] The server collects data on professional players. Specifically, it retrieves the game history of professional Go players from online databases and official websites. This data is recorded in coordinate format and includes the coordinate information of each move, the time of each move, and the result of the game.

[0393] Data Preprocessing

[0394] The server preprocesses the collected data. This process converts raw data into a format that can be applied to machine learning. For example, Go game data is converted into coordinate data, and the overall game progress and the efficiency and impact of each move are quantified.

[0395] AI model generation and learning

[0396] The server uses the preprocessed data to generate and train an AI model. Deep learning techniques are used to build a neural network and create a model that reproduces the playing style of a professional. During this process, libraries such as Keras and TensorFlow are used to define the model, and the model's performance is evaluated using training and validation data.

[0397] Model Optimization

[0398] The server optimizes the model through cross-validation and hyperparameter tuning. This prevents overfitting and provides a model that can maintain high accuracy in actual games. Specifically, it uses methods such as GridSearchCV and RandomSearchCV to find the optimal hyperparameters.

[0399] Deploying and serving the model

[0400] The server deploys the optimized AI model to a cloud environment and provides it to the user's device via an API. Applications installed on the user's device download the model via the API and store it locally. This process allows users to use the latest AI model.

[0401] Match and Feedback

[0402] The user uses the app to play against an AI model. During the match, the device calculates and displays the AI's next move. After the match ends, the user sends the match log to the server. The server analyzes the match data and generates feedback on the user's weaknesses and areas for improvement during play, which is displayed on the user's device. This feedback is provided in detail to help the user improve their skills.

[0403] Specific examples

[0404] For example, in a Go match, the user opens the smartphone app and taps the start button. The device uses an AI model downloaded from the cloud to display the next move. The user plays moves in response to the move, progressing the match. After the match ends, the user sends the match log to the server and receives analysis results and feedback in real time. This feedback includes where the wrong move was made, the reasons for the move, and alternatives. By providing an experience that feels like playing against a professional player and specific advice for improving their skills, users can efficiently hone their skills.

[0405] Prompt Sentence Examples

[0406] "Build an AI model that predicts the next move from Go game data. The data used is the game history of professional Go players, and create a model using deep learning. Use Keras or TensorFlow to build a neural network that replicates the playing style of a professional."

[0407] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0408] Step 1: Data collection

[0409] The server collects data from professional players. The input data is collected from online databases and official website APIs. Specifically, the server calls the API to retrieve the game history of the professional players. This data includes the coordinate information of each move, the time of each move, the result of the game, etc. The output data is stored in a local database.

[0410] Step 2: Data Preprocessing

[0411] The data collected by the server is preprocessed. The input is the raw data collected in step 1. Specific operations include shaping the data, filling in missing values, and filtering outliers. For example, Go game record data is converted into coordinate format, and the overall game progress and the efficiency and impact of each move are quantified. The output is preprocessed data that can be applied to machine learning.

[0412] Step 3: AI model generation and training

[0413] The server uses the preprocessed data to generate and train an AI model. The input is the preprocessed data obtained in step 2. Specifically, it uses deep learning techniques to build a neural network and defines the model using Keras or TensorFlow. It evaluates the model's performance using training and validation data and learns optimal weights. The output is an initial AI model.

[0414] Step 4: Model optimization

[0415] The server optimizes the generated AI model. The input is the initial AI model generated in step 3. Specifically, it performs cross-validation and hyperparameter tuning to find optimal hyperparameters using techniques such as GridSearchCV and RandomSearchCV. The output is an optimized AI model.

[0416] Step 5: Deploy and serve the model

[0417] The server deploys the optimized AI model to the cloud environment. The input is the AI ​​model optimized in step 4. Specifically, the model is uploaded to the cloud server and published as an API endpoint. The user device then downloads the AI ​​model via this API and saves it locally. The output is the AI ​​model on the user device.

[0418] Step 6: Match and feedback

[0419] The user uses the app to play against an AI model. The input is the downloaded AI model. Specifically, the user taps the start button on the app, and the device calculates and displays the AI's next move. After the match ends, the user sends the match log to the server. The server analyzes the match data and generates feedback on the user's weaknesses and areas for improvement during play, which is displayed on the user's device. The output is detailed feedback and advice on how to improve the user's skills.

[0420] (Application example 1)

[0421] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0422] In traditional manufacturing, it has been difficult for new employees and employees in training to efficiently learn the skills of professional workers. In particular, there is a need to pass on the skilled techniques and know-how of professional workers and improve production efficiency while maintaining consistent quality. However, current training methods require time and effort and pose a risk of affecting the operating rate of the production line. Therefore, a new system is needed to efficiently and reliably transfer the skills and know-how of professional workers to new employees.

[0423] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0424] In this invention, the server includes means for collecting data of professional workers, means for training a generative AI model based on the collected data of professional workers, means for optimizing the generative AI model, means for providing the optimized generative AI model to a user terminal, and means for receiving and analyzing the results of a user's competition or collaboration with the generative AI model and providing feedback. This enables new employees and employees in training to learn skills effectively and efficiently by competing or collaborating with professional workers and receiving feedback based on real-life experience.

[0425] A "professional worker" refers to an individual who is engaged in manufacturing work with advanced specialized knowledge and skills and who has a certain level of proficiency.

[0426] "Data" refers to information such as operation records, measurement values, and operation logs generated when professional workers perform their work.

[0427] A "generative artificial intelligence model" refers to an artificial intelligence system that is generated by applying machine learning algorithms based on collected data.

[0428] "Learning" refers to the process by which an AI model uses collected data to understand behaviors and patterns and improve its capabilities.

[0429] "Optimization" refers to the process of making adjustments to an artificial intelligence model to maximize its performance and minimize error.

[0430] "User terminal" refers to a device for using an artificial intelligence model, and specifically includes hardware such as smartphones, tablets, PCs, and robots.

[0431] "Competition or collaboration" refers to the process in which a user actually works or collaborates with an artificial intelligence model.

[0432] "Receiving" refers to a user obtaining data generated as a result of a match or collaboration.

[0433] "Analysis" refers to the detailed examination of the results of a competition or collaboration and the processing of data to understand its significance and patterns.

[0434] "Feedback" refers to information that indicates evaluations and areas for improvement based on a user's actions and results.

[0435] The system of the present invention consists of the following main steps: First, collect data from professional workers, generate and optimize a generative AI model based on the data, and then provide the model to the user terminal, analyze the results of the competitive or collaborative play, and provide feedback.

[0436] 1. Professional data collection

[0437] The server automatically collects various data, such as motion records, measurements, and operation logs, from professional workers as they perform their tasks. This data is collected using high-performance sensors, cameras, and other measuring devices. This data is then aggregated in one place and used as a basic dataset for machine learning.

[0438] 2. Data preprocessing and feature extraction

[0439] The server preprocesses the collected data and formats it into a format suitable for machine learning. Specifically, it normalizes the data, fills in missing values, extracts features, etc. This process prepares the data in a way that maximizes the performance of the AI ​​model.

[0440] 3. Generating and training AI models

[0441] The server uses the preprocessed data to train a generative artificial intelligence model using deep learning techniques (specifically TensorFlow and Keras), which imitates the actions and patterns of professional workers and develops highly accurate prediction capabilities.

[0442] 4. Optimize and deploy the model

[0443] The server optimizes the generated AI model using cross-validation and hyperparameter tuning. The optimized model is then uploaded to a cloud server and provided to user devices via API. User devices can be a variety of devices, including smartphones, tablets, PCs, and robots.

[0444] 5. Competition and Feedback

[0445] Users compete or collaborate with the AI ​​model through the application. The results of the competition or collaboration are sent to the server in real time. The server analyzes the results and generates detailed feedback based on the user's actions and performance. This feedback is displayed on the user's device, allowing the user to effectively improve their skills and work efficiency.

[0446] Examples:

[0447] For example, when new employees undergo training to operate a robot arm on a manufacturing site, the system uses an AI model generated by a server. This system allows the AI ​​model, which has the same skills as a professional worker, to support the new employee's training. After the training session is over, the system provides analysis results and feedback, specifically indicating which movements the new employee needs to improve. This allows employees to smoothly improve their skills.

[0448] Example prompt sentence:

[0449] "I want to design a training system for new employees. How can I build a system that collects data from professional workers, generates an AI model based on that data, and provides real-time feedback to employees?"

[0450] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0451] Step 1:

[0452] The server collects data from professional workers. Specifically, it automatically acquires data such as operation records, measurements, and operation logs from sensors, cameras, and other measuring devices used by professional workers in the manufacturing process and stores the data in a database.

[0453] Input: Real-time data from sensors and cameras.

[0454] Output: Data of professional workers accumulated in a database.

[0455] Specific operation: Collecting data from various sensors and storing it in a database in real time.

[0456] Step 2:

[0457] The server preprocesses the collected data and formats it into a format suitable for machine learning. Specifically, it normalizes the data, fills in missing values, and extracts necessary features.

[0458] Input: Raw data in the database.

[0459] Output: A dataset formatted for machine learning.

[0460] Specific operations: Normalize the data, impute missing values, and select and extract the features to use.

[0461] Step 3:

[0462] The server uses the preprocessed data to train a generative artificial intelligence model. Specifically, it builds a neural network model using a deep learning framework (e.g., TensorFlow, Keras) and trains it using the collected data.

[0463] Input: The preprocessed dataset.

[0464] Output: A trained artificial intelligence model.

[0465] Specific operations: Set up the neural network architecture and train the model while optimizing the learning parameters.

[0466] Step 4:

[0467] The server optimizes the generated artificial intelligence model, specifically by performing cross-validation and hyperparameter tuning to improve the accuracy of the model.

[0468] Input: A trained artificial intelligence model.

[0469] Output: An optimized artificial intelligence model.

[0470] What it does: It performs cross-validation to find the optimal set of hyperparameters and tune them to maximize the model's accuracy.

[0471] Step 5:

[0472] The server uploads the optimized artificial intelligence model to a cloud server and provides it to the user's device via an API.

[0473] Input: An optimized artificial intelligence model.

[0474] Output: An artificial intelligence model that is downloaded to the user's device.

[0475] Specific operation: Deploying a model to a cloud server and downloading the model via an API call from the user's device.

[0476] Step 6:

[0477] Users use their devices to compete or collaborate with artificial intelligence models. Specifically, users operate the application and work in conjunction with the AI ​​model to carry out manufacturing tasks.

[0478] Input: User operation instructions.

[0479] Output: Work assistance data generated by the AI ​​model.

[0480] Specific behavior: The user performs an operation through the application, and the AI ​​model responds with assistance and instructions.

[0481] Step 7:

[0482] The server analyzes the results of the match or collaboration and provides feedback. Specifically, it analyzes the user's operation log in detail, generates advice on areas for improvement and technical enhancement, and provides it to the user's device.

[0483] Input: User operation log.

[0484] Output: Feedback content.

[0485] Specific operation: Analyzes operation logs using analytical algorithms and generates and returns feedback based on the results.

[0486] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0487] This invention is a system that collects data from professional players, generates an artificial intelligence model based on that data, and allows users to use it. It also recognizes the user's emotions in real time and adjusts the content of the match and feedback. Users can experience playing against professional players, receive accurate feedback after the match, and receive instruction that takes their own emotions into consideration.

[0488] Specific processing of the program

[0489] Professional data collection

[0490] The server collects data on professional players. For example, it retrieves the game history of a professional Go player from an online database or an official website and stores the data in a database.

[0491] Data Preprocessing

[0492] The server formats the collected data into a format that can be applied to machine learning. It converts the Go game data into coordinate format and extracts the features of each position. This ensures the consistency and quality of the data.

[0493] AI model generation and training

[0494] The server uses the preprocessed data to build a deep learning model. It designs a neural network using frameworks such as TensorFlow and Keras, and trains it to reproduce the playing style of a professional. The AI ​​model has the ability to predict the next move from the professional game record data.

[0495] Model optimization

[0496] The server evaluates and optimizes the performance of the generated model. It performs cross-validation and hyperparameter tuning to prevent overfitting and improve versatility. The model is then adjusted based on the evaluation results.

[0497] Deploying and serving the model

[0498] The server uploads the optimized model to a cloud server and provides it to the user's device via an API. The app installed on the user's device downloads the AI ​​model via this API and stores it locally, allowing users to play against the AI ​​model even offline.

[0499] Introducing the Emotion Engine

[0500] The user's device uses a camera and microphone to collect the user's facial expressions and voice in real time. The device processes the user's emotional data and uses a specific algorithm to recognize the user's emotional state. This emotional data is reflected in the AI's responses and feedback during the match.

[0501] Battles and Emotion Recognition

[0502] A user opens the app and begins a match against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. For example, if the user is feeling impatient or irritated, the AI ​​can display a message encouraging the user to pause.

[0503] Sending and analyzing match results

[0504] Once a user has finished a match, the results are sent to the server. The server analyzes the match data and identifies weaknesses and areas for improvement in the user's play. At the same time, feedback based on emotional data is also generated. This allows the user to receive guidance not only on technical but also psychological aspects.

[0505] Providing Feedback

[0506] The server sends the analysis results and feedback to the user's device, which displays the feedback to help the user review the game afterward. The feedback includes advice on mistakes, alternatives, and even advice based on the user's emotional state.

[0507] Specific examples

[0508] For example, in a Go match, a user opens a smartphone app and taps the start button. The device uses an AI model downloaded from the cloud to suggest the next move. During the match, the user's face is captured on camera, and their facial expression data is analyzed by the emotion engine. The engine determines in real time whether the user is concentrating on the match or feeling anxious or irritated, and suggests an appropriate response. After the match ends, the user sends the match log and emotional data to the server, and receives the analysis results along with technical and psychological advice in real time. In this way, users can learn professional techniques while also mastering how to control their own mental state.

[0509] The processing flow will be explained below.

[0510] Step 1:

[0511] The server collects data on professional players, specifically, obtaining their game history and strategy data from online databases and official websites, and storing the data in a database.

[0512] Step 2:

[0513] The server converts the collected data into a format that can be applied to machine learning. For example, it converts Go game record data into coordinate format and extracts the features of each position. This ensures the consistency and quality of the data.

[0514] Step 3:

[0515] The server uses the preprocessed data to build a deep learning model. It designs a neural network using frameworks such as TensorFlow and Keras, and trains it to reproduce the playing style of a professional. The AI ​​model has the ability to predict the next move from the professional game record data.

[0516] Step 4:

[0517] The server evaluates and optimizes the performance of the generated model. It performs cross-validation and hyperparameter tuning to prevent overfitting and improve versatility. The model is then adjusted based on the evaluation results.

[0518] Step 5:

[0519] The server uploads the optimized model to the cloud server, and an API is built to make it easy for users to access, and the model is provided through this API.

