System

Abstract

A system is provided.SOLUTION: A system comprising: means for collecting vast amounts of data; means for pre-processing and integrating the collected data; means for training a generative artificial intelligence model using the pre-processed and integrated data; means for generating a virtual environment similar to a real world using the trained generative artificial intelligence model; means for performing a simulation within the generated virtual environment; and means for analyzing and displaying results of the simulation.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] This invention aims to solve the problem of limited information required for decision-making. In the real world, individuals and companies require a large amount of information and experience to make accurate decisions, but if these are insufficient, they are more likely to make incorrect decisions or fail. Furthermore, while failure is considered part of learning, it involves actual losses and risks and is therefore something that should be avoided if possible. Therefore, there is a need for a system that can improve decision-making accuracy and minimize risk by virtualizing real-world failures and simulating a large number of scenarios. [Means for solving the problem]

[0005] The present invention provides a means for collecting massive amounts of data, preprocessing and integrating the collected data, training a generative artificial intelligence model using the preprocessed and integrated data, and generating a virtual environment similar to the real world using the trained model. By running a simulation in the generated virtual environment and analyzing and displaying the results of the simulation, users can input specific scenarios and make decisions based on the simulation results. This series of means makes it possible to reduce the risk of failure in the real world within the virtual environment and support optimal decision-making.

[0006] "Big data" refers to a collection of large amounts of information collected from various sources, and is data of such a large scale that it cannot be processed using conventional methods.

[0007] "Preprocessing" refers to the initial processing of collected data, such as removing noise, filling in missing values, and standardizing data, to ensure that subsequent processing can be performed accurately and efficiently.

[0008] "Integration" refers to the process of bringing together data collected from different sources into a single, unified format or structure, making it consistent.

[0009] A "generative artificial intelligence model" is an AI model that uses machine learning algorithms to learn patterns and trends in input data and generate and predict future data.

[0010] "Training" refers to the process of using a dataset to train a generative artificial intelligence model and improve its performance.

[0011] A "virtual environment" is a simulated environment that has similar characteristics to the real world but is generated in a digital space, allowing users to test various scenarios.

[0012] "Simulation" refers to the process of modeling specific conditions or behaviors in a virtual environment and observing and analyzing the results.

[0013] "Analysis" refers to the process of examining the results of a simulation in detail, identifying trends and patterns in the data, and deepening understanding.

[0014] "Display" refers to the process of visually presenting analytical results in a format that is easy for users to understand (e.g., graphs or reports).

[0015] "User" refers to an individual or organization that uses the system to run simulations and make decisions based on the results.

[0016] A "specific scenario" refers to a description, in words or numbers, of the specific conditions or actions that a user wants to test in a virtual environment.

[0017] "Decision support" refers to a function that assists users in making decisions based on simulation results and suggests optimal choices.

[0018] "Feedback" refers to reactions and comments from users regarding results and suggestions obtained from the system, including information that can be used to improve the system. [Brief explanation of the drawings]

[0019] [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

[0020] 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.

[0021] First, the terms used in the following description will be explained.

[0022] 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).

[0023] 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.

[0024] 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.

[0025] 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.

[0026] 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."

[0027] [First embodiment]

[0028] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0029] 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.

[0030] 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).

[0031] 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.

[0032] 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.

[0033] 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.

[0034] 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.

[0035] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0036] 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.

[0037] 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.

[0038] 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.

[0039] 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."

[0040] This invention provides a system that uses generative artificial intelligence to build a virtual environment based on a huge amount of data and performs simulations within that virtual environment. This system aims to minimize risk by simulating real-world failures in the virtual environment in advance in order to improve the accuracy of decision-making.

[0041] System configuration and operation

[0042] 1. Data collection and integration

[0043] The servers collect vast amounts of data from various sources, including corporate, personal, and public data, using techniques such as APIs, database queries, and web scraping.

[0044] The server pre-processes the collected data, cleaning, normalizing, and consolidating it, transforming data from different sources into a unified format and making it consistent.

[0045] 2. Training a generative AI model

[0046] The server uses the preprocessed and integrated data to train a generative artificial intelligence model, selecting appropriate machine learning algorithms and training the model to learn patterns and trends in the data.

[0047] The training dataset is split into training, validation, and testing sets and is used to evaluate the performance of the model.

[0048] 3. Building a Virtual Environment

[0049] The server uses trained generative artificial intelligence models to generate virtual environments similar to the real world, allowing users to freely configure the scenarios they want to try out, recreating a wide range of scenarios and variables.

[0050] 4. Running the Simulation

[0051] The user inputs the specific scenario they want to simulate through the terminal, including specific settings such as new product features, pricing, and carrier selection.

[0052] The terminal transmits the user's input to the server, which runs the simulation within the virtual environment.

[0053] 5. Analyzing and displaying simulation results

[0054] The server analyzes the simulation results, finding trends and patterns in the data, and then uses specific statistical analysis and machine learning techniques to summarize the results in an easy-to-understand format.

[0055] The terminal visually displays the analysis results and provides them to the user, including graphs, charts, reports, etc.

[0056] 6. Decision support

[0057] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback and performs further simulations or re-adjusts the model as needed.

[0058] Specific examples

[0059] Case 1: A company decides to launch a new product

[0060] The user (corporate decision maker) inputs scenarios such as the characteristics and pricing of new products into the terminal.

[0061] The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions.

[0062] The terminal presents the analysis results to the user in the form of graphs and reports.