[0520] Step 6:

[0521] The user's device downloads the latest AI model from the cloud server and stores it locally, allowing the user to play against the AI ​​model even offline.

[0522] Step 7:

[0523] The user device uses a camera and microphone to collect the user's facial expressions and voice in real time, and the collected data is sent to an emotion engine to analyze the user's emotional state.

[0524] Step 8:

[0525] The user opens the app and begins a match against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. For example, if the user becomes impatient, the AI ​​displays a message suggesting a pause.

[0526] Step 9:

[0527] The user's device records the user's emotional data and battle data during the match, allowing the user to track emotional changes along with the match history.

[0528] Step 10:

[0529] After a user finishes a match, the match results and emotional data are sent to the server, which analyzes the match data and identifies weaknesses and areas for improvement.

[0530] Step 11:

[0531] The server sends the analysis results and feedback to the user's device. The feedback includes error detection, alternative solutions, and advice based on the user's emotional state. This provides not only technical guidance but also emotional and mental support.

[0532] In this way, users can learn professional techniques while also mastering how to control their own mental state, thereby efficiently improving their skills.

[0533] Example 2

[0534] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0535] Conventional technology not only provides users with the experience of playing against professional players, but also fails to provide feedback that reflects their emotional state during the match in real time. As a result, while users can learn technical skills, it is difficult for them to improve their mental control or emotional state. To solve this problem, a system is needed that collects users' emotional data and adjusts the content of the match and feedback based on that data.

[0536] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for collecting data of professional players, a means for training an AI model to be generated based on the collected data of professional players, and a means for optimizing the generated AI model. This makes it possible to faithfully reproduce the playing style of a professional and provide it to the user. Furthermore, the user terminal includes a means for collecting and analyzing user emotional data using a camera and microphone, a means for adjusting the content of the match and feedback based on the user emotional data in real time, and a means for receiving, analyzing, and providing feedback on the results of the user's match against the generated AI model. This allows the user to receive guidance on both technical and mental aspects.

[0537] "Professional Player Data" means data including the match history, playing style, and related statistical information of players who are recognized as professionals in the game.

[0538] An "artificial intelligence model" is a mathematical structure trained using machine learning techniques, and refers to a system that makes predictions and judgments based on input data.

[0539] A "user terminal" is a device, such as a personal computer, smartphone, or tablet, that is directly operated by a user to send and receive information.

[0540] "Emotion data" is data that indicates the psychological state of the user, obtained from facial expressions, voice, etc., and is quantified or classified using an emotion analysis engine.

[0541] "Feedback" refers to guidance and advice provided based on the results of a match or the user's performance, and includes advice on technical improvements and mental aspects.

[0542] "Collection means" refers to software and hardware configurations for obtaining data via the Internet.

[0543] A "training tool" is a process and tool for analyzing data and training a model using machine learning algorithms.

[0544] "Optimization means" refers to a method of adjusting and evaluating the generated artificial intelligence model to improve its performance, including hyperparameter adjustment and cross-validation.

[0545] "Means of providing" refers to a deployment method for making the optimized artificial intelligence model available to user devices, including cloud APIs and download links.

[0546] "Means of analysis" refers to the processes and tools used to analyze collected data statistically or through machine learning to extract meaningful information.

[0547] The present invention is a system that collects data from professional players and allows users to use an artificial intelligence model generated based on this data. This system also recognizes the user's emotions in real time and adjusts the match content and feedback, allowing users to experience playing against professional players. In addition, users receive accurate feedback after the match and can receive instruction that takes their own emotions into consideration.

[0548] First, the server collects data on professional players. For example, it retrieves the game history of a professional Go player from an online database or official website and stores that data in a database. Specific technologies used include web scraping using Python's BeautifulSoup and Scrapy, and storing the retrieved data in AWS RDS.

[0549] Next, the server converts the collected data into a format that can be applied to machine learning. For example, Go game data is converted into coordinate format, and the features of each position are extracted. At this stage, data processing libraries such as pandas and NumPy are used.

[0550] The server then uses the preprocessed data to build a deep learning model. Using deep learning frameworks such as TensorFlow and Keras, it designs a neural network and trains it to replicate the playing style of a professional. This gives the AI ​​model the ability to predict the next move from professional match data.

[0551] Cross-validation and hyperparameter tuning are required to evaluate and optimize the performance of the constructed model. The server performs cross-validation using scikit-learn and finds optimal hyperparameters using methods such as grid search and random search.

[0552] The optimized model is uploaded to a cloud server by the server and provided to user devices via an API. The server is built using Heroku or AWS Lambda, and a RESTful API is configured. User devices can download the AI ​​model via this API and save it locally. This allows users to play against the AI ​​model even offline.

[0553] The user's device uses a camera and microphone to collect the user's facial expressions and voice in real time. The facial expression data is analyzed using OpenCV or Microsoft Azure's Face API, and a specific algorithm is used to recognize the user's emotional state. This emotional data is reflected in the AI's responses and feedback during the match.

[0554] The user opens the app and begins a match against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. If the user becomes impatient or irritated, the device can display a message encouraging the user to pause.

[0555] After a match, the user sends the match results to the server. The server analyzes the match data and identifies the user's weaknesses and areas for improvement. At the same time, feedback based on emotional data is generated, allowing the user to receive guidance not only on technical but also psychological aspects.

[0556] This system not only allows users to experience playing against professional players, but also provides instruction on both technical and mental aspects, allowing users to learn professional techniques while also mastering their own mental control.

[0557] Prompt Sentence Examples

[0558] For example, "Please tell me the specific steps to develop a Go game application with emotion recognition capabilities using Python. Please explain how to use Azure's Face API to recognize the user's emotions and adjust the AI's response based on that data."

[0559] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0560] Step 1: Professional data collection

[0561] The server collects data on professional players. Specifically, it uses Python's BeautifulSoup and Scrapy to scrape the game history of professional players from online databases and official websites. Website URLs are used as input, and game data in JSON format is output. This data is then uploaded to AWS's RDS for efficient management and storage.

[0562] Step 2: Preprocessing the data

[0563] The data collected by the server is formatted into a format that can be applied to machine learning. Specifically, the Python pandas library is used to convert the JSON data into a data frame. The input is the JSON-formatted game data obtained in step 1, and the output is a structured data frame with missing values ​​imputed. Next, the Go game record data is converted into coordinate format, and features of each position are extracted.

[0564] Step 3: Generate and train an AI model

[0565] The server uses the preprocessed data to build a deep learning model. It designs a neural network using frameworks such as TensorFlow or Keras. The input is the data frame shaped in step 2, and the output is a trained AI model. Specifically, it supplies the training data to the model and optimizes the model parameters using backpropagation.

[0566] Step 4: Optimize the model

[0567] The server evaluates and optimizes the generated model. Specifically, it uses scikit-learn to perform cross-validation and adjusts hyperparameters using grid search and random search. The trained AI model obtained in step 3 is used as input, and an optimized AI model is obtained as output. This prevents the model from overfitting and increases its versatility.

[0568] Step 5: Deploy and serve the model

[0569] The server uploads the optimized model to a cloud server and provides it to the user device via an API. Specifically, the model is deployed to Heroku or AWS Lambda, and a RESTful API is built using Flask or FastAPI. The optimized model obtained in step 4 is used as input, and the model provided to the user device via the API is obtained as output. The user device downloads the AI ​​model via this API and stores it locally.

[0570] Step 6: Implementing the Emotion Engine

[0571] The user's device uses a camera and microphone to collect the user's facial expressions and voice in real time. Specifically, the facial expression data is analyzed using OpenCV and Microsoft Azure's Face API. The input is real-time media data obtained from the camera and microphone, and the output is the user's emotional data. This emotional data is reflected in the AI's responses and feedback during the match.

[0572] Step 7: Match and Emotion Recognition

[0573] The user opens the app and begins playing against a professional copy AI. Specifically, the device uses an AI model downloaded from the cloud to predict the next move and presents it to the user. The input is the next move information entered by the user into the app and real-time emotional data, and the output is the optimal move and a response message based on the user's emotion. If the user becomes impatient or irritated, the device displays a message encouraging them to pause.

[0574] Step 8: Submitting and analyzing match results

[0575] When a user finishes a match, the match results are sent to the server. Specifically, the device converts the match results and emotional data into JSON format and sends it to the server via an API. The input is the user's match results and emotional data, and the output is the analysis results. Based on this, the server analyzes the user's weaknesses and areas for improvement during play and generates technical and mental feedback.

[0576] Step 9: Provide feedback

[0577] The server sends the analysis results and feedback to the user device. Specifically, the analysis results are generated in JSON format and sent to the user device via API. The analysis results from step 8 are used as input, and feedback is obtained as output, which is displayed on the user device. The user device displays the received feedback in a format that is easy for the user to understand, and provides technical advice and advice based on the user's emotional state.

[0578] (Application example 2)

[0579] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the smart glasses 214 will be referred to as a "terminal."

[0580] Conventional AI models based on data from professional players can provide users with technical feedback, but they have the problem of being unable to provide feedback or adjust game content that takes into account the user's emotional state. This can potentially reduce users' learning efficiency and satisfaction. The present invention aims to provide a system that recognizes a user's emotional state in real time and provides appropriate feedback and adjusts game content based on that emotion.

[0581] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for collecting data of professional players, a means for training a generative AI model based on the collected data of professional players, and a means for optimizing the generative AI model. This makes it possible to experience a realistic battle based on the skills of professional players.

[0582] The server further includes means for providing the optimized generative AI model to the user terminal, means for receiving and analyzing the results of the user's match against the generative AI model and providing feedback, and means for recognizing the user's emotional state in real time and adjusting the content of the match and feedback based on the user's emotions. This allows the user to receive feedback that takes into account their own emotional state, improving learning efficiency and satisfaction.

[0583] "Professional player data" refers to information that indicates the match records and playing styles of professional players in sports such as Go, Shogi, Chess, or card games.

[0584] A "generative AI model" is an AI model built using machine learning based on data from professional players, and has the ability to predict the next move during a match.

[0585] "Optimization" refers to the process of evaluating the performance of a generated artificial intelligence model and using machine learning techniques to improve the model's accuracy.

[0586] "User terminal" refers to any device used by a user, such as a smartphone, smart glasses, head-mounted display, or robot.

[0587] "Emotional state" refers to the psychological state that a user exhibits during a match, such as anxiety, irritation, or concentration.

[0588] "Feedback" is information on improvements and advice for the user's playing style that is provided based on the results of matches between the AI ​​model and the user.

[0589] "Deep learning technology" is a type of machine learning technology used to train artificial intelligence models, and is a method of learning complex patterns and features using multi-layer neural networks.

[0590] An "algorithm" refers to a procedure or computational method for solving a specific problem, and in this case refers to a series of processing steps used to analyze a user's emotional state.

[0591] A "cloud server" refers to a server that provides large-scale computer resources that can be remotely accessed via the Internet.

[0592] This invention is a system that collects data from professional players and generates a generative AI model based on that data. In this system, the user terminal recognizes the user's emotional state in real time and provides appropriate feedback corresponding to that emotion. The specific configuration and operating procedure are shown below.

[0593] System Configuration

[0594] 1. Data collection methods:

[0595] The server collects data on professional players of games such as Go, Shogi, Chess, and card games, including game records and playing styles. The data is obtained from online databases or official websites and stored in a database.

[0596] 2. Means of generating artificial intelligence models:

[0597] The server trains an artificial intelligence model based on the collected data. Deep learning technology is used to design a neural network and train it to reproduce the playing style of a professional. Frameworks such as TensorFlow and Keras are used.

[0598] 3. Optimization methods for artificial intelligence models:

[0599] The server evaluates the performance of the generated AI model and performs optimization, improving the model's accuracy through cross-validation and hyperparameter tuning.

[0600] 4. Means of provision to user terminal:

[0601] The server uploads the optimized AI model to a cloud server and provides it to the user's device via an API, allowing the user to play against the AI ​​model even offline.

[0602] 5. Emotion recognition means:

[0603] The user device uses a camera and microphone to collect the user's facial expressions and voice in real time, and uses algorithms to recognize the user's emotional state. This data is reflected in AI responses and feedback during the match.

[0604] 6. Competition and feedback channels:

[0605] The user opens the app and begins a match against a professional copy AI. During the match, the user's device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. After the match ends, the match results and emotion data are sent to the server, which provides analysis results and feedback.

[0606] Hardware and software used

[0607] Hardware: Smartphones, smart glasses, head-mounted displays, robots, webcams

[0608] Software: TensorFlow, Keras, facial_expression_recognition library

[0609] Examples of concrete examples and prompts

[0610] For example, in a Go match, a user opens a smartphone app and taps the "Start Match" button. During the match, the camera captures the user's face and analyzes their facial expressions. If the user shows signs of anxiety or irritation, the app displays a message saying, "Would you like to take a break?" After the match ends, the results and emotional data are sent to a server, and technical and psychological feedback is received.

[0611] Example prompt sentence:

[0612] "If the user's emotions show signs of anxiety or irritation during a Go match, implement a system that displays an appropriate message to encourage the user to take a break."

[0613] With the above system, users can learn the techniques of professional players while also learning efficiently while controlling their own emotions.

[0614] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0615] Step 1:

[0616] The server collects data from professional players. Specifically, it retrieves match records for Go, Shogi, chess, and card games from online databases and official websites, and stores the data in an internal database. The input is the match record data of professional players, and the output is formatted match record data. To ensure data consistency and quality, data processing such as format conversion and deletion of duplicate data is performed.

[0617] Step 2:

[0618] The server trains an AI model based on the collected data. Specifically, the collected match records are input into a neural network using deep learning frameworks such as TensorFlow and Keras, and training is performed to reproduce the playing style of professional players. The input is formatted match record data, and the output is a trained AI model. Here, data calculations such as model construction, preprocessing of learning data, and training are performed.

[0619] Step 3:

[0620] The server evaluates and optimizes the performance of the generated AI model. Specifically, it evaluates the predictive accuracy of the trained model and improves the model's accuracy using cross-validation and hyperparameter tuning. The input is the trained AI model and evaluation data, and the output is an optimized AI model. Parameters are adjusted to prevent overfitting and increase versatility.

[0621] Step 4:

[0622] The server uploads the optimized AI model to the cloud server and provides it to the user's device via API. The input is the optimized AI model, and the output is the model distribution to the user's device. The model is downloaded to the user's device so that the user can play offline.

[0623] Step 5:

[0624] The user device uses a camera and microphone to collect the user's facial expressions and voice in real time and recognize their emotional state. Specifically, the images captured by the camera are input into an emotion recognition algorithm to classify the user's emotional state. The input is the user's facial images and voice data captured in real time, and the output is the user's emotional state data. Here, facial expression analysis and voice analysis are performed.

[0625] Step 6:

[0626] A user starts a match against an artificial intelligence model, and the user's device predicts the AI's next move during the match. The input is the user's movements during the match (such as the board state), and the output is the AI's next move. During the match, the emotion engine monitors the user's emotional state in real time and adjusts the AI's response and feedback content as necessary. This is where real-time predictions and emotional response adjustments take place.

[0627] Step 7:

[0628] After a user finishes a match, the match results and the user's emotional state data are sent to the server. The server analyzes this data and generates feedback including the user's weaknesses and areas for improvement during play. The input is the match results and emotional state data, and the output is the analysis results and feedback. Data analysis and feedback generation are performed.

[0629] Step 8:

[0630] The server sends the generated feedback to the user's device, which then displays it to the user. The input is the analysis results and feedback, and the output is the feedback display to the user. The feedback includes not only technical advice but also mental support based on the user's emotional state. This allows the user to improve both their playing skills and emotional management.

[0631] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0632] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0633] In the above embodiment, an example in which the specific processing is performed by the data processing device 12 has been given, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the smart glasses 214.

[0634] [Third embodiment]

[0635] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0636] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[0637] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0638] The headset type terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a display 343. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the display 343 are also connected to the bus 52.

[0639] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0640] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0641] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0642] Fig. 6 shows an example of the main functions of the data processing device 12 and the headset type terminal 314. As shown in Fig. 6, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0643] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0644] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0645] In the headset type terminal 314, a reception output process is performed by the processor 46. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0646] Next, a description will be given of the identification process performed by the identification processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as the "server" and the headset type terminal 314 will be referred to as the "terminal."

[0647] The present invention is a system that allows users to use an artificial intelligence model created based on collected data from professional players. With this system, users can experience playing against professional players and receive accurate feedback after the match.

[0648] Specific processing of the program

[0649] Professional data collection

[0650] The server collects data on professional players. For example, it retrieves the game history of a professional Go player from an online database or an official website. The data is recorded in coordinate format and used for subsequent processing.

[0651] Data Preprocessing

[0652] The server converts the collected data into a format that can be applied to machine learning. This process extracts features that represent the moves and strategic patterns at each stage. For example, Go game records can be converted into coordinate data, and the overall game progress and the strength of the moves can be quantified.

[0653] AI model generation and training

[0654] The server uses the formatted data to train a generative AI model. Deep learning techniques are used to build a model that reproduces the playing style of a professional. For example, a neural network is built using Keras or TensorFlow, and a model is trained to predict the next move from professional game records.

[0655] Model optimization

[0656] The server uses cross-validation and hyperparameter tuning to optimize the performance of the generated AI model, thereby avoiding overfitting and providing a model that can maintain high accuracy in actual games.

[0657] Deploying and serving the model

[0658] The server uploads the optimized AI model to the cloud server and provides it to the user's device via API. The app installed on the user's device downloads the AI ​​model via this API and stores it locally, allowing users to play against the AI ​​anytime, anywhere.

[0659] Match and Feedback

[0660] The user uses the app to play against the AI ​​model. The device predicts the AI's next move and displays its response. After the match ends, the user sends the match data to the server. The server analyzes the match data, generates feedback on the user's weaknesses and areas for improvement during play, and displays the feedback on the user's device. This feedback helps the user improve their skills.

[0661] Specific examples

[0662] For example, in a Go match, a user opens the smartphone app and taps the start button. The device uses an AI model downloaded from the cloud to suggest the next move. The user continues to make moves in response to the move, progressing the match. After the match ends, the user sends the match log to the server and receives analysis results and feedback in real time. This feedback includes where the wrong move was made, the reasons for the move, and alternatives. By providing an experience that mimics playing against a professional player and specific advice for improving their skills, users can hone their skills efficiently.

[0663] The processing flow will be explained below.

[0664] Step 1:

[0665] The server collects data on professional players, specifically, obtaining their game history and strategy data from online databases and official websites, and storing the data in a database.

[0666] Step 2:

[0667] The server converts the collected data into a format that can be applied to machine learning. For example, Go game record data is converted into coordinate format and the features of each position are extracted. This process maintains the consistency and quality of the data.

[0668] Step 3:

[0669] The server uses the preprocessed data to build a deep learning model, specifically, designing a neural network using frameworks such as TensorFlow and Keras, and training it to reproduce the playing style of a professional.

[0670] Step 4:

[0671] The server evaluates and optimizes the performance of the generated model. It performs cross-validation and hyperparameter tuning to prevent overfitting and improve versatility. The model is then modified based on the evaluation results.

[0672] Step 5:

[0673] The server uploads the optimized model to the cloud server, and an API is built to make it easy for users to access, and the model is provided through this API.

[0674] Step 6:

[0675] The user's device downloads the latest AI model from the cloud server and stores it locally, allowing the user to play against the AI ​​model even offline.

[0676] Step 7:

[0677] The user opens the app and begins playing against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The user then inputs their own next move and continues the match.

[0678] Step 8:

[0679] When a user finishes a match, the results are sent to the server, which analyzes the match data and identifies weaknesses and areas for improvement.

[0680] Step 9:

[0681] The server sends the analysis results and feedback to the user's device. The user's device displays the feedback, helping the user to easily review their game after it has finished. The feedback includes suggestions for mistakes and alternative methods, helping the user improve their skills.

[0682] Example 1

[0683] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0684] Conventional AI-based game systems have struggled to provide users with the same high-level experience as if they were playing against a professional player. Furthermore, they lacked mechanisms for providing detailed feedback to help users improve their skills. Furthermore, AI model optimization and model provisioning were often inefficient.

[0685] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0686] In this invention, the server includes means for collecting data of professional players, means for preprocessing data based on the collected data of professional players, means for training a generative AI model using the preprocessed data, means for optimizing the generated AI model using cross-validation and hyperparameter tuning, means for deploying the optimized generative AI model to a cloud environment and providing it to a user terminal, and means for a user to play against the generative AI model, receive and analyze the results of the game, and provide feedback. This allows users to enjoy an advanced experience similar to playing against professional players, and enables them to improve their skills through detailed feedback.

[0687] "Professional Player Data" means information about an expert or skilled player, including the results of previous matches or games.

[0688] "Data preprocessing" is the process of converting raw data into a form that can be applied to machine learning, including shaping the data, filling in missing values, and treating outliers.

[0689] A "generative artificial intelligence model" is an algorithm or model built using machine learning or deep learning techniques and trained to perform a specific task or prediction.

[0690] "Cross-validation" is a method of evaluating the performance of a machine learning model by dividing the data into multiple parts and checking the model's generalization ability.

[0691] "Hyperparameter tuning" is the process of adjusting parameters related to the model's settings and structure in order to optimize the performance of a machine learning model.

[0692] "Deploying to a cloud environment" means placing optimized models and applications on remote servers available via the Internet.

[0693] "User terminal" refers to a device that is directly operated by a user, and includes smartphones, PCs, tablets, etc.

[0694] "Feedback" refers to advice and comments provided to improve a user's skills based on an analysis of the user's match results.

[0695] The present invention is a system that allows users to use artificial intelligence (AI) models generated based on collected data from professional players. With this system, users can experience playing against professional players and receive accurate feedback after the match.

[0696] System Configuration

[0697] Data collection

[0698] The server collects data on professional players. Specifically, it retrieves the game history of professional Go players from online databases and official websites. This data is recorded in coordinate format and includes the coordinate information of each move, the time of each move, and the result of the game.

[0699] Data Preprocessing

[0700] The server preprocesses the collected data. This process converts raw data into a format that can be applied to machine learning. For example, Go game data is converted into coordinate data, and the overall game progress and the efficiency and impact of each move are quantified.

[0701] AI model generation and learning

[0702] The server uses the preprocessed data to generate and train an AI model. Deep learning techniques are used to build a neural network and create a model that reproduces the playing style of a professional. During this process, libraries such as Keras and TensorFlow are used to define the model, and the model's performance is evaluated using training and validation data.

[0703] Model Optimization

[0704] The server optimizes the model through cross-validation and hyperparameter tuning. This prevents overfitting and provides a model that can maintain high accuracy in actual games. Specifically, it uses methods such as GridSearchCV and RandomSearchCV to find the optimal hyperparameters.

[0705] Deploying and serving the model

[0706] The server deploys the optimized AI model to a cloud environment and provides it to the user's device via an API. Applications installed on the user's device download the model via the API and store it locally. This process allows users to use the latest AI model.

[0707] Match and Feedback

[0708] The user uses the app to play against an AI model. During the match, the device calculates and displays the AI's next move. After the match ends, the user sends the match log to the server. The server analyzes the match data and generates feedback on the user's weaknesses and areas for improvement during play, which is displayed on the user's device. This feedback is provided in detail to help the user improve their skills.

[0709] Specific examples

[0710] For example, in a Go match, the user opens the smartphone app and taps the start button. The device uses an AI model downloaded from the cloud to display the next move. The user plays moves in response to the move, progressing the match. After the match ends, the user sends the match log to the server and receives analysis results and feedback in real time. This feedback includes where the wrong move was made, the reasons for the move, and alternatives. By providing an experience that feels like playing against a professional player and specific advice for improving their skills, users can efficiently hone their skills.

[0711] Prompt Sentence Examples

[0712] "Build an AI model that predicts the next move from Go game data. The data used is the game history of professional Go players, and create a model using deep learning. Use Keras or TensorFlow to build a neural network that replicates the playing style of a professional."

[0713] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0714] Step 1: Data collection

[0715] The server collects data from professional players. The input data is collected from online databases and official website APIs. Specifically, the server calls the API to retrieve the game history of the professional players. This data includes the coordinate information of each move, the time of each move, the result of the game, etc. The output data is stored in a local database.

[0716] Step 2: Data Preprocessing

[0717] The data collected by the server is preprocessed. The input is the raw data collected in step 1. Specific operations include shaping the data, filling in missing values, and filtering outliers. For example, Go game record data is converted into coordinate format, and the overall game progress and the efficiency and impact of each move are quantified. The output is preprocessed data that can be applied to machine learning.

[0718] Step 3: AI model generation and training

[0719] The server uses the preprocessed data to generate and train an AI model. The input is the preprocessed data obtained in step 2. Specifically, it uses deep learning techniques to build a neural network and defines the model using Keras or TensorFlow. It evaluates the model's performance using training and validation data and learns optimal weights. The output is an initial AI model.

[0720] Step 4: Model optimization

[0721] The server optimizes the generated AI model. The input is the initial AI model generated in step 3. Specifically, it performs cross-validation and hyperparameter tuning to find optimal hyperparameters using techniques such as GridSearchCV and RandomSearchCV. The output is an optimized AI model.

[0722] Step 5: Deploy and serve the model

[0723] The server deploys the optimized AI model to the cloud environment. The input is the AI ​​model optimized in step 4. Specifically, the model is uploaded to the cloud server and published as an API endpoint. The user device then downloads the AI ​​model via this API and saves it locally. The output is the AI ​​model on the user device.

[0724] Step 6: Match and feedback

[0725] The user uses the app to play against an AI model. The input is the downloaded AI model. Specifically, the user taps the start button on the app, and the device calculates and displays the AI's next move. After the match ends, the user sends the match log to the server. The server analyzes the match data and generates feedback on the user's weaknesses and areas for improvement during play, which is displayed on the user's device. The output is detailed feedback and advice on how to improve the user's skills.

[0726] (Application example 1)

[0727] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0728] In traditional manufacturing, it has been difficult for new employees and employees in training to efficiently learn the skills of professional workers. In particular, there is a need to pass on the skilled techniques and know-how of professional workers and improve production efficiency while maintaining consistent quality. However, current training methods require time and effort and pose a risk of affecting the operating rate of the production line. Therefore, a new system is needed to efficiently and reliably transfer the skills and know-how of professional workers to new employees.

[0729] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[0730] In this invention, the server includes means for collecting data of professional workers, means for training a generative AI model based on the collected data of professional workers, means for optimizing the generative AI model, means for providing the optimized generative AI model to a user terminal, and means for receiving and analyzing the results of a user's competition or collaboration with the generative AI model and providing feedback. This enables new employees and employees in training to learn skills effectively and efficiently by competing or collaborating with professional workers and receiving feedback based on real-life experience.

[0731] A "professional worker" refers to an individual who is engaged in manufacturing work with advanced specialized knowledge and skills and who has a certain level of proficiency.

[0732] "Data" refers to information such as operation records, measurement values, and operation logs generated when professional workers perform their work.

[0733] A "generative artificial intelligence model" refers to an artificial intelligence system that is generated by applying machine learning algorithms based on collected data.

[0734] "Learning" refers to the process by which an AI model uses collected data to understand behaviors and patterns and improve its capabilities.

[0735] "Optimization" refers to the process of making adjustments to an artificial intelligence model to maximize its performance and minimize error.

[0736] "User terminal" refers to a device for using an artificial intelligence model, and specifically includes hardware such as smartphones, tablets, PCs, and robots.

[0737] "Competition or collaboration" refers to the process in which a user actually works or collaborates with an artificial intelligence model.

[0738] "Receiving" refers to a user obtaining data generated as a result of a match or collaboration.

[0739] "Analysis" refers to the detailed examination of the results of a competition or collaboration and the processing of data to understand its significance and patterns.

[0740] "Feedback" refers to information that indicates evaluations and areas for improvement based on a user's actions and results.

[0741] The system of the present invention consists of the following main steps: First, collect data from professional workers, generate and optimize a generative AI model based on the data, and then provide the model to the user terminal, analyze the results of the competitive or collaborative play, and provide feedback.

[0742] 1. Professional data collection

[0743] The server automatically collects various data, such as motion records, measurements, and operation logs, from professional workers as they perform their tasks. This data is collected using high-performance sensors, cameras, and other measuring devices. This data is then aggregated in one place and used as a basic dataset for machine learning.

[0744] 2. Data preprocessing and feature extraction

[0745] The server preprocesses the collected data and formats it into a format suitable for machine learning. Specifically, it normalizes the data, fills in missing values, extracts features, etc. This process prepares the data in a way that maximizes the performance of the AI ​​model.

[0746] 3. Generating and training AI models

[0747] The server uses the preprocessed data to train a generative artificial intelligence model using deep learning techniques (specifically TensorFlow and Keras), which imitates the actions and patterns of professional workers and develops highly accurate prediction capabilities.

[0748] 4. Optimize and deploy the model

[0749] The server optimizes the generated AI model using cross-validation and hyperparameter tuning. The optimized model is then uploaded to a cloud server and provided to user devices via API. User devices can be a variety of devices, including smartphones, tablets, PCs, and robots.

[0750] 5. Competition and Feedback

[0751] Users compete or collaborate with the AI ​​model through the application. The results of the competition or collaboration are sent to the server in real time. The server analyzes the results and generates detailed feedback based on the user's actions and performance. This feedback is displayed on the user's device, allowing the user to effectively improve their skills and work efficiency.