[0063] The user decides on a new product launch strategy based on the simulation results.

[0064] Case 2: Individual career choices

[0065] The user inputs a scenario regarding career choices (school choice, occupation choice, etc.) into the terminal.

[0066] The server simulates scenarios after career choices in a virtual environment and analyzes future annual income and job satisfaction.

[0067] The terminal presents the analysis results to the user as a report.

[0068] The user selects a carrier based on the simulation results.

[0069] In this way, the system of the present invention can support user decision-making and minimize risks through highly accurate simulations based on a variety of data.

[0070] The processing flow will be explained below.

[0071] Step 1: Data collection

[0072] The server retrieves data from a variety of sources, including corporate databases, public data services, and public data sources on the internet, using techniques such as APIs, database queries, and web scraping.

[0073] The server periodically collects data and keeps the necessary information up to date.

[0074] Step 2: Data Preprocessing

[0075] The server removes noise from the acquired data and fills in missing values. For example, missing data is interpolated using the average value or deleted.

[0076] The server performs data standardization and normalization, converting data from various formats into a unified format.

[0077] Step 3: Data Integration

[0078] The server consolidates the pre-processed data into one unified database.

[0079] The server maps and cross-references data schemas so that data from different sources is consistent.

[0080] Step 4: Prepare the dataset

[0081] The server extracts the training dataset from the integrated database and splits it into training, validation, and test datasets.

[0082] The server labels the data and performs feature engineering to prepare it in the optimal format for model training.

[0083] Step 5: Training the generative AI model

[0084] The server selects an appropriate generative AI algorithm and trains the model using a training dataset.

[0085] The server monitors the training process and tunes hyperparameters as needed.

[0086] Step 6: Evaluate the generative AI model

[0087] The server evaluates the model's performance using a validation dataset, using metrics such as precision, recall, and F1 score.

[0088] The server improves and retrains the model based on the evaluation results.

[0089] Step 7: Generate a Virtual Environment

[0090] The server uses a trained generative AI model to generate a virtual environment similar to the real world.

[0091] The server sets variables and parameters within the virtual environment to recreate realistic scenarios.

[0092] Step 8: Entering the Scenario

[0093] The user inputs the specific scenario he or she wants to simulate through the terminal.

[0094] The terminal collects the user's input and sends it to the server.

[0095] Step 9: Run the simulation

[0096] The server executes a simulation in a virtual environment based on the received specific scenario.

[0097] The server simulates multiple scenarios in parallel and collects data.

[0098] Step 10: Analyze the simulation results

[0099] The server analyzes the simulation results and finds trends and patterns in the data.

[0100] The server uses statistical analysis and machine learning techniques to compile the results into an easy-to-understand format.

[0101] Step 11: View the results

[0102] The terminal visually displays the simulation results sent from the server, including graphs, charts, and reports.

[0103] The user checks the displayed results and decides on the next action to take.

[0104] Step 12: Gather feedback

[0105] The user provides feedback on the simulation results through the terminal.

[0106] The terminal sends the collected feedback to the server.

[0107] Step 13: Further simulation

[0108] The server simulates additional scenarios or retunes the model based on user feedback.

[0109] The user inputs a new scenario as necessary and runs the simulation again.

[0110] In this way, the systems work together to help users make better decisions.

[0111] Example 1

[0112] 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."

[0113] Decisions in the real world involve many risks and uncertainties, so prior simulations are required to minimize risk. However, with conventional systems, the process from collecting and integrating massive amounts of data to training AI models, running simulations, and analyzing the results is often complex and inefficient, resulting in insufficient accuracy. It is also difficult for users to efficiently input specific scenarios and quickly make decisions based on the analysis results.

[0114] 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.

[0115] In this invention, the server includes means for collecting a huge amount of data, means for preprocessing and integrating the collected data, means for training a generative AI model using the preprocessed and integrated data, means for generating a virtual environment similar to the real world using the trained generative AI model, means for executing a simulation in the generated virtual environment, means for analyzing and displaying the results of the simulation, means for a user to input a specific scenario they wish to simulate, and means for automatically generating and analyzing multiple scenarios in the virtual environment. This enables users to make quick and accurate decisions through highly accurate simulations based on a huge amount of data.

[0116] "Big data" refers to data collected in large quantities from various sources, including corporate data, personal data, and public data.

[0117] "Preprocessing" refers to the process of cleaning (filling in missing values, removing outliers, etc.) and normalizing (formatting the data) the collected data.

[0118] "Integration" is the process of transforming data collected from different sources into a consistent format so that it can be treated as a single data set.

[0119] A "generative artificial intelligence model" is a model trained using machine learning algorithms to learn patterns and trends in data and perform simulations and predictions.

[0120] "Training" refers to the process of training a generative artificial intelligence model using preprocessed and integrated data.

[0121] "Virtual environment" refers to a simulated space that resembles the real world and is generated using a trained generative artificial intelligence model.

[0122] "Simulation" refers to the process of conducting experiments and verifications in a virtual environment based on scenarios and conditions specified by the user.

[0123] "Analysis" refers to the process of finding trends and patterns in the data from the simulation results, specifically using statistical analysis and machine learning techniques.

[0124] "Display" refers to providing the analyzed simulation results to the user in a visual format (graphs, charts, reports, etc.).

[0125] A "specific scenario" refers to the specific conditions or settings that the user wants to simulate (e.g., new product characteristics, pricing, carrier selection, etc.).