[0752] Examples:

[0753] For example, when new employees undergo training to operate a robot arm on a manufacturing site, the system uses an AI model generated by a server. This system allows the AI ​​model, which has the same skills as a professional worker, to support the new employee's training. After the training session is over, the system provides analysis results and feedback, specifically indicating which movements the new employee needs to improve. This allows employees to smoothly improve their skills.

[0754] Example prompt sentence:

[0755] "I want to design a training system for new employees. How can I build a system that collects data from professional workers, generates an AI model based on that data, and provides real-time feedback to employees?"

[0756] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0757] Step 1:

[0758] The server collects data from professional workers. Specifically, it automatically acquires data such as operation records, measurements, and operation logs from sensors, cameras, and other measuring devices used by professional workers in the manufacturing process and stores the data in a database.

[0759] Input: Real-time data from sensors and cameras.

[0760] Output: Data of professional workers accumulated in a database.

[0761] Specific operation: Collecting data from various sensors and storing it in a database in real time.

[0762] Step 2:

[0763] The server preprocesses the collected data and formats it into a format suitable for machine learning. Specifically, it normalizes the data, fills in missing values, and extracts necessary features.

[0764] Input: Raw data in the database.

[0765] Output: A dataset formatted for machine learning.

[0766] Specific operations: Normalize the data, impute missing values, and select and extract the features to use.

[0767] Step 3:

[0768] The server uses the preprocessed data to train a generative artificial intelligence model. Specifically, it builds a neural network model using a deep learning framework (e.g., TensorFlow, Keras) and trains it using the collected data.

[0769] Input: The preprocessed dataset.

[0770] Output: A trained artificial intelligence model.

[0771] Specific operations: Set up the neural network architecture and train the model while optimizing the learning parameters.

[0772] Step 4:

[0773] The server optimizes the generated artificial intelligence model, specifically by performing cross-validation and hyperparameter tuning to improve the accuracy of the model.

[0774] Input: A trained artificial intelligence model.

[0775] Output: An optimized artificial intelligence model.

[0776] What it does: It performs cross-validation to find the optimal set of hyperparameters and tune them to maximize the model's accuracy.

[0777] Step 5:

[0778] The server uploads the optimized artificial intelligence model to a cloud server and provides it to the user's device via an API.

[0779] Input: An optimized artificial intelligence model.

[0780] Output: An artificial intelligence model that is downloaded to the user's device.

[0781] Specific operation: Deploying a model to a cloud server and downloading the model via an API call from the user's device.

[0782] Step 6:

[0783] Users use their devices to compete or collaborate with artificial intelligence models. Specifically, users operate the application and work in conjunction with the AI ​​model to carry out manufacturing tasks.

[0784] Input: User operation instructions.

[0785] Output: Work assistance data generated by the AI ​​model.

[0786] Specific behavior: The user performs an operation through the application, and the AI ​​model responds with assistance and instructions.

[0787] Step 7:

[0788] The server analyzes the results of the match or collaboration and provides feedback. Specifically, it analyzes the user's operation log in detail, generates advice on areas for improvement and technical enhancement, and provides it to the user's device.

[0789] Input: User operation log.

[0790] Output: Feedback content.

[0791] Specific operation: Analyzes operation logs using analytical algorithms and generates and returns feedback based on the results.

[0792] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[0793] This invention is a system that collects data from professional players, generates an artificial intelligence model based on that data, and allows users to use it. It also recognizes the user's emotions in real time and adjusts the content of the match and feedback. Users can experience playing against professional players, receive accurate feedback after the match, and receive instruction that takes their own emotions into consideration.

[0794] Specific processing of the program

[0795] Professional data collection

[0796] The server collects data on professional players. For example, it retrieves the game history of a professional Go player from an online database or an official website and stores the data in a database.

[0797] Data Preprocessing

[0798] The server formats the collected data into a format that can be applied to machine learning. It converts the Go game data into coordinate format and extracts the features of each position. This ensures the consistency and quality of the data.

[0799] AI model generation and training

[0800] The server uses the preprocessed data to build a deep learning model. It designs a neural network using frameworks such as TensorFlow and Keras, and trains it to reproduce the playing style of a professional. The AI ​​model has the ability to predict the next move from the professional game record data.

[0801] Model optimization

[0802] The server evaluates and optimizes the performance of the generated model. It performs cross-validation and hyperparameter tuning to prevent overfitting and improve versatility. The model is then adjusted based on the evaluation results.

[0803] Deploying and serving the model

[0804] The server uploads the optimized model to a cloud server and provides it to the user's device via an API. The app installed on the user's device downloads the AI ​​model via this API and stores it locally, allowing users to play against the AI ​​model even offline.

[0805] Introducing the Emotion Engine

[0806] The user's device uses a camera and microphone to collect the user's facial expressions and voice in real time. The device processes the user's emotional data and uses a specific algorithm to recognize the user's emotional state. This emotional data is reflected in the AI's responses and feedback during the match.

[0807] Battles and Emotion Recognition

[0808] A user opens the app and begins a match against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. For example, if the user is feeling impatient or irritated, the AI ​​can display a message encouraging the user to pause.

[0809] Sending and analyzing match results

[0810] Once a user has finished a match, the results are sent to the server. The server analyzes the match data and identifies weaknesses and areas for improvement in the user's play. At the same time, feedback based on emotional data is also generated. This allows the user to receive guidance not only on technical but also psychological aspects.

[0811] Providing Feedback

[0812] The server sends the analysis results and feedback to the user's device, which displays the feedback to help the user review the game afterward. The feedback includes advice on mistakes, alternatives, and even advice based on the user's emotional state.

[0813] Specific examples

[0814] For example, in a Go match, a user opens a smartphone app and taps the start button. The device uses an AI model downloaded from the cloud to suggest the next move. During the match, the user's face is captured on camera, and their facial expression data is analyzed by the emotion engine. The engine determines in real time whether the user is concentrating on the match or feeling anxious or irritated, and suggests an appropriate response. After the match ends, the user sends the match log and emotional data to the server, and receives the analysis results along with technical and psychological advice in real time. In this way, users can learn professional techniques while also mastering how to control their own mental state.

[0815] The processing flow will be explained below.

[0816] Step 1:

[0817] The server collects data on professional players, specifically, obtaining their game history and strategy data from online databases and official websites, and storing the data in a database.

[0818] Step 2:

[0819] The server converts the collected data into a format that can be applied to machine learning. For example, it converts Go game record data into coordinate format and extracts the features of each position. This ensures the consistency and quality of the data.

[0820] Step 3:

[0821] The server uses the preprocessed data to build a deep learning model. It designs a neural network using frameworks such as TensorFlow and Keras, and trains it to reproduce the playing style of a professional. The AI ​​model has the ability to predict the next move from the professional game record data.

[0822] Step 4:

[0823] The server evaluates and optimizes the performance of the generated model. It performs cross-validation and hyperparameter tuning to prevent overfitting and improve versatility. The model is then adjusted based on the evaluation results.

[0824] Step 5:

[0825] The server uploads the optimized model to the cloud server, and an API is built to make it easy for users to access, and the model is provided through this API.

[0826] Step 6:

[0827] The user's device downloads the latest AI model from the cloud server and stores it locally, allowing the user to play against the AI ​​model even offline.

[0828] Step 7:

[0829] The user device uses a camera and microphone to collect the user's facial expressions and voice in real time, and the collected data is sent to an emotion engine to analyze the user's emotional state.

[0830] Step 8:

[0831] The user opens the app and begins a match against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. For example, if the user becomes impatient, the AI ​​displays a message suggesting a pause.

[0832] Step 9:

[0833] The user's device records the user's emotional data and battle data during the match, allowing the user to track emotional changes along with the match history.

[0834] Step 10:

[0835] After a user finishes a match, the match results and emotional data are sent to the server, which analyzes the match data and identifies weaknesses and areas for improvement.

[0836] Step 11:

[0837] The server sends the analysis results and feedback to the user's device. The feedback includes error detection, alternative solutions, and advice based on the user's emotional state. This provides not only technical guidance but also emotional and mental support.

[0838] In this way, users can learn professional techniques while also mastering how to control their own mental state, thereby efficiently improving their skills.

[0839] Example 2

[0840] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0841] Conventional technology not only provides users with the experience of playing against professional players, but also fails to provide feedback that reflects their emotional state during the match in real time. As a result, while users can learn technical skills, it is difficult for them to improve their mental control or emotional state. To solve this problem, a system is needed that collects users' emotional data and adjusts the content of the match and feedback based on that data.

[0842] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for collecting data of professional players, a means for training an AI model to be generated based on the collected data of professional players, and a means for optimizing the generated AI model. This makes it possible to faithfully reproduce the playing style of a professional and provide it to the user. Furthermore, the user terminal includes a means for collecting and analyzing user emotional data using a camera and microphone, a means for adjusting the content of the match and feedback based on the user emotional data in real time, and a means for receiving, analyzing, and providing feedback on the results of the user's match against the generated AI model. This allows the user to receive guidance on both technical and mental aspects.

[0843] "Professional Player Data" means data including the match history, playing style, and related statistical information of players who are recognized as professionals in the game.

[0844] An "artificial intelligence model" is a mathematical structure trained using machine learning techniques, and refers to a system that makes predictions and judgments based on input data.

[0845] A "user terminal" is a device, such as a personal computer, smartphone, or tablet, that is directly operated by a user to send and receive information.

[0846] "Emotion data" is data that indicates the psychological state of the user, obtained from facial expressions, voice, etc., and is quantified or classified using an emotion analysis engine.

[0847] "Feedback" refers to guidance and advice provided based on the results of a match or the user's performance, and includes advice on technical improvements and mental aspects.

[0848] "Collection means" refers to software and hardware configurations for obtaining data via the Internet.

[0849] A "training tool" is a process and tool for analyzing data and training a model using machine learning algorithms.

[0850] "Optimization means" refers to a method of adjusting and evaluating the generated artificial intelligence model to improve its performance, including hyperparameter adjustment and cross-validation.

[0851] "Means of providing" refers to a deployment method for making the optimized artificial intelligence model available to user devices, including cloud APIs and download links.

[0852] "Means of analysis" refers to the processes and tools used to analyze collected data statistically or through machine learning to extract meaningful information.

[0853] The present invention is a system that collects data from professional players and allows users to use an artificial intelligence model generated based on this data. This system also recognizes the user's emotions in real time and adjusts the match content and feedback, allowing users to experience playing against professional players. In addition, users receive accurate feedback after the match and can receive instruction that takes their own emotions into consideration.

[0854] First, the server collects data on professional players. For example, it retrieves the game history of a professional Go player from an online database or official website and stores that data in a database. Specific technologies used include web scraping using Python's BeautifulSoup and Scrapy, and storing the retrieved data in AWS RDS.

[0855] Next, the server converts the collected data into a format that can be applied to machine learning. For example, Go game data is converted into coordinate format, and the features of each position are extracted. At this stage, data processing libraries such as pandas and NumPy are used.

[0856] The server then uses the preprocessed data to build a deep learning model. Using deep learning frameworks such as TensorFlow and Keras, it designs a neural network and trains it to replicate the playing style of a professional. This gives the AI ​​model the ability to predict the next move from professional match data.

[0857] Cross-validation and hyperparameter tuning are required to evaluate and optimize the performance of the constructed model. The server performs cross-validation using scikit-learn and finds optimal hyperparameters using methods such as grid search and random search.

[0858] The optimized model is uploaded to a cloud server by the server and provided to user devices via an API. The server is built using Heroku or AWS Lambda, and a RESTful API is configured. User devices can download the AI ​​model via this API and save it locally. This allows users to play against the AI ​​model even offline.

[0859] The user's device uses a camera and microphone to collect the user's facial expressions and voice in real time. The facial expression data is analyzed using OpenCV or Microsoft Azure's Face API, and a specific algorithm is used to recognize the user's emotional state. This emotional data is reflected in the AI's responses and feedback during the match.

[0860] The user opens the app and begins a match against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. If the user becomes impatient or irritated, the device can display a message encouraging the user to pause.

[0861] After a match, the user sends the match results to the server. The server analyzes the match data and identifies the user's weaknesses and areas for improvement. At the same time, feedback based on emotional data is generated, allowing the user to receive guidance not only on technical but also psychological aspects.

[0862] This system not only allows users to experience playing against professional players, but also provides instruction on both technical and mental aspects, allowing users to learn professional techniques while also mastering their own mental control.

[0863] Prompt Sentence Examples

[0864] For example, "Please tell me the specific steps to develop a Go game application with emotion recognition capabilities using Python. Please explain how to use Azure's Face API to recognize the user's emotions and adjust the AI's response based on that data."

[0865] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0866] Step 1: Professional data collection

[0867] The server collects data on professional players. Specifically, it uses Python's BeautifulSoup and Scrapy to scrape the game history of professional players from online databases and official websites. Website URLs are used as input, and game data in JSON format is output. This data is then uploaded to AWS's RDS for efficient management and storage.

[0868] Step 2: Preprocessing the data

[0869] The data collected by the server is formatted into a format that can be applied to machine learning. Specifically, the Python pandas library is used to convert the JSON data into a data frame. The input is the JSON-formatted game data obtained in step 1, and the output is a structured data frame with missing values ​​imputed. Next, the Go game record data is converted into coordinate format, and features of each position are extracted.

[0870] Step 3: Generate and train an AI model

[0871] The server uses the preprocessed data to build a deep learning model. It designs a neural network using frameworks such as TensorFlow or Keras. The input is the data frame shaped in step 2, and the output is a trained AI model. Specifically, it supplies the training data to the model and optimizes the model parameters using backpropagation.

[0872] Step 4: Optimize the model

[0873] The server evaluates and optimizes the generated model. Specifically, it uses scikit-learn to perform cross-validation and adjusts hyperparameters using grid search and random search. The trained AI model obtained in step 3 is used as input, and an optimized AI model is obtained as output. This prevents the model from overfitting and increases its versatility.

[0874] Step 5: Deploy and serve the model

[0875] The server uploads the optimized model to a cloud server and provides it to the user device via an API. Specifically, the model is deployed to Heroku or AWS Lambda, and a RESTful API is built using Flask or FastAPI. The optimized model obtained in step 4 is used as input, and the model provided to the user device via the API is obtained as output. The user device downloads the AI ​​model via this API and stores it locally.

[0876] Step 6: Implementing the Emotion Engine

[0877] The user's device uses a camera and microphone to collect the user's facial expressions and voice in real time. Specifically, the facial expression data is analyzed using OpenCV and Microsoft Azure's Face API. The input is real-time media data obtained from the camera and microphone, and the output is the user's emotional data. This emotional data is reflected in the AI's responses and feedback during the match.

[0878] Step 7: Match and Emotion Recognition

[0879] The user opens the app and begins playing against a professional copy AI. Specifically, the device uses an AI model downloaded from the cloud to predict the next move and presents it to the user. The input is the next move information entered by the user into the app and real-time emotional data, and the output is the optimal move and a response message based on the user's emotion. If the user becomes impatient or irritated, the device displays a message encouraging them to pause.

[0880] Step 8: Submitting and analyzing match results

[0881] When a user finishes a match, the match results are sent to the server. Specifically, the device converts the match results and emotional data into JSON format and sends it to the server via an API. The input is the user's match results and emotional data, and the output is the analysis results. Based on this, the server analyzes the user's weaknesses and areas for improvement during play and generates technical and mental feedback.

[0882] Step 9: Provide feedback

[0883] The server sends the analysis results and feedback to the user device. Specifically, the analysis results are generated in JSON format and sent to the user device via API. The analysis results from step 8 are used as input, and feedback is obtained as output, which is displayed on the user device. The user device displays the received feedback in a format that is easy for the user to understand, and provides technical advice and advice based on the user's emotional state.

[0884] (Application example 2)

[0885] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the headset type terminal 314 will be referred to as a "terminal."

[0886] Conventional AI models based on data from professional players can provide users with technical feedback, but they have the problem of being unable to provide feedback or adjust game content that takes into account the user's emotional state. This can potentially reduce users' learning efficiency and satisfaction. The present invention aims to provide a system that recognizes a user's emotional state in real time and provides appropriate feedback and adjusts game content based on that emotion.

[0887] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for collecting data of professional players, a means for training a generative AI model based on the collected data of professional players, and a means for optimizing the generative AI model. This makes it possible to experience a realistic battle based on the skills of professional players.

[0888] The server further includes means for providing the optimized generative AI model to the user terminal, means for receiving and analyzing the results of the user's match against the generative AI model and providing feedback, and means for recognizing the user's emotional state in real time and adjusting the content of the match and feedback based on the user's emotions. This allows the user to receive feedback that takes into account their own emotional state, improving learning efficiency and satisfaction.

[0889] "Professional player data" refers to information that indicates the match records and playing styles of professional players in sports such as Go, Shogi, Chess, or card games.

[0890] A "generative AI model" is an AI model built using machine learning based on data from professional players, and has the ability to predict the next move during a match.

[0891] "Optimization" refers to the process of evaluating the performance of a generated artificial intelligence model and using machine learning techniques to improve the model's accuracy.

[0892] "User terminal" refers to any device used by a user, such as a smartphone, smart glasses, head-mounted display, or robot.

[0893] "Emotional state" refers to the psychological state that a user exhibits during a match, such as anxiety, irritation, or concentration.

[0894] "Feedback" is information on improvements and advice for the user's playing style that is provided based on the results of matches between the AI ​​model and the user.

[0895] "Deep learning technology" is a type of machine learning technology used to train artificial intelligence models, and is a method of learning complex patterns and features using multi-layer neural networks.

[0896] An "algorithm" refers to a procedure or computational method for solving a specific problem, and in this case refers to a series of processing steps used to analyze a user's emotional state.

[0897] A "cloud server" refers to a server that provides large-scale computer resources that can be remotely accessed via the Internet.

[0898] This invention is a system that collects data from professional players and generates a generative AI model based on that data. In this system, the user terminal recognizes the user's emotional state in real time and provides appropriate feedback corresponding to that emotion. The specific configuration and operating procedure are shown below.

[0899] System Configuration

[0900] 1. Data collection methods:

[0901] The server collects data on professional players of games such as Go, Shogi, Chess, and card games, including game records and playing styles. The data is obtained from online databases or official websites and stored in a database.

[0902] 2. Means of generating artificial intelligence models:

[0903] The server trains an artificial intelligence model based on the collected data. Deep learning technology is used to design a neural network and train it to reproduce the playing style of a professional. Frameworks such as TensorFlow and Keras are used.

[0904] 3. Optimization methods for artificial intelligence models:

[0905] The server evaluates the performance of the generated AI model and performs optimization, improving the model's accuracy through cross-validation and hyperparameter tuning.

[0906] 4. Means of provision to user terminal:

[0907] The server uploads the optimized AI model to a cloud server and provides it to the user's device via an API, allowing the user to play against the AI ​​model even offline.

[0908] 5. Emotion recognition means:

[0909] The user device uses a camera and microphone to collect the user's facial expressions and voice in real time, and uses algorithms to recognize the user's emotional state. This data is reflected in AI responses and feedback during the match.

[0910] 6. Competition and feedback channels:

[0911] The user opens the app and begins a match against a professional copy AI. During the match, the user's device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. After the match ends, the match results and emotion data are sent to the server, which provides analysis results and feedback.

[0912] Hardware and software used

[0913] Hardware: Smartphones, smart glasses, head-mounted displays, robots, webcams

[0914] Software: TensorFlow, Keras, facial_expression_recognition library

[0915] Examples of concrete examples and prompts

[0916] For example, in a Go match, a user opens a smartphone app and taps the "Start Match" button. During the match, the camera captures the user's face and analyzes their facial expressions. If the user shows signs of anxiety or irritation, the app displays a message saying, "Would you like to take a break?" After the match ends, the results and emotional data are sent to a server, and technical and psychological feedback is received.

[0917] Example prompt sentence:

[0918] "If the user's emotions show signs of anxiety or irritation during a Go match, implement a system that displays an appropriate message to encourage the user to take a break."

[0919] With the above system, users can learn the techniques of professional players while also learning efficiently while controlling their own emotions.

[0920] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0921] Step 1:

[0922] The server collects data from professional players. Specifically, it retrieves match records for Go, Shogi, chess, and card games from online databases and official websites, and stores the data in an internal database. The input is the match record data of professional players, and the output is formatted match record data. To ensure data consistency and quality, data processing such as format conversion and deletion of duplicate data is performed.

[0923] Step 2:

[0924] The server trains an AI model based on the collected data. Specifically, the collected match records are input into a neural network using deep learning frameworks such as TensorFlow and Keras, and training is performed to reproduce the playing style of professional players. The input is formatted match record data, and the output is a trained AI model. Here, data calculations such as model construction, preprocessing of learning data, and training are performed.

[0925] Step 3:

[0926] The server evaluates and optimizes the performance of the generated AI model. Specifically, it evaluates the predictive accuracy of the trained model and improves the model's accuracy using cross-validation and hyperparameter tuning. The input is the trained AI model and evaluation data, and the output is an optimized AI model. Parameters are adjusted to prevent overfitting and increase versatility.

[0927] Step 4:

[0928] The server uploads the optimized AI model to the cloud server and provides it to the user's device via API. The input is the optimized AI model, and the output is the model distribution to the user's device. The model is downloaded to the user's device so that the user can play offline.

[0929] Step 5:

[0930] The user device uses a camera and microphone to collect the user's facial expressions and voice in real time and recognize their emotional state. Specifically, the images captured by the camera are input into an emotion recognition algorithm to classify the user's emotional state. The input is the user's facial images and voice data captured in real time, and the output is the user's emotional state data. Here, facial expression analysis and voice analysis are performed.

[0931] Step 6:

[0932] A user starts a match against an artificial intelligence model, and the user's device predicts the AI's next move during the match. The input is the user's movements during the match (such as the board state), and the output is the AI's next move. During the match, the emotion engine monitors the user's emotional state in real time and adjusts the AI's response and feedback content as necessary. This is where real-time predictions and emotional response adjustments take place.

[0933] Step 7:

[0934] After a user finishes a match, the match results and the user's emotional state data are sent to the server. The server analyzes this data and generates feedback including the user's weaknesses and areas for improvement during play. The input is the match results and emotional state data, and the output is the analysis results and feedback. Data analysis and feedback generation are performed.

[0935] Step 8:

[0936] The server sends the generated feedback to the user's device, which then displays it to the user. The input is the analysis results and feedback, and the output is the feedback display to the user. The feedback includes not only technical advice but also mental support based on the user's emotional state. This allows the user to improve both their playing skills and emotional management.

[0937] The specific processing unit 290 transmits the result of the specific processing to the headset type terminal 314. In the headset type terminal 314, the control unit 46A causes the speaker 240 and the display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating a user input regarding the result of the specific processing. The control unit 46A transmits audio data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0938] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0939] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the headset type terminal 314.

[0940] [Fourth embodiment]

[0941] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[0942] 7, a data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0943] The data processing device 12 includes a computer 22, a database 24, and a communication I / F 26. The computer 22 is an example of a "computer" according to the technology of the present disclosure. The computer 22 includes a processor 28, a RAM 30, and a storage 32. The processor 28, the RAM 30, and the storage 32 are connected to a bus 34. The database 24 and the communication I / F 26 are also connected to the bus 34. The communication I / F 26 is connected to a network 54. Examples of the network 54 include a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0944] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication I / F 44, and a control target 443. The computer 36 includes a processor 46, a RAM 48, and a storage 50. The processor 46, the RAM 48, and the storage 50 are connected to a bus 52. The microphone 238, the speaker 240, the camera 42, and the control target 443 are also connected to the bus 52.

[0945] The microphone 238 receives instructions and the like from the user 20 by receiving voice uttered by the user 20. The microphone 238 captures the voice uttered by the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio in accordance with instructions from the processor 46.

[0946] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an imaging element such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the surroundings of user 20 (for example, an imaging range defined by an angle of view equivalent to the field of vision of a typical healthy person).

[0947] The communication I / F 44 is connected to a network 54. The communication I / Fs 44 and 26 are responsible for the exchange of various information between the processor 46 and the processor 28 via the network 54. The exchange of various information between the processor 46 and the processor 28 using the communication I / Fs 44 and 26 is carried out in a secure state.

[0948] The control object 443 includes a display device, LEDs in the eyes, and motors for driving the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the emotions of the robot 414 can be expressed by controlling these motors. In addition, the facial expressions of the robot 414 can also be expressed by controlling the light emission state of the LEDs in the eyes of the robot 414.

[0949] Fig. 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Fig. 8, in the data processing device 12, a specific process is performed by the processor 28. A specific process program 56 is stored in the storage 32.

[0950] The specific processing program 56 is an example of a "program" according to the technology of the present disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0951] The storage 32 stores a data generation model 58 and an emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0952] In the robot 414, the processor 46 performs the reception output process. A reception output program 60 is stored in the storage 50. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output process is realized by the processor 46 operating as the control unit 46A in accordance with the reception output program 60 executed on the RAM 48.

[0953] Next, a description will be given of the specific processing performed by the specific processing unit 290 of the data processing device 12. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0954] The present invention is a system that allows users to use an artificial intelligence model created based on collected data from professional players. With this system, users can experience playing against professional players and receive accurate feedback after the match.

[0955] Specific processing of the program

[0956] Professional data collection

[0957] The server collects data on professional players. For example, it retrieves the game history of a professional Go player from an online database or an official website. The data is recorded in coordinate format and used for subsequent processing.

[0958] Data Preprocessing

[0959] The server converts the collected data into a format that can be applied to machine learning. This process extracts features that represent the moves and strategic patterns at each stage. For example, Go game records can be converted into coordinate data, and the overall game progress and the strength of the moves can be quantified.

[0960] AI model generation and training

[0961] The server uses the formatted data to train a generative AI model. Deep learning techniques are used to build a model that reproduces the playing style of a professional. For example, a neural network is built using Keras or TensorFlow, and a model is trained to predict the next move from professional game records.

[0962] Model optimization

[0963] The server uses cross-validation and hyperparameter tuning to optimize the performance of the generated AI model, thereby avoiding overfitting and providing a model that can maintain high accuracy in actual games.

[0964] Deploying and serving the model

[0965] The server uploads the optimized AI model to the cloud server and provides it to the user's device via API. The app installed on the user's device downloads the AI ​​model via this API and stores it locally, allowing users to play against the AI ​​anytime, anywhere.

[0966] Match and Feedback

[0967] The user uses the app to play against the AI ​​model. The device predicts the AI's next move and displays its response. After the match ends, the user sends the match data to the server. The server analyzes the match data, generates feedback on the user's weaknesses and areas for improvement during play, and displays the feedback on the user's device. This feedback helps the user improve their skills.

[0968] Specific examples

[0969] For example, in a Go match, a user opens the smartphone app and taps the start button. The device uses an AI model downloaded from the cloud to suggest the next move. The user continues to make moves in response to the move, progressing the match. After the match ends, the user sends the match log to the server and receives analysis results and feedback in real time. This feedback includes where the wrong move was made, the reasons for the move, and alternatives. By providing an experience that mimics playing against a professional player and specific advice for improving their skills, users can hone their skills efficiently.

[0970] The processing flow will be explained below.

[0971] Step 1:

[0972] The server collects data on professional players, specifically, obtaining their game history and strategy data from online databases and official websites, and storing the data in a database.

[0973] Step 2:

[0974] The server converts the collected data into a format that can be applied to machine learning. For example, Go game record data is converted into coordinate format and the features of each position are extracted. This process maintains the consistency and quality of the data.

[0975] Step 3:

[0976] The server uses the preprocessed data to build a deep learning model, specifically, designing a neural network using frameworks such as TensorFlow and Keras, and training it to reproduce the playing style of a professional.

[0977] Step 4:

[0978] The server evaluates and optimizes the performance of the generated model. It performs cross-validation and hyperparameter tuning to prevent overfitting and improve versatility. The model is then modified based on the evaluation results.

[0979] Step 5:

[0980] The server uploads the optimized model to the cloud server, and an API is built to make it easy for users to access, and the model is provided through this API.

[0981] Step 6:

[0982] The user's device downloads the latest AI model from the cloud server and stores it locally, allowing the user to play against the AI ​​model even offline.

[0983] Step 7:

[0984] The user opens the app and begins playing against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The user then inputs their own next move and continues the match.

[0985] Step 8:

[0986] When a user finishes a match, the results are sent to the server, which analyzes the match data and identifies weaknesses and areas for improvement.

[0987] Step 9:

[0988] The server sends the analysis results and feedback to the user's device. The user's device displays the feedback, helping the user to easily review their game after it has finished. The feedback includes suggestions for mistakes and alternative methods, helping the user improve their skills.

[0989] Example 1

[0990] Next, a description will be given of Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[0991] Conventional AI-based game systems have struggled to provide users with the same high-level experience as if they were playing against a professional player. Furthermore, they lacked mechanisms for providing detailed feedback to help users improve their skills. Furthermore, AI model optimization and model provisioning were often inefficient.

[0992] The specific processing by the specific processing unit 290 of the data processing device 12 in the first embodiment is realized by the following means.