[0126] "Automatically generating and analyzing multiple scenarios" refers to the process of automatically generating multiple scenarios within a virtual environment based on user input and analyzing the results of each scenario.

[0127] MODE FOR CARRYING OUT THE INVENTION

[0128] This invention relates to a system that uses a generative artificial intelligence model based on a huge amount of data to create a virtual environment and perform simulations within that virtual environment. The system aims to minimize risk by simulating real-world failures in the virtual environment in advance in order to improve the accuracy of user decision-making.

[0129] System configuration

[0130] Hardware and Software Configuration

[0131] server:

[0132] Data collection is done using APIs (e.g., Twitter API, Google Analytics API), databases (e.g., MySQL, PostgreSQL), and web scraping tools (e.g., BeautifulSoup, Scrapy).

[0133] Use data cleansing tools (e.g., Pandas) for data preprocessing and integration.

[0134] Use machine learning libraries (e.g., TensorFlow, PyTorch) to train generative artificial intelligence models.

[0135] Statistical analysis tools (e.g., R, SciPy) and data analysis platforms (e.g., Jupyter Notebook, MATLAB) will be used for simulation and result analysis.

[0136] Device:

[0137] Use visualization tools (e.g., Matplotlib, Tableau) to display the results.

[0138] Data collection and preprocessing

[0139] The servers collect vast amounts of data, including corporate, personal, and public data, from various sources, including APIs, databases, and web scraping.

[0140] Example: A server retrieves the latest tweets from the Twitter API at regular intervals and stores them in a MySQL database.

[0141] The server cleans, normalizes, and consolidates the collected data, transforming it from different sources into a consistent format.

[0142] For example, the server uses Pandas to impute missing values, remove outliers, and convert each data set into a standard format.

[0143] Training generative artificial intelligence models

[0144] Based on the preprocessed data, the server selects an appropriate machine learning algorithm and trains a generative artificial intelligence model.

[0145] Example: The server uses TensorFlow to split the dataset into training, validation, and test data, and train and evaluate the model.

[0146] Building a virtual environment

[0147] The server uses a trained generative AI model to generate a virtual environment similar to the real world, which can be flexibly changed depending on the scenario or conditions the user wants to try.

[0148] Example: The server uses a generative AI model to generate various market scenarios based on the price range and market conditions set by the user.

[0149] Running and analyzing the simulation

[0150] Users input the specific scenario they want to simulate through the device, including the features and pricing of new products, carrier selection, and so on.

[0151] Example: A user enters prices and feature settings into the device interface and sends them to the server.

[0152] The server receives the scenario sent by the user and executes the simulation in the virtual environment.

[0153] Example: The server uses a generative AI model to run parallel simulations of scenarios based on input conditions.

[0154] The server analyzes the simulation results and finds trends and patterns in the data, using statistical analysis and machine learning techniques.

[0155] Example: The server uses SciPy to perform statistical analysis of the simulation results and graph sales forecasts and market reactions.

[0156] Results presentation and decision support

[0157] The terminal presents the analysis results to the user in a visual format, for example in the form of graphs, charts, or reports.

[0158] Example: The terminal receives the analysis results from the server and displays the results as a graph using Matplotlib.

[0159] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback from the user and performs further simulations or readjusts the model as needed.

[0160] Example: The user decides on a new product launch strategy based on the results and sends that feedback to the server, which uses the feedback to recalibrate the model.

[0161] Examples of concrete examples and prompts

[0162] Case 1: A company decides to launch a new product

[0163] Example of prompt text entered by the user (corporate decision maker):

[0164] Please define the following characteristics for your new product:

[0165] Function A

[0166] Price B

[0167] Target Market C

[0168] Simulate market reactions based on these conditions and analyze projected sales and market share.

[0169] The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions.

[0170] The terminal presents the analysis results to the user in the form of graphs and reports.

[0171] The user decides on a new product launch strategy based on the simulation results.

[0172] Case 2: Individual career choices

[0173] An example of a prompt that the user might enter:

[0174] "Enter the following career choice scenario:

[0175] Occupation X

[0176] University Y

[0177] Based on these, please analyze your future annual income and job satisfaction.

[0178] The server simulates scenarios after career choices in a virtual environment and analyzes future annual income and job satisfaction.

[0179] The terminal presents the analysis results to the user as a report.

[0180] The user makes a carrier selection based on the simulation results.

[0181] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0182] Step 1: Data collection

[0183] The server uses APIs (e.g., Twitter API, Google Analytics API), databases (e.g., MySQL, PostgreSQL), and web scraping tools (e.g., BeautifulSoup, Scrapy) to collect huge amounts of data, including corporate data, personal data, and public data.

[0184] Input: API endpoint, database query, website URL.

[0185] What it does: The server periodically fetches data from the API and stores it in a MySQL database. It also uses a web scraping tool to crawl websites and retrieve the required data.

[0186] Output: The raw data collected.

[0187] Step 2: Data Preprocessing

[0188] The server pre-processes the collected data, cleaning, normalizing, and integrating it, transforming data from different sources into a consistent format.

[0189] Input: Raw data collected.

[0190] Specific operation: The server uses Pandas to impute missing values, remove outliers, and convert each data into a standard format.

[0191] Output: The preprocessed dataset.

[0192] Step 3: Training the generative artificial intelligence model

[0193] Based on the preprocessed data, the server selects an appropriate machine learning algorithm and trains a generative artificial intelligence model.

[0194] Input: The preprocessed dataset.

[0195] Specific operation: The server uses TensorFlow to split the dataset into training data, validation data, and test data, and trains and evaluates the model.

[0196] Output: A trained generative AI model.

[0197] Step 4: Build a virtual environment

[0198] The server uses a trained generative AI model to generate a virtual environment similar to the real world, which can be flexibly changed depending on the scenario or conditions the user wants to try.

[0199] Input: A trained generative AI model.

[0200] Specific operation: The server uses the generative AI model to set various parameters within the virtual environment and provide scenario templates that can be customized by the user.

[0201] Output: A customizable virtual environment.

[0202] Step 5: Enter the simulation scenario

[0203] Users input the specific scenario they want to simulate through the device, including new product features, pricing, carrier selection, etc.

[0204] Input: Scenario conditions set by the user (e.g., new product characteristics, pricing).

[0205] Specific operation: The user inputs the required information into the terminal interface and sends it to the server.

[0206] Output: User's scenario configuration data.

[0207] Step 6: Run the simulation

[0208] The server receives the scenario sent by the user and executes the simulation in the virtual environment. The simulation automatically generates and analyzes multiple scenarios based on the specified conditions.

[0209] Input: User-submitted scenario configuration data, customizable virtual environment.

[0210] Specific operation: The server uses the generative AI model to simulate multiple scenarios in parallel and record the results.

[0211] Output: Simulation result data.

[0212] Step 7: Analyze the simulation results

[0213] The server analyzes the simulation results and finds trends and patterns in the data, using statistical analysis and machine learning techniques.

[0214] Input: Simulation result data.

[0215] How it works: The server uses SciPy to statistically analyze the data and visualize the results, such as sales forecasts and market reactions.

[0216] Output: Analyzed simulation results.

[0217] Step 8: View the results

[0218] The terminal visually displays the analysis results and provides them to the user, specifically in the form of graphs, charts, and reports.

[0219] Input: Analyzed simulation results.

[0220] Specific operation: The terminal receives the analysis results sent from the server and visually displays the results using Matplotlib.

[0221] Output: The visualization data that is presented to the user.

[0222] Step 9: Decision making and feedback

[0223] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback from the user and performs further simulations or readjusts the model as needed.

[0224] Input: User decision results, feedback data.

[0225] How it works: The user decides on a strategy based on the results and enters it into the device. The server receives the feedback and retrains the model as needed.

[0226] Output: Retuned generative AI model, further simulation results.

[0227] (Application example 1)

[0228] 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."

[0229] In content distribution services, accurately predicting user viewing trends and engagement and determining optimal distribution schedules based on the results is a very difficult challenge. Conventional methods only allow for limited predictions based on past data and experience, and are unable to effectively capture viewer reactions. Therefore, new technologies are needed to optimize content distribution schedules and improve engagement.

[0230] 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.

[0231] In this invention, the server includes means for collecting a huge amount of data, means for preprocessing and integrating the collected data, means for training a generative artificial intelligence model using the preprocessed and integrated data, means for generating a virtual environment similar to the real world using the trained generative artificial intelligence model, means for running a simulation in the generated virtual environment, means for analyzing and displaying the results of the simulation, means for predicting viewer responses based on the collected data, means for predicting an engagement score for adjusting a content delivery schedule, and means for visualizing the simulation results and providing them to a user. This makes it possible to predict viewer responses with high accuracy and to formulate an optimal delivery schedule based on the predictions.

[0232] "Big data" refers to large amounts of digital information collected from a wide variety of sources.

[0233] "Preprocessing" refers to a series of steps that transform collected data into a form that is applicable for analysis and modeling.

[0234] "Integration" refers to the act of bringing together data collected from different sources into a consistent format.

[0235] A "generative artificial intelligence model" refers to a machine learning model trained on collected and preprocessed data.

[0236] A "virtual environment" refers to a digital space created to simulate real-world actions or events.

[0237] "Simulation" refers to the process of experimenting with behaviors and outcomes under specific conditions within a virtual environment.

[0238] "Analysis" refers to the process of summarizing simulation results and data trends in an easy-to-understand format.

[0239] "Display" refers to the act of visually presenting the analysis results to the user.

[0240] "Viewer response" refers to the viewer's behavior and reaction to content.

[0241] "Engagement score" refers to a numerical indicator of the level of engagement that viewers show with content.

[0242] "Content delivery schedule" refers to a plan that determines when particular content should be delivered.

[0243] A "scenario" refers to the specific conditions and settings that a user inputs to perform a simulation.

[0244] "Optimization" refers to adjusting a system or process to its best state according to a specific goal.

[0245] The system of the present invention provides a means to collect, preprocess, and integrate massive amounts of data, train generative artificial intelligence models to generate virtual environments that resemble the real world, and run simulations within those virtual environments to determine optimal content delivery schedules.

[0246] The system includes a server, a terminal, and a user.

[0247] Data collection and preprocessing

[0248] The server collects a huge amount of data using techniques such as APIs, database queries, and web scraping. The collected data includes information such as viewer viewing history, viewing times, and behavioral patterns. The server then cleans and normalizes the data, integrating it for consistency.

[0249] Training generative artificial intelligence models

[0250] Using the preprocessed and integrated data, the server trains a generative artificial intelligence model. The training dataset is divided into training, validation, and testing datasets, and the target machine learning algorithm predicts viewer viewing habits. The model is designed to predict viewer behavior and reactions, and the algorithm used for training can be a linear regression model.