[0993] In this invention, the server includes means for collecting data of professional players, means for preprocessing data based on the collected data of professional players, means for training a generative AI model using the preprocessed data, means for optimizing the generated AI model using cross-validation and hyperparameter tuning, means for deploying the optimized generative AI model to a cloud environment and providing it to a user terminal, and means for a user to play against the generative AI model, receive and analyze the results of the game, and provide feedback. This allows users to enjoy an advanced experience similar to playing against professional players, and enables them to improve their skills through detailed feedback.

[0994] "Professional Player Data" means information about an expert or skilled player, including the results of previous matches or games.

[0995] "Data preprocessing" is the process of converting raw data into a form that can be applied to machine learning, including shaping the data, filling in missing values, and treating outliers.

[0996] A "generative artificial intelligence model" is an algorithm or model built using machine learning or deep learning techniques and trained to perform a specific task or prediction.

[0997] "Cross-validation" is a method of evaluating the performance of a machine learning model by dividing the data into multiple parts and checking the model's generalization ability.

[0998] "Hyperparameter tuning" is the process of adjusting parameters related to the model's settings and structure in order to optimize the performance of a machine learning model.

[0999] "Deploying to a cloud environment" means placing optimized models and applications on remote servers available via the Internet.

[1000] "User terminal" refers to a device that is directly operated by a user, and includes smartphones, PCs, tablets, etc.

[1001] "Feedback" refers to advice and comments provided to improve a user's skills based on an analysis of the user's match results.

[1002] The present invention is a system that allows users to use artificial intelligence (AI) models generated based on collected data from professional players. With this system, users can experience playing against professional players and receive accurate feedback after the match.

[1003] System Configuration

[1004] Data collection

[1005] The server collects data on professional players. Specifically, it retrieves the game history of professional Go players from online databases and official websites. This data is recorded in coordinate format and includes the coordinate information of each move, the time of each move, and the result of the game.

[1006] Data Preprocessing

[1007] The server preprocesses the collected data. This process converts raw data into a format that can be applied to machine learning. For example, Go game data is converted into coordinate data, and the overall game progress and the efficiency and impact of each move are quantified.

[1008] AI model generation and learning

[1009] The server uses the preprocessed data to generate and train an AI model. Deep learning techniques are used to build a neural network and create a model that reproduces the playing style of a professional. During this process, libraries such as Keras and TensorFlow are used to define the model, and the model's performance is evaluated using training and validation data.

[1010] Model Optimization

[1011] The server optimizes the model through cross-validation and hyperparameter tuning. This prevents overfitting and provides a model that can maintain high accuracy in actual games. Specifically, it uses methods such as GridSearchCV and RandomSearchCV to find the optimal hyperparameters.

[1012] Deploying and serving the model

[1013] The server deploys the optimized AI model to a cloud environment and provides it to the user's device via an API. Applications installed on the user's device download the model via the API and store it locally. This process allows users to use the latest AI model.

[1014] Match and Feedback

[1015] The user uses the app to play against an AI model. During the match, the device calculates and displays the AI's next move. After the match ends, the user sends the match log to the server. The server analyzes the match data and generates feedback on the user's weaknesses and areas for improvement during play, which is displayed on the user's device. This feedback is provided in detail to help the user improve their skills.

[1016] Specific examples

[1017] For example, in a Go match, the user opens the smartphone app and taps the start button. The device uses an AI model downloaded from the cloud to display the next move. The user plays moves in response to the move, progressing the match. After the match ends, the user sends the match log to the server and receives analysis results and feedback in real time. This feedback includes where the wrong move was made, the reasons for the move, and alternatives. By providing an experience that feels like playing against a professional player and specific advice for improving their skills, users can efficiently hone their skills.

[1018] Prompt Sentence Examples

[1019] "Build an AI model that predicts the next move from Go game data. The data used is the game history of professional Go players, and create a model using deep learning. Use Keras or TensorFlow to build a neural network that replicates the playing style of a professional."

[1020] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1021] Step 1: Data collection

[1022] The server collects data from professional players. The input data is collected from online databases and official website APIs. Specifically, the server calls the API to retrieve the game history of the professional players. This data includes the coordinate information of each move, the time of each move, the result of the game, etc. The output data is stored in a local database.

[1023] Step 2: Data Preprocessing

[1024] The data collected by the server is preprocessed. The input is the raw data collected in step 1. Specific operations include shaping the data, filling in missing values, and filtering outliers. For example, Go game record data is converted into coordinate format, and the overall game progress and the efficiency and impact of each move are quantified. The output is preprocessed data that can be applied to machine learning.

[1025] Step 3: AI model generation and training

[1026] The server uses the preprocessed data to generate and train an AI model. The input is the preprocessed data obtained in step 2. Specifically, it uses deep learning techniques to build a neural network and defines the model using Keras or TensorFlow. It evaluates the model's performance using training and validation data and learns optimal weights. The output is an initial AI model.

[1027] Step 4: Model optimization

[1028] The server optimizes the generated AI model. The input is the initial AI model generated in step 3. Specifically, it performs cross-validation and hyperparameter tuning to find optimal hyperparameters using techniques such as GridSearchCV and RandomSearchCV. The output is an optimized AI model.

[1029] Step 5: Deploy and serve the model

[1030] The server deploys the optimized AI model to the cloud environment. The input is the AI ​​model optimized in step 4. Specifically, the model is uploaded to the cloud server and published as an API endpoint. The user device then downloads the AI ​​model via this API and saves it locally. The output is the AI ​​model on the user device.

[1031] Step 6: Match and feedback

[1032] The user uses the app to play against an AI model. The input is the downloaded AI model. Specifically, the user taps the start button on the app, and the device calculates and displays the AI's next move. After the match ends, the user sends the match log to the server. The server analyzes the match data and generates feedback on the user's weaknesses and areas for improvement during play, which is displayed on the user's device. The output is detailed feedback and advice on how to improve the user's skills.

[1033] (Application example 1)

[1034] Next, a description will be given of Application Example 1. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1035] In traditional manufacturing, it has been difficult for new employees and employees in training to efficiently learn the skills of professional workers. In particular, there is a need to pass on the skilled techniques and know-how of professional workers and improve production efficiency while maintaining consistent quality. However, current training methods require time and effort and pose a risk of affecting the operating rate of the production line. Therefore, a new system is needed to efficiently and reliably transfer the skills and know-how of professional workers to new employees.

[1036] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 1 is realized by the following means.

[1037] In this invention, the server includes means for collecting data of professional workers, means for training a generative AI model based on the collected data of professional workers, means for optimizing the generative AI model, means for providing the optimized generative AI model to a user terminal, and means for receiving and analyzing the results of a user's competition or collaboration with the generative AI model and providing feedback. This enables new employees and employees in training to learn skills effectively and efficiently by competing or collaborating with professional workers and receiving feedback based on real-life experience.

[1038] A "professional worker" refers to an individual who is engaged in manufacturing work with advanced specialized knowledge and skills and who has a certain level of proficiency.

[1039] "Data" refers to information such as operation records, measurement values, and operation logs generated when professional workers perform their work.

[1040] A "generative artificial intelligence model" refers to an artificial intelligence system that is generated by applying machine learning algorithms based on collected data.

[1041] "Learning" refers to the process by which an AI model uses collected data to understand behaviors and patterns and improve its capabilities.

[1042] "Optimization" refers to the process of making adjustments to an artificial intelligence model to maximize its performance and minimize error.

[1043] "User terminal" refers to a device for using an artificial intelligence model, and specifically includes hardware such as smartphones, tablets, PCs, and robots.

[1044] "Competition or collaboration" refers to the process in which a user actually works or collaborates with an artificial intelligence model.

[1045] "Receiving" refers to a user obtaining data generated as a result of a match or collaboration.

[1046] "Analysis" refers to the detailed examination of the results of a competition or collaboration and the processing of data to understand its significance and patterns.

[1047] "Feedback" refers to information that indicates evaluations and areas for improvement based on a user's actions and results.

[1048] The system of the present invention consists of the following main steps: First, collect data from professional workers, generate and optimize a generative AI model based on the data, and then provide the model to the user terminal, analyze the results of the competitive or collaborative play, and provide feedback.

[1049] 1. Professional data collection

[1050] The server automatically collects various data, such as motion records, measurements, and operation logs, from professional workers as they perform their tasks. This data is collected using high-performance sensors, cameras, and other measuring devices. This data is then aggregated in one place and used as a basic dataset for machine learning.

[1051] 2. Data preprocessing and feature extraction

[1052] The server preprocesses the collected data and formats it into a format suitable for machine learning. Specifically, it normalizes the data, fills in missing values, extracts features, etc. This process prepares the data in a way that maximizes the performance of the AI ​​model.

[1053] 3. Generating and training AI models

[1054] The server uses the preprocessed data to train a generative artificial intelligence model using deep learning techniques (specifically TensorFlow and Keras), which imitates the actions and patterns of professional workers and develops highly accurate prediction capabilities.

[1055] 4. Optimize and deploy the model

[1056] The server optimizes the generated AI model using cross-validation and hyperparameter tuning. The optimized model is then uploaded to a cloud server and provided to user devices via API. User devices can be a variety of devices, including smartphones, tablets, PCs, and robots.

[1057] 5. Competition and Feedback

[1058] Users compete or collaborate with the AI ​​model through the application. The results of the competition or collaboration are sent to the server in real time. The server analyzes the results and generates detailed feedback based on the user's actions and performance. This feedback is displayed on the user's device, allowing the user to effectively improve their skills and work efficiency.

[1059] Examples:

[1060] For example, when new employees undergo training to operate a robot arm on a manufacturing site, the system uses an AI model generated by a server. This system allows the AI ​​model, which has the same skills as a professional worker, to support the new employee's training. After the training session is over, the system provides analysis results and feedback, specifically indicating which movements the new employee needs to improve. This allows employees to smoothly improve their skills.

[1061] Example prompt sentence:

[1062] "I want to design a training system for new employees. How can I build a system that collects data from professional workers, generates an AI model based on that data, and provides real-time feedback to employees?"

[1063] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1064] Step 1:

[1065] The server collects data from professional workers. Specifically, it automatically acquires data such as operation records, measurements, and operation logs from sensors, cameras, and other measuring devices used by professional workers in the manufacturing process and stores the data in a database.

[1066] Input: Real-time data from sensors and cameras.

[1067] Output: Data of professional workers accumulated in a database.

[1068] Specific operation: Collecting data from various sensors and storing it in a database in real time.

[1069] Step 2:

[1070] The server preprocesses the collected data and formats it into a format suitable for machine learning. Specifically, it normalizes the data, fills in missing values, and extracts necessary features.

[1071] Input: Raw data in the database.

[1072] Output: A dataset formatted for machine learning.

[1073] Specific operations: Normalize the data, impute missing values, and select and extract the features to use.

[1074] Step 3:

[1075] The server uses the preprocessed data to train a generative artificial intelligence model. Specifically, it builds a neural network model using a deep learning framework (e.g., TensorFlow, Keras) and trains it using the collected data.

[1076] Input: The preprocessed dataset.

[1077] Output: A trained artificial intelligence model.

[1078] Specific operations: Set up the neural network architecture and train the model while optimizing the learning parameters.

[1079] Step 4:

[1080] The server optimizes the generated artificial intelligence model, specifically by performing cross-validation and hyperparameter tuning to improve the accuracy of the model.

[1081] Input: A trained artificial intelligence model.

[1082] Output: An optimized artificial intelligence model.

[1083] What it does: It performs cross-validation to find the optimal set of hyperparameters and tune them to maximize the model's accuracy.

[1084] Step 5:

[1085] The server uploads the optimized artificial intelligence model to a cloud server and provides it to the user's device via an API.

[1086] Input: An optimized artificial intelligence model.

[1087] Output: An artificial intelligence model that is downloaded to the user's device.

[1088] Specific operation: Deploying a model to a cloud server and downloading the model via an API call from the user's device.

[1089] Step 6:

[1090] Users use their devices to compete or collaborate with artificial intelligence models. Specifically, users operate the application and work in conjunction with the AI ​​model to carry out manufacturing tasks.

[1091] Input: User operation instructions.

[1092] Output: Work assistance data generated by the AI ​​model.

[1093] Specific behavior: The user performs an operation through the application, and the AI ​​model responds with assistance and instructions.

[1094] Step 7:

[1095] The server analyzes the results of the match or collaboration and provides feedback. Specifically, it analyzes the user's operation log in detail, generates advice on areas for improvement and technical enhancement, and provides it to the user's device.

[1096] Input: User operation log.

[1097] Output: Feedback content.

[1098] Specific operation: Analyzes operation logs using analytical algorithms and generates and returns feedback based on the results.

[1099] Furthermore, an emotion engine that estimates the user's emotion may be further combined. That is, the identification processing unit 290 may estimate the user's emotion using the emotion identification model 59, and perform identification processing using the user's emotion.

[1100] This invention is a system that collects data from professional players, generates an artificial intelligence model based on that data, and allows users to use it. It also recognizes the user's emotions in real time and adjusts the content of the match and feedback. Users can experience playing against professional players, receive accurate feedback after the match, and receive instruction that takes their own emotions into consideration.

[1101] Specific processing of the program

[1102] Professional data collection

[1103] The server collects data on professional players. For example, it retrieves the game history of a professional Go player from an online database or an official website and stores the data in a database.

[1104] Data Preprocessing

[1105] The server formats the collected data into a format that can be applied to machine learning. It converts the Go game data into coordinate format and extracts the features of each position. This ensures the consistency and quality of the data.

[1106] AI model generation and training

[1107] The server uses the preprocessed data to build a deep learning model. It designs a neural network using frameworks such as TensorFlow and Keras, and trains it to reproduce the playing style of a professional. The AI ​​model has the ability to predict the next move from the professional game record data.

[1108] Model optimization

[1109] The server evaluates and optimizes the performance of the generated model. It performs cross-validation and hyperparameter tuning to prevent overfitting and improve versatility. The model is then adjusted based on the evaluation results.

[1110] Deploying and serving the model

[1111] The server uploads the optimized model to a cloud server and provides it to the user's device via an API. The app installed on the user's device downloads the AI ​​model via this API and stores it locally, allowing users to play against the AI ​​model even offline.

[1112] Introducing the Emotion Engine

[1113] The user's device uses a camera and microphone to collect the user's facial expressions and voice in real time. The device processes the user's emotional data and uses a specific algorithm to recognize the user's emotional state. This emotional data is reflected in the AI's responses and feedback during the match.

[1114] Battles and Emotion Recognition

[1115] A user opens the app and begins a match against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. For example, if the user is feeling impatient or irritated, the AI ​​can display a message encouraging the user to pause.

[1116] Sending and analyzing match results

[1117] Once a user has finished a match, the results are sent to the server. The server analyzes the match data and identifies weaknesses and areas for improvement in the user's play. At the same time, feedback based on emotional data is also generated. This allows the user to receive guidance not only on technical but also psychological aspects.