[0251] Virtual environment generation and simulation

[0252] Using a trained generative artificial intelligence model, the server generates a virtual environment similar to the real world. Users can input specific scenarios via their devices, specifying specific dates and times, the number of views, likes, shares, etc. The server then runs a simulation based on these scenarios and predicts viewer reactions.

[0253] Analyzing and displaying results

[0254] The results of the simulation are analyzed by the server. The analysis results include viewer reactions and engagement scores. The device visualizes these results and provides them to the user. The user can view the results in the form of graphs, charts, and reports.

[0255] Example

[0256] For example, if a user wants to stream at 8pm with the hope of getting 5000 views, 300 likes, and 50 shares, they can enter a specific scenario. This scenario can be expressed with a prompt like this:

[0257] Based on your viewing data, predict your viewer engagement score for the following stream schedule:

[0258] Views: 5000

[0259] Likes: 300

[0260] Shares: 50

[0261] Time: 8 PM

[0262] Based on this prompt, the server runs a simulation in the virtual environment to predict the viewer's reaction. The prediction results are provided to the user via their device, and the user can then decide on the optimal distribution schedule.

[0263] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0264] Step 1:

[0265] The server collects a huge amount of data. The types of data include viewer viewing history, viewing times, and behavioral patterns. This data is obtained using APIs, database queries, and web scraping techniques. Specifically, data is obtained by issuing queries to a viewing history database. The input in this step is the viewer data source, and the output is raw viewer data.

[0266] Step 2:

[0267] The server preprocesses and integrates the collected data. It removes noise from the data, imputes missing values, and converts various data sources into a consistent format. This step includes cleaning and normalizing the data. Specifically, it uses data cleaning tools to handle missing values. The input to this step is the collected data from step 1, and the output is the preprocessed and integrated data.

[0268] Step 3:

[0269] The server trains a generative artificial intelligence model using the preprocessed and integrated data. Specifically, it splits the dataset into training, validation, and test sets, and trains the model using a machine learning algorithm such as a linear regression model. The input in this step is the preprocessed data from step 2, and the output is a trained AI model.

[0270] Step 4:

[0271] The server generates a virtual environment using a trained generative AI model. The virtual environment is a digital space for simulating viewer behavior. Various scenarios and variables can be set, enabling various simulations. The input in this step is the trained AI model, and the output is the virtual environment.

[0272] Step 5:

[0273] Through the terminal, the user inputs the specific scenario they want to simulate. The scenario includes parameters such as a specific date and time, expected number of viewers, number of likes, number of shares, etc. The input in this step is the user's scenario information, and the output is the scenario data sent to the server.

[0274] Step 6:

[0275] The server runs a simulation in the virtual environment based on the input scenario. Specifically, it uses the model to generate data to predict viewer reactions and then runs the simulation. The input in this step is the scenario data from step 5, and the output is the simulation results.

[0276] Step 7:

[0277] The server analyzes the simulation results. The analysis aims to predict viewer responses and engagement scores. This procedure includes analyzing data trends using statistical analysis and machine learning techniques. The input in this step is the simulation results, and the output is the analysis results.

[0278] Step 8:

[0279] The device visualizes the analysis results and provides them to the user. Specifically, the analysis results are displayed in the form of graphs, charts, and reports. The user views these results and determines the optimal content distribution schedule based on the engagement scores. The input in this step is the analysis results, and the output is the visualized analysis results provided to the user.

[0280] By following the steps above, it becomes possible to predict viewer reactions with high accuracy and formulate an optimal distribution schedule.

[0281] 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.

[0282] The present invention is a system that collects massive amounts of data, integrates and preprocesses them, trains a generative artificial intelligence model, and runs simulations in a virtual environment using the generated model. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to support more accurate decision-making that takes into account the user's emotional state. The present invention is implemented using the following processing steps and configuration.

[0283] System configuration and operation

[0284] 1. Data collection and integration

[0285] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data, and uses techniques such as APIs, database queries, and web scraping to efficiently retrieve the data.

[0286] The server performs preprocessing on the collected data, such as noise removal, missing value filling, and data standardization, to ensure consistency.

[0287] 2. Training a generative AI model

[0288] The server uses the preprocessed and integrated data to train a generative artificial intelligence model, selecting an appropriate machine learning algorithm and splitting the dataset into training, validation, and testing sections.

[0289] Evaluate the performance of the model and adjust the hyperparameters as necessary.

[0290] 3. Creating a virtual environment

[0291] The server uses a trained generative artificial intelligence model to generate a virtual environment similar to the real world, with the flexibility to recreate a variety of scenarios.

[0292] 4. Enter the scenario

[0293] The user inputs the specific scenario they want to simulate through the terminal, including, for example, the characteristics and pricing of a new product, carrier selection, etc.

[0294] The terminal transmits the input scenario to the server.

[0295] 5. Collecting Emotion Data Using an Emotion Engine

[0296] The device collects emotional data from the user's facial expressions, voice, input, etc.

[0297] The emotion engine analyzes the collected emotion data and recognizes the user's emotional state in real time.

[0298] 6. Running the Simulation

[0299] The server runs a simulation in the virtual environment based on the received scenario and emotion data, dynamically adjusting the scenario based on the emotion data to provide more realistic results.

[0300] 7. Analyzing and displaying simulation results

[0301] The server analyzes the simulation results and uses statistical analysis and machine learning techniques to summarize the results in an easy-to-understand format.

[0302] The terminal provides the user with a visual display of the analysis results, including graphs, charts, and reports.

[0303] 8. Decision support and feedback gathering

[0304] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback from the user and performs additional simulations or refines the model.

[0305] Specific examples

[0306] Case 1: A company decides to launch a new product

[0307] Users (corporate decision makers) input scenarios such as new product characteristics and pricing into the terminal, and their emotional state (e.g., excitement, anxiety, etc.) is collected at the same time.

[0308] The emotion engine analyzes the user's emotional state in real time and transmits it to the server.

[0309] The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions taking into account emotional data.

[0310] The terminal presents the analysis results to the user in the form of graphs and reports.

[0311] The user decides on a new product launch strategy based on the simulation results.

[0312] Case 2: Individual career choices

[0313] The user inputs a scenario regarding career choices (such as the school to attend, the occupation to choose, etc.) into the terminal, and their emotional state (such as anxiety, expectations, etc.) is collected at the same time.

[0314] The emotion engine analyzes the user's emotional state and transmits it to the server.

[0315] The server simulates scenarios after career selection and analyzes future annual income and job satisfaction taking into account emotional data.

[0316] The terminal presents the analysis results to the user as a report.

[0317] The user selects a carrier based on the simulation results.

[0318] In this way, the system of the present invention, which is combined with an emotion engine, provides highly accurate simulation and decision support that takes into account the user's emotional state, helping the user make optimal choices.

[0319] The processing flow will be explained below.

[0320] Step 1: Data collection

[0321] The server retrieves data from corporate databases, public data services, and public data sources on the Internet, using techniques such as APIs, database queries, and web scraping to efficiently gather data.

[0322] The server periodically updates the data to keep it up to date.

[0323] Step 2: Data Preprocessing

[0324] The server removes noise from the acquired data and fills in missing values. Specifically, it interpolates the average value of missing data and removes outliers.

[0325] The server performs data standardization (eg, normalization, scaling) and converts data from different sources into a consistent format.

[0326] Step 3: Data Integration

[0327] The server consolidates the pre-processed data into one unified database.

[0328] The server maps the schema of the data from different sources and makes it consistent.

[0329] Step 4: Prepare the dataset

[0330] The server extracts training, validation, and test datasets from the integrated database.

[0331] The server labels the data and performs feature engineering to prepare it in a format that is easy for the model to learn.

[0332] Step 5: Training the generative AI model

[0333] The server selects an appropriate generative AI algorithm (e.g., a deep learning model) and trains the model using a training dataset.

[0334] The server monitors the training process and optimizes hyperparameters as needed.

[0335] Step 6: Evaluate the generative AI model

[0336] The server evaluates the model's performance using a validation dataset, using metrics such as precision, recall, and F1 score.

[0337] The server will make any necessary improvements based on the evaluation results.

[0338] Step 7: Generate a Virtual Environment

[0339] The server uses a trained generative AI model to generate a virtual environment that resembles the real world.

[0340] The server sets the variables and parameters to be simulated within the virtual environment.

[0341] Step 8: Entering the Scenario

[0342] The user inputs the specific scenario he or she wants to simulate (for example, the characteristics and pricing of a new product, carrier selection, etc.) through the terminal.

[0343] The terminal transmits the user's input to the server.

[0344] Step 9: Collect emotion data

[0345] The device collects emotional data from the user's facial expressions, voice, input, etc. This is typically done using devices such as a camera and microphone.

[0346] The emotion engine analyzes the collected data and recognizes the user's emotional state in real time.

[0347] Step 10: Run the simulation

[0348] The server executes a simulation in the virtual environment based on the scenario received from the user and the emotion data from the emotion engine.

[0349] The server dynamically adjusts the simulation to produce results that reflect real-world emotional states.

[0350] Step 11: Analyze the simulation results

[0351] The server collects and analyzes the simulation results, using statistical analysis and machine learning techniques to extract trends and patterns from the results.

[0352] The server summarizes the analysis results in an easy-to-understand format (e.g., graphs, charts, reports).

[0353] Step 12: View the results

[0354] The terminal visually displays the simulation results sent from the server, presenting the results using visual elements (e.g., graphs, charts, reports) so that the user can easily understand them.

[0355] The user checks the results displayed on the terminal and makes a decision.

[0356] Step 13: Gather feedback

[0357] The user provides feedback on the simulation results and the displayed content through the terminal.

[0358] The terminal sends the collected feedback to the server.

[0359] Step 14: Further simulation

[0360] The server simulates additional scenarios or retunes the model based on user feedback.

[0361] The user inputs a new scenario as needed, and the server executes a re-simulation based on that.

[0362] In this way, the systems work together to help users make better, more emotionally informed decisions.

[0363] Example 2

[0364] 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."

[0365] Currently, there are systems that utilize a large amount of data to provide highly accurate decision-making support, but few of them take into account the user's emotional state. As a result, conventional systems are unable to fully reflect the impact of the user's emotions on decision-making, making it difficult to make optimal decisions.

[0366] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0367] In this invention, the server includes means for collecting a huge amount of data, means for preprocessing and integrating the collected data, means for training a generative AI model using the preprocessed and integrated data, means for generating a virtual environment similar to the real world using the trained generative AI model, means for analyzing the collected emotional data and recognizing a user's emotional state in real time, means for dynamically adjusting a simulation scenario based on the emotional data, and means for analyzing and displaying the results of the simulation, thereby enabling highly accurate simulation and decision-making support that takes the user's emotional state into account.