[1118] Providing Feedback

[1119] The server sends the analysis results and feedback to the user's device, which displays the feedback to help the user review the game afterward. The feedback includes advice on mistakes, alternatives, and even advice based on the user's emotional state.

[1120] Specific examples

[1121] For example, in a Go match, a user opens a smartphone app and taps the start button. The device uses an AI model downloaded from the cloud to suggest the next move. During the match, the user's face is captured on camera, and their facial expression data is analyzed by the emotion engine. The engine determines in real time whether the user is concentrating on the match or feeling anxious or irritated, and suggests an appropriate response. After the match ends, the user sends the match log and emotional data to the server, and receives the analysis results along with technical and psychological advice in real time. In this way, users can learn professional techniques while also mastering how to control their own mental state.

[1122] The processing flow will be explained below.

[1123] Step 1:

[1124] The server collects data on professional players, specifically, obtaining their game history and strategy data from online databases and official websites, and storing the data in a database.

[1125] Step 2:

[1126] The server converts the collected data into a format that can be applied to machine learning. For example, it converts Go game record data into coordinate format and extracts the features of each position. This ensures the consistency and quality of the data.

[1127] Step 3:

[1128] The server uses the preprocessed data to build a deep learning model. It designs a neural network using frameworks such as TensorFlow and Keras, and trains it to reproduce the playing style of a professional. The AI ​​model has the ability to predict the next move from the professional game record data.

[1129] Step 4:

[1130] The server evaluates and optimizes the performance of the generated model. It performs cross-validation and hyperparameter tuning to prevent overfitting and improve versatility. The model is then adjusted based on the evaluation results.

[1131] Step 5:

[1132] The server uploads the optimized model to the cloud server, and an API is built to make it easy for users to access, and the model is provided through this API.

[1133] Step 6:

[1134] The user's device downloads the latest AI model from the cloud server and stores it locally, allowing the user to play against the AI ​​model even offline.

[1135] Step 7:

[1136] The user device uses a camera and microphone to collect the user's facial expressions and voice in real time, and the collected data is sent to an emotion engine to analyze the user's emotional state.

[1137] Step 8:

[1138] The user opens the app and begins a match against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. For example, if the user becomes impatient, the AI ​​displays a message suggesting a pause.

[1139] Step 9:

[1140] The user's device records the user's emotional data and battle data during the match, allowing the user to track emotional changes along with the match history.

[1141] Step 10:

[1142] After a user finishes a match, the match results and emotional data are sent to the server, which analyzes the match data and identifies weaknesses and areas for improvement.

[1143] Step 11:

[1144] The server sends the analysis results and feedback to the user's device. The feedback includes error detection, alternative solutions, and advice based on the user's emotional state. This provides not only technical guidance but also emotional and mental support.

[1145] In this way, users can learn professional techniques while also mastering how to control their own mental state, thereby efficiently improving their skills.

[1146] Example 2

[1147] Next, a description will be given of Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1148] Conventional technology not only provides users with the experience of playing against professional players, but also fails to provide feedback that reflects their emotional state during the match in real time. As a result, while users can learn technical skills, it is difficult for them to improve their mental control or emotional state. To solve this problem, a system is needed that collects users' emotional data and adjusts the content of the match and feedback based on that data.

[1149] The identification process by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes a means for collecting data of professional players, a means for training an AI model to be generated based on the collected data of professional players, and a means for optimizing the generated AI model. This makes it possible to faithfully reproduce the playing style of a professional and provide it to the user. Furthermore, the user terminal includes a means for collecting and analyzing user emotional data using a camera and microphone, a means for adjusting the content of the match and feedback based on the user emotional data in real time, and a means for receiving, analyzing, and providing feedback on the results of the user's match against the generated AI model. This allows the user to receive guidance on both technical and mental aspects.

[1150] "Professional Player Data" means data including the match history, playing style, and related statistical information of players who are recognized as professionals in the game.

[1151] An "artificial intelligence model" is a mathematical structure trained using machine learning techniques, and refers to a system that makes predictions and judgments based on input data.

[1152] A "user terminal" is a device, such as a personal computer, smartphone, or tablet, that is directly operated by a user to send and receive information.

[1153] "Emotion data" is data that indicates the psychological state of the user, obtained from facial expressions, voice, etc., and is quantified or classified using an emotion analysis engine.

[1154] "Feedback" refers to guidance and advice provided based on the results of a match or the user's performance, and includes advice on technical improvements and mental aspects.

[1155] "Collection means" refers to software and hardware configurations for obtaining data via the Internet.

[1156] A "training tool" is a process and tool for analyzing data and training a model using machine learning algorithms.

[1157] "Optimization means" refers to a method of adjusting and evaluating the generated artificial intelligence model to improve its performance, including hyperparameter adjustment and cross-validation.

[1158] "Means of providing" refers to a deployment method for making the optimized artificial intelligence model available to user devices, including cloud APIs and download links.

[1159] "Means of analysis" refers to the processes and tools used to analyze collected data statistically or through machine learning to extract meaningful information.

[1160] The present invention is a system that collects data from professional players and allows users to use an artificial intelligence model generated based on this data. This system also recognizes the user's emotions in real time and adjusts the match content and feedback, allowing users to experience playing against professional players. In addition, users receive accurate feedback after the match and can receive instruction that takes their own emotions into consideration.

[1161] First, the server collects data on professional players. For example, it retrieves the game history of a professional Go player from an online database or official website and stores that data in a database. Specific technologies used include web scraping using Python's BeautifulSoup and Scrapy, and storing the retrieved data in AWS RDS.

[1162] Next, the server converts the collected data into a format that can be applied to machine learning. For example, Go game data is converted into coordinate format, and the features of each position are extracted. At this stage, data processing libraries such as pandas and NumPy are used.

[1163] The server then uses the preprocessed data to build a deep learning model. Using deep learning frameworks such as TensorFlow and Keras, it designs a neural network and trains it to replicate the playing style of a professional. This gives the AI ​​model the ability to predict the next move from professional match data.

[1164] Cross-validation and hyperparameter tuning are required to evaluate and optimize the performance of the constructed model. The server performs cross-validation using scikit-learn and finds optimal hyperparameters using methods such as grid search and random search.

[1165] The optimized model is uploaded to a cloud server by the server and provided to user devices via an API. The server is built using Heroku or AWS Lambda, and a RESTful API is configured. User devices can download the AI ​​model via this API and save it locally. This allows users to play against the AI ​​model even offline.

[1166] The user's device uses a camera and microphone to collect the user's facial expressions and voice in real time. The facial expression data is analyzed using OpenCV or Microsoft Azure's Face API, and a specific algorithm is used to recognize the user's emotional state. This emotional data is reflected in the AI's responses and feedback during the match.

[1167] The user opens the app and begins a match against a professional copy AI. During the match, the device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. If the user becomes impatient or irritated, the device can display a message encouraging the user to pause.

[1168] After a match, the user sends the match results to the server. The server analyzes the match data and identifies the user's weaknesses and areas for improvement. At the same time, feedback based on emotional data is generated, allowing the user to receive guidance not only on technical but also psychological aspects.

[1169] This system not only allows users to experience playing against professional players, but also provides instruction on both technical and mental aspects, allowing users to learn professional techniques while also mastering their own mental control.

[1170] Prompt Sentence Examples

[1171] For example, "Please tell me the specific steps to develop a Go game application with emotion recognition capabilities using Python. Please explain how to use Azure's Face API to recognize the user's emotions and adjust the AI's response based on that data."

[1172] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1173] Step 1: Professional data collection

[1174] The server collects data on professional players. Specifically, it uses Python's BeautifulSoup and Scrapy to scrape the game history of professional players from online databases and official websites. Website URLs are used as input, and game data in JSON format is output. This data is then uploaded to AWS's RDS for efficient management and storage.

[1175] Step 2: Preprocessing the data

[1176] The data collected by the server is formatted into a format that can be applied to machine learning. Specifically, the Python pandas library is used to convert the JSON data into a data frame. The input is the JSON-formatted game data obtained in step 1, and the output is a structured data frame with missing values ​​imputed. Next, the Go game record data is converted into coordinate format, and features of each position are extracted.

[1177] Step 3: Generate and train an AI model

[1178] The server uses the preprocessed data to build a deep learning model. It designs a neural network using frameworks such as TensorFlow or Keras. The input is the data frame shaped in step 2, and the output is a trained AI model. Specifically, it supplies the training data to the model and optimizes the model parameters using backpropagation.

[1179] Step 4: Optimize the model

[1180] The server evaluates and optimizes the generated model. Specifically, it uses scikit-learn to perform cross-validation and adjusts hyperparameters using grid search and random search. The trained AI model obtained in step 3 is used as input, and an optimized AI model is obtained as output. This prevents the model from overfitting and increases its versatility.

[1181] Step 5: Deploy and serve the model

[1182] The server uploads the optimized model to a cloud server and provides it to the user device via an API. Specifically, the model is deployed to Heroku or AWS Lambda, and a RESTful API is built using Flask or FastAPI. The optimized model obtained in step 4 is used as input, and the model provided to the user device via the API is obtained as output. The user device downloads the AI ​​model via this API and stores it locally.

[1183] Step 6: Implementing the Emotion Engine

[1184] The user's device uses a camera and microphone to collect the user's facial expressions and voice in real time. Specifically, the facial expression data is analyzed using OpenCV and Microsoft Azure's Face API. The input is real-time media data obtained from the camera and microphone, and the output is the user's emotional data. This emotional data is reflected in the AI's responses and feedback during the match.

[1185] Step 7: Match and Emotion Recognition

[1186] The user opens the app and begins playing against a professional copy AI. Specifically, the device uses an AI model downloaded from the cloud to predict the next move and presents it to the user. The input is the next move information entered by the user into the app and real-time emotional data, and the output is the optimal move and a response message based on the user's emotion. If the user becomes impatient or irritated, the device displays a message encouraging them to pause.

[1187] Step 8: Submitting and analyzing match results

[1188] When a user finishes a match, the match results are sent to the server. Specifically, the device converts the match results and emotional data into JSON format and sends it to the server via an API. The input is the user's match results and emotional data, and the output is the analysis results. Based on this, the server analyzes the user's weaknesses and areas for improvement during play and generates technical and mental feedback.

[1189] Step 9: Provide feedback

[1190] The server sends the analysis results and feedback to the user device. Specifically, the analysis results are generated in JSON format and sent to the user device via API. The analysis results from step 8 are used as input, and feedback is obtained as output, which is displayed on the user device. The user device displays the received feedback in a format that is easy for the user to understand, and provides technical advice and advice based on the user's emotional state.

[1191] (Application example 2)

[1192] Next, a description will be given of Application Example 2. In the following description, the data processing device 12 will be referred to as a "server" and the robot 414 will be referred to as a "terminal."

[1193] Conventional AI models based on data from professional players can provide users with technical feedback, but they have the problem of being unable to provide feedback or adjust game content that takes into account the user's emotional state. This can potentially reduce users' learning efficiency and satisfaction. The present invention aims to provide a system that recognizes a user's emotional state in real time and provides appropriate feedback and adjusts game content based on that emotion.

[1194] The specific processing by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means. In this invention, the server includes a means for collecting data of professional players, a means for training a generative AI model based on the collected data of professional players, and a means for optimizing the generative AI model. This makes it possible to experience a realistic battle based on the skills of professional players.

[1195] The server further includes means for providing the optimized generative AI model to the user terminal, means for receiving and analyzing the results of the user's match against the generative AI model and providing feedback, and means for recognizing the user's emotional state in real time and adjusting the content of the match and feedback based on the user's emotions. This allows the user to receive feedback that takes into account their own emotional state, improving learning efficiency and satisfaction.

[1196] "Professional player data" refers to information that indicates the match records and playing styles of professional players in sports such as Go, Shogi, Chess, or card games.

[1197] A "generative AI model" is an AI model built using machine learning based on data from professional players, and has the ability to predict the next move during a match.

[1198] "Optimization" refers to the process of evaluating the performance of a generated artificial intelligence model and using machine learning techniques to improve the model's accuracy.

[1199] "User terminal" refers to any device used by a user, such as a smartphone, smart glasses, head-mounted display, or robot.

[1200] "Emotional state" refers to the psychological state that a user exhibits during a match, such as anxiety, irritation, or concentration.

[1201] "Feedback" is information on improvements and advice for the user's playing style that is provided based on the results of matches between the AI ​​model and the user.

[1202] "Deep learning technology" is a type of machine learning technology used to train artificial intelligence models, and is a method of learning complex patterns and features using multi-layer neural networks.

[1203] An "algorithm" refers to a procedure or computational method for solving a specific problem, and in this case refers to a series of processing steps used to analyze a user's emotional state.

[1204] A "cloud server" refers to a server that provides large-scale computer resources that can be remotely accessed via the Internet.

[1205] This invention is a system that collects data from professional players and generates a generative AI model based on that data. In this system, the user terminal recognizes the user's emotional state in real time and provides appropriate feedback corresponding to that emotion. The specific configuration and operating procedure are shown below.

[1206] System Configuration

[1207] 1. Data collection methods:

[1208] The server collects data on professional players of games such as Go, Shogi, Chess, and card games, including game records and playing styles. The data is obtained from online databases or official websites and stored in a database.

[1209] 2. Means of generating artificial intelligence models:

[1210] The server trains an artificial intelligence model based on the collected data. Deep learning technology is used to design a neural network and train it to reproduce the playing style of a professional. Frameworks such as TensorFlow and Keras are used.

[1211] 3. Optimization methods for artificial intelligence models:

[1212] The server evaluates the performance of the generated AI model and performs optimization, improving the model's accuracy through cross-validation and hyperparameter tuning.

[1213] 4. Means of provision to user terminal:

[1214] The server uploads the optimized AI model to a cloud server and provides it to the user's device via an API, allowing the user to play against the AI ​​model even offline.

[1215] 5. Emotion recognition means:

[1216] The user device uses a camera and microphone to collect the user's facial expressions and voice in real time, and uses algorithms to recognize the user's emotional state. This data is reflected in AI responses and feedback during the match.

[1217] 6. Competition and feedback channels:

[1218] The user opens the app and begins a match against a professional copy AI. During the match, the user's device predicts the AI's next move and displays it to the user. The emotion engine monitors the user's emotional state in real time and adjusts the AI's response. After the match ends, the match results and emotion data are sent to the server, which provides analysis results and feedback.

[1219] Hardware and software used

[1220] Hardware: Smartphones, smart glasses, head-mounted displays, robots, webcams

[1221] Software: TensorFlow, Keras, facial_expression_recognition library

[1222] Examples of concrete examples and prompts

[1223] For example, in a Go match, a user opens a smartphone app and taps the "Start Match" button. During the match, the camera captures the user's face and analyzes their facial expressions. If the user shows signs of anxiety or irritation, the app displays a message saying, "Would you like to take a break?" After the match ends, the results and emotional data are sent to a server, and technical and psychological feedback is received.

[1224] Example prompt sentence:

[1225] "If the user's emotions show signs of anxiety or irritation during a Go match, implement a system that displays an appropriate message to encourage the user to take a break."

[1226] With the above system, users can learn the techniques of professional players while also learning efficiently while controlling their own emotions.

[1227] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1228] Step 1:

[1229] The server collects data from professional players. Specifically, it retrieves match records for Go, Shogi, chess, and card games from online databases and official websites, and stores the data in an internal database. The input is the match record data of professional players, and the output is formatted match record data. To ensure data consistency and quality, data processing such as format conversion and deletion of duplicate data is performed.

[1230] Step 2:

[1231] The server trains an AI model based on the collected data. Specifically, the collected match records are input into a neural network using deep learning frameworks such as TensorFlow and Keras, and training is performed to reproduce the playing style of professional players. The input is formatted match record data, and the output is a trained AI model. Here, data calculations such as model construction, preprocessing of learning data, and training are performed.

[1232] Step 3:

[1233] The server evaluates and optimizes the performance of the generated AI model. Specifically, it evaluates the predictive accuracy of the trained model and improves the model's accuracy using cross-validation and hyperparameter tuning. The input is the trained AI model and evaluation data, and the output is an optimized AI model. Parameters are adjusted to prevent overfitting and increase versatility.

[1234] Step 4:

[1235] The server uploads the optimized AI model to the cloud server and provides it to the user's device via API. The input is the optimized AI model, and the output is the model distribution to the user's device. The model is downloaded to the user's device so that the user can play offline.

[1236] Step 5:

[1237] The user device uses a camera and microphone to collect the user's facial expressions and voice in real time and recognize their emotional state. Specifically, the images captured by the camera are input into an emotion recognition algorithm to classify the user's emotional state. The input is the user's facial images and voice data captured in real time, and the output is the user's emotional state data. Here, facial expression analysis and voice analysis are performed.

[1238] Step 6:

[1239] A user starts a match against an artificial intelligence model, and the user's device predicts the AI's next move during the match. The input is the user's movements during the match (such as the board state), and the output is the AI's next move. During the match, the emotion engine monitors the user's emotional state in real time and adjusts the AI's response and feedback content as necessary. This is where real-time predictions and emotional response adjustments take place.

[1240] Step 7:

[1241] After a user finishes a match, the match results and the user's emotional state data are sent to the server. The server analyzes this data and generates feedback including the user's weaknesses and areas for improvement during play. The input is the match results and emotional state data, and the output is the analysis results and feedback. Data analysis and feedback generation are performed.

[1242] Step 8:

[1243] The server sends the generated feedback to the user's device, which then displays it to the user. The input is the analysis results and feedback, and the output is the feedback display to the user. The feedback includes not only technical advice but also mental support based on the user's emotional state. This allows the user to improve both their playing skills and emotional management.

[1244] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the control target 443 to output the result of the specific processing. The microphone 238 acquires voice indicating a user input regarding the result of the specific processing. The control unit 46A transmits voice data indicating the user input acquired by the microphone 238 to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the voice data.

[1245] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is ChatGPT (Internet Search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search <url: https: gemini.google.com ?hl="ja">) and other generation AIs. The data generation model 58 is obtained by performing deep learning on a neural network. A prompt including an instruction is input to the data generation model 58, and inference data such as voice data indicating voice, text data indicating text, and image data indicating an image is also input. The data generation model 58 performs inference on the input inference data in accordance with the instruction indicated by the prompt, and outputs the inference result in a data format such as voice data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[1246] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of the present disclosure is not limited to this, and the specific processing may be performed by the robot 414.

[1247] The emotion identification model 59 as an emotion engine may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to an emotion map (see FIG. 9), which is a specific mapping. Similarly, the emotion identification model 59 may determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[1248] FIG. 9 is a diagram illustrating an emotion map 400 on which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. Emotions closer to the center of the concentric circles are more primitive. Emotions representing states and actions arising from a state of mind are arranged on the outer edges of the concentric circles. The concept of emotion includes both affect and mental states. Emotions generally generated from reactions occurring in the brain are arranged on the left side of the concentric circles. Emotions generally induced by situational judgment are arranged on the right side of the concentric circles. Emotions generally generated from reactions occurring in the brain and induced by situational judgment are arranged on the upper and lower sides of the concentric circles. Furthermore, the emotion of "pleasure" is arranged on the upper side of the concentric circles, and the emotion of "discomfort" is arranged on the lower side. In this way, in the emotion map 400, multiple emotions are mapped based on the structure by which emotions are generated, and emotions that tend to occur simultaneously are mapped close to each other.

[1249] These emotions are distributed in the 3 o'clock direction on emotion map 400, and typically fluctuate between relief and anxiety. In the right half of emotion map 400, situational awareness dominates over internal sensations, resulting in a sense of calm.

[1250] The inside of emotion map 400 represents what is going on in the mind, and the outside of emotion map 400 represents behavior, so the further you go outside emotion map 400, the more visible the emotions become (the more they are expressed in behavior).

[1251] Human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. Emotions can also be created for robots, automobiles, and motorcycles, based on various balances, such as posture and remaining battery life. When these balances deviate from the ideal, a state of discomfort is indicated, and when they approach the ideal, a state of pleasure is indicated. An emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on Voice Emotion Recognition and Emotional Brain Physiological Signal Analysis Systems, Tokushima University, Doctoral Dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map lists emotions belonging to the "reaction" domain, where sensation is dominant. The right half of the emotion map lists emotions belonging to the "situation" domain, where situational awareness is dominant.

[1252] The emotion map defines two emotions that promote learning. One is a negative emotion on the situation side, around the middle of "repentance" or "reflection." In other words, this occurs when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is a positive emotion on the response side, around "desire." In other words, this occurs when the robot experiences positive feelings such as "I want more" or "I want to know more."

[1253] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​indicating each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple pieces of training data that are combinations of user input and emotion values ​​indicating each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions that are located close to each other have similar values, as in the emotion map 900 shown in FIG. 10. FIG. 10 shows an example in which multiple emotions, "relieved," "calm," and "reassuring," have similar emotion values.

[1254] The system according to the present disclosure has been described above mainly with respect to the functions of the data processing device 12, but the system according to the present disclosure is not necessarily implemented on a server. The system according to the present disclosure may be implemented as a general information processing system. The present disclosure may be implemented, for example, as a software program running on a personal computer or an application running on a smartphone, etc. The method according to the present disclosure may be provided to users in the form of SaaS (Software as a Service).

[1255] In the above embodiment, an example was given in which the specific processing is performed by one computer 22, but the technology of the present disclosure is not limited to this, and the specific processing may be distributed and performed by a plurality of computers including the computer 22. For example, the data generation model 58 may be provided in an external device of the data processing device 12, and data may be generated in the external device in accordance with input data.

[1256] In the above embodiment, an example in which the specific processing program 56 is stored in the storage 32 has been described, but the technology of the present disclosure is not limited to this. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-transitory storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-transitory storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes the specific processing in accordance with the specific processing program 56.

[1257] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[1258] It is not necessary to store all of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store all of the specific processing program 56 in the storage 32; only a portion of the specific processing program 56 may be stored.

[1259] The hardware resource for executing a specific process can be any of the following processors: An example of a processor is a CPU, which is a general-purpose processor that functions as a hardware resource for executing a specific process by executing software, i.e., a program. Another example of a processor is a dedicated electrical circuit, such as an FPGA (Field-Programmable Gate Array), a PLD (Programmable Logic Device), or an ASIC (Application Specific Integrated Circuit), which is a processor with a circuit configuration designed specifically for executing a specific process. Each processor has built-in or connected memory, and each processor uses the memory to execute the specific process.

[1260] The hardware resource that executes the specific processing may be configured with one of these various processors, or may be configured with a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Also, the hardware resource that executes the specific processing may be a single processor.

[1261] As an example of a system configured with a single processor, first, one processor is configured by combining one or more CPUs and software, and this processor functions as a hardware resource that executes a specific process. Second, there is a system that uses a processor that realizes the functions of an entire system including multiple hardware resources that execute a specific process on a single IC chip, as typified by SoC (System-on-a-chip). In this way, a specific process is realized using one or more of the above-mentioned various processors as hardware resources.

[1262] Furthermore, the hardware structure of these various processors can be, more specifically, an electric circuit that combines circuit elements such as semiconductor devices. The specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps may be deleted, new steps may be added, or the processing order may be rearranged, without departing from the spirit of the invention.

[1263] The above-described description and illustrations are a detailed explanation of the parts related to the technology of the present disclosure and are merely an example of the technology of the present disclosure. For example, the above description of the configuration, functions, actions, and effects is an explanation of an example of the configuration, functions, actions, and effects of the parts related to the technology of the present disclosure. Therefore, it goes without saying that unnecessary parts may be deleted, new elements may be added, or replacements may be made to the above-described description and illustrations within the scope of the gist of the technology of the present disclosure. Furthermore, to avoid confusion and facilitate understanding of the parts related to the technology of the present disclosure, the above-described description and illustrations omit explanations of common technical knowledge that do not require particular explanation to enable the implementation of the technology of the present disclosure.

[1264] All publications, patent applications, and technical standards mentioned in this specification are herein incorporated by reference to the same extent as if each individual publication, patent application, or technical standard was specifically and individually indicated to be incorporated by reference.

[1265] The following is further disclosed regarding the above embodiment.

[1266] (Claim 1)

[1267] A means of collecting data on professional players,

[1268] A means for training a generative artificial intelligence model based on the collected data of professional players;

[1269] a means for optimizing the generative artificial intelligence model;

[1270] A means for providing an optimized generative artificial intelligence model to a user terminal;

[1271] A means for receiving and analyzing the results of a user's competition against the generating artificial intelligence model and providing feedback;

[1272] A system including:

[1273] (Claim 2)

[1274] 2. The system of claim 1, wherein the professional player data is data relating to any of Go, Shogi, Chess, and card games.

[1275] (Claim 3)

[1276] 2. The system of claim 1, wherein the generative artificial intelligence model is trained using deep learning techniques.

[1277] (Claim 4)

[1278] 2. The system according to claim 1, wherein the generative artificial intelligence model provided to the user terminal is provided via a cloud server.

[1279] (Claim 5)

[1280] 2. The system according to claim 1, wherein the means for analyzing the match data and providing feedback after the user plays against the generated artificial intelligence model provides optimal instruction based on the user's match history.

[1281] "Example 1"

[1282] (Claim 1)

[1283] A means of collecting data on professional players,

[1284] A means of preprocessing data based on the collected data of professional players;

[1285] means for training a generative artificial intelligence model using the preprocessed data;

[1286] A means for optimizing the generated artificial intelligence model using cross-validation and hyperparameter tuning;

[1287] A means for deploying the optimized generative artificial intelligence model in a cloud environment and providing it to a user terminal;

[1288] A means for a user to play against the generated AI model, receive and analyze the results of the game, and provide feedback;

[1289] A system including:

[1290] (Claim 2)

[1291] 2. The system of claim 1, wherein the professional player data is data relating to a strategy board game, such as Go, Shogi, Chess, or a card game.

[1292] (Claim 3)

[1293] 2. The system of claim 1, wherein the generative artificial intelligence model is trained using deep learning techniques.

[1294] "Application Example 1"

[1295] (Claim 1)

[1296] a means for collecting data on professional workers;

[1297] A means for training a generative artificial intelligence model based on the collected data of professional workers;

[1298] a means for optimizing the generative artificial intelligence model;

[1299] A means for providing an optimized generative artificial intelligence model to a user terminal;

[1300] A means for receiving and analyzing the results of a user's competition or collaboration with the generated artificial intelligence model and providing feedback;

[1301] A system including:

[1302] (Claim 2)

[1303] 2. The system of claim 1, wherein the professional worker data is data relating to any of a manufacturing process, equipment operation, or quality control.

[1304] (Claim 3)

[1305] 2. The system of claim 1, wherein the generative artificial intelligence model is trained using deep learning techniques.

[1306] "Example 2: Combining Emotion Engines"

[1307] (Claim 1)

[1308] A means of collecting data on professional players,

[1309] A means for training an artificial intelligence model generated based on the collected data of professional players;

[1310] a means for optimizing the generative artificial intelligence model;

[1311] A means for providing an optimized generative artificial intelligence model to a user terminal;

[1312] A means for the user terminal to collect and analyze user emotion data using a camera or microphone;

[1313] A means to adjust match content and feedback based on user emotional data in real time;

[1314] A means for receiving and analyzing the results of a user's competition against the generating artificial intelligence model and providing feedback;

[1315] A system including:

[1316] (Claim 2)

[1317] 2. The system of claim 1, wherein the professional player data is data relating to a board game or a card game.

[1318] (Claim 3)

[1319] 2. The system of claim 1, wherein the generative artificial intelligence model is trained using deep learning techniques.

[1320] "Application example 2 when combining emotion engines"

[1321] (Claim 1)

[1322] A means of collecting data on professional players,

[1323] A means for training a generative artificial intelligence model based on the collected data of professional players;

[1324] a means for optimizing the generative artificial intelligence model;

[1325] A means for providing an optimized generative artificial intelligence model to a user terminal;

[1326] A means for receiving and analyzing the results of a user's competition against the generating artificial intelligence model and providing feedback;

[1327] A means for recognizing a user's emotional state in real time and adjusting the content of the match or feedback based on the user's emotional state;

[1328] A system including:

[1329] (Claim 2)

[1330] The system of claim 1, wherein the professional player data is data relating to any of Go, Shogi, chess, or card games, and the user uses a smartphone, smart glasses, a head-mounted display, or a robot.

[1331] (Claim 3)

[1332] 2. The system of claim 1, wherein the generative artificial intelligence model is trained using deep learning techniques and further includes an algorithm used to analyze the user's emotional state. [Explanation of symbols]

[1333] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Device 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robot< / url:> < / url:> < / url:> < / url:>

Claims

1. A means of collecting data on professional players, A means for training a generative artificial intelligence model based on the collected data of professional players; a means for optimizing the generative artificial intelligence model; A means for providing an optimized generative artificial intelligence model to a user terminal; A means for receiving and analyzing the results of a user's competition against the generating artificial intelligence model and providing feedback; A system including:

2. 2. The system of claim 1, wherein the professional player data is data relating to any one of Go, Shogi, Chess, and card games.

3. The system of claim 1 , wherein the generative artificial intelligence model is trained using deep learning techniques.

4. The system according to claim 1, wherein the generative artificial intelligence model provided to the user terminal is provided via a cloud server.

5. 2. The system according to claim 1, wherein the means for analyzing the battle data and providing feedback after the user plays a game against the generated artificial intelligence model provides optimal instruction based on the user's battle history.

Images