[0368] "Big data" refers to a large and diverse collection of information, collected from sources such as corporate data, personal data, and public data.

[0369] "Preprocessing" refers to the process of carrying out procedures such as noise removal, missing value filling, and data standardization on collected data to make the data consistent.

[0370] "Synthesis" is the process of combining pre-processed data into one unified data set.

[0371] "Training a generative artificial intelligence model" is the process of applying machine learning algorithms to preprocessed and integrated data to train the model for optimal performance.

[0372] A "virtual environment" is an artificial environment that uses a generative artificial intelligence model to create a simulated environment similar to the real world.

[0373] "Emotional data" refers to information collected from the user's facial expressions, voice, input content, etc., that indicates the user's emotional state in real time.

[0374] "Analyzing emotional data" refers to the process of recognizing the user's emotional state based on the collected emotional data, and detecting classifications or specific emotional states as needed.

[0375] "Dynamic adjustment of the simulation scenario" refers to the process of appropriately changing the content of the scenario based on the user's emotional state during the simulation, in order to bring the results closer to reality.

[0376] "Displaying the simulation results" refers to the process of visually expressing the data obtained from the simulation and presenting it in a form that is easy for the user to understand.

[0377] MODE FOR CARRYING OUT THE INVENTION

[0378] The present invention is a system that collects massive amounts of data, integrates and preprocesses them, trains a generative artificial intelligence model, and runs simulations in a virtual environment using the generated model. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to support more accurate decision-making that takes into account the user's emotional state. This system is configured as follows:

[0379] 1. Data collection and integration

[0380] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data. Collection methods include APIs, database queries, and web scraping tools (e.g., Beautiful Soup and Scrapy). The collected data is preprocessed using the Python Pandas library, which removes noise, fills in missing values, and standardizes the data to ensure consistency.

[0381] 2. Training a generative AI model

[0382] The server trains a generative artificial intelligence model using the preprocessed and integrated data. It uses machine learning algorithms such as TensorFlow and PyTorch to split the dataset into training, validation, and testing sets. It also uses Optuna and Grid Search to tune the model's hyperparameters and optimize accuracy.

[0383] 3. Creating a virtual environment

[0384] The server uses the trained generative AI model to generate a virtual environment similar to the real world, using a virtual environment generation tool (e.g., Unity or Unreal Engine) to recreate the scenario required for the simulation.

[0385] 4. Enter the scenario

[0386] The user inputs a specific simulation scenario through the terminal. This input process involves using a web application form or an interactive chat window to enter details such as new product features, pricing, and carrier selection. The terminal then sends the input scenario data to the server via an HTTP request.

[0387] 5. Emotional Data Collection and Analysis Using an Emotional Engine

[0388] The device collects emotion data from the user's facial expressions, voice, and input. For example, it captures facial expressions with a camera, analyzes them using OpenCV, and converts the speech into text using a speech recognition engine (e.g., Google Speech-to-Text). The emotion engine analyzes the collected emotion data and uses a deep learning model to classify and recognize emotions into multiple categories, such as "happiness," "sadness," and "excitement."

[0389] 6. Running the Simulation

[0390] The server executes a simulation based on the received scenario and emotion data. The scenario is dynamically adjusted based on the emotion data, and the simulation is performed in real time within the virtual environment. For example, if a user wants to predict market reaction to a new product, they can input the characteristics and pricing of the new product, and if the emotion engine interprets this as "excitement," it will affect the simulation results.

[0391] 7. Analyzing and displaying simulation results

[0392] The server analyzes the simulation results and uses statistical analysis and machine learning techniques to compile the results into easy-to-understand formats, such as Pandas and Matplotlib, to convert the data into graphs and charts. The terminal visually displays these results and provides them to the user in an easy-to-understand format.

[0393] 8. Decision support and feedback gathering

[0394] The user makes optimal decisions based on the displayed simulation results. The server collects feedback from the user and uses it to conduct additional simulations and refine the model, thereby enabling more accurate decision-making support.

[0395] Specific examples

[0396] Case 1: A company decides to launch a new product

[0397] 1. The user (corporate decision maker) inputs scenarios such as new product characteristics and pricing into the terminal, and their emotional state (e.g., excitement) is collected at the same time.

[0398] 2. The emotion engine analyzes the user's emotional state in real time and sends it to the server.

[0399] 3. The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions taking into account emotional data.

[0400] 4. The terminal presents the analysis results to the user in the form of graphs and reports.

[0401] 5. The user decides on a new product launch strategy based on the simulation results.

[0402] Case 2: Individual career choices

[0403] 1. The user inputs a scenario about career choices (such as the school they will attend, the occupation they will choose, etc.) into the terminal, and their emotional state (e.g., anxiety, expectation) is collected at the same time.

[0404] 2. The emotion engine analyzes the user's emotional state and sends it to the server.

[0405] 3. The server simulates scenarios after career selection and analyzes future annual income and job satisfaction taking into account emotional data.

[0406] 4. The device presents the analysis results to the user as a report.

[0407] 5. The user makes a carrier selection based on the simulation results.

[0408] Prompt Sentence Examples

[0409] If you want to predict market reaction to a new product:

[0410] Predict the market reaction to a newly developed smartphone. The characteristics are 5G, a 6.5-inch display, a triple camera, and a price of $800. The emotional state is "Excited."

[0411] If you would like to discuss your career options:

[0412] I'm looking for help choosing a future career. My options are to go to college for engineering or to go to design school. My current emotional state is a mixture of anxiety and excitement.

[0413] In this way, the system of the present invention, which is combined with an emotion engine, provides highly accurate simulation and decision support that takes into account the user's emotional state, helping the user make optimal choices.

[0414] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0415] Step 1:

[0416] The server collects data from various sources, including corporate data, personal data, and public data. It uses APIs, database queries, and web scraping tools (such as Beautiful Soup and Scrapy) as input. It saves the collected data in JSON or CSV format as output. Specifically, it periodically accesses a specified API endpoint, retrieves the latest data, and saves it in a local database.

[0417] Step 2:

[0418] The server preprocesses and standardizes the collected data. It receives the data collected in the previous step as input. It denoises the data, fills missing values, and standardizes the data. For example, it uses the Python Pandas library to generate a data frame, removes inaccurate values, and fills missing values ​​with the median. As output, it obtains a preprocessed, clean dataset.

[0419] Step 3:

[0420] The server trains a generative AI model using the preprocessed data. It receives the preprocessed data as input and trains the model using a machine learning algorithm (e.g., TensorFlow or PyTorch). It splits the dataset into training, validation, and test datasets and feeds each dataset into the model. It uses Optuna or Grid Search to tune the hyperparameters. As output, it obtains a trained generative AI model.

[0421] Step 4:

[0422] The server generates a virtual environment using a trained generative AI model. It receives the trained model and scenario data as input. It uses a virtual environment generation tool (such as Unity or Unreal Engine) to create an environment similar to the real world. As output, it obtains a virtual environment for simulation.

[0423] Step 5:

[0424] The user inputs a specific simulation scenario through a terminal. For input, the scenario (such as the characteristics and pricing of a new product) is entered using a form or chat window in the web application. For output, the scenario data is sent to the server.

[0425] Step 6:

[0426] The device collects the user's emotional data. As input, the user's facial expressions and voice are captured using a camera and microphone. OpenCV and Google Speech-to-Text are used to analyze the emotional data. The analyzed emotional data is obtained as output.

[0427] Step 7:

[0428] The emotion engine analyzes collected emotion data in real time to recognize the user's emotional state. It receives data from the camera and microphone as input and applies emotion recognition algorithms. The output is classified emotion data (e.g., joy, sadness, excitement).

[0429] Step 8:

[0430] The server executes a simulation using the received scenario and emotion data. It receives the scenario data and emotion data as input. It performs a simulation in the virtual environment while dynamically adjusting the scenario based on the emotion data. It obtains the simulation results as output.

[0431] Step 9:

[0432] The server analyzes the simulation results and generates data for visual display. It receives the simulation results as input and organizes them into an easy-to-understand format using statistical analysis and machine learning techniques. The output is analysis data, graphs, and charts.

[0433] Step 10:

[0434] The terminal visually displays the analysis results and provides them to the user. It receives the analysis data as input. It uses a web interface to display the results in the form of interactive graphs and reports. It outputs the results in a format that is easy for the user to understand.

[0435] Step 11:

[0436] The user makes optimal decisions based on the displayed simulation results. The analysis results are used as input. The information necessary for decision-making is collected and reflected in actual actions. For example, decisions are made regarding new product launch strategies or career choices. The user's decision is obtained as output.

[0437] Step 12:

[0438] The server collects user feedback and uses it to refine the model and conduct additional simulations. It receives user feedback as input, updates the model based on it, and retrains it to improve its accuracy. The output is an improved generative AI model.

[0439] (Application example 2)

[0440] 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."

[0441] Conventional autonomous driving systems control operation based on external environmental data, but do not take into account the emotional state of passengers, which means they are unable to fully alleviate the anxiety and stress felt by passengers. Furthermore, due to the lack of technology to collect and analyze emotional data and dynamically adjust driving modes, there is an issue of not being able to improve passenger safety and comfort.

[0442] 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 means for collecting a huge amount of data, means for preprocessing and integrating the collected data, and means for training a generative artificial intelligence model using the preprocessed and integrated data. This enables the operation of an autonomous vehicle that improves safety and comfort by analyzing passenger emotional states in real time and reflecting this in driving control.

[0443] "Big data" refers to data collected in large quantities from various sources, including corporate data, personal data, and public data.

[0444] "Preprocessing" refers to the process of performing operations such as noise removal, missing value filling, and data standardization on collected data to improve consistency and quality.

[0445] "Integration" is the process of bringing together data collected from different data sources, making it consistent and easier to process and analyze.

[0446] A "generative artificial intelligence model" is a model trained using machine learning algorithms that is generated to perform specific tasks based on large amounts of data.

[0447] A "virtual environment" is a simulation environment that imitates the real world, and is an environment that reproduces various situations based on scenarios and conditions specified by the user.

[0448] "Simulation" is a technique for reproducing phenomena based on specific scenarios and conditions in a virtual environment and analyzing the results.

[0449] The "emotion engine" is a system that recognizes a user's emotional state in real time by collecting and analyzing emotional data from the user's facial expressions, voice, input content, etc.

[0450] "Dynamic adjustment" refers to the act of changing and optimizing system behavior and settings on the fly based on real-time information such as emotional data.

[0451] "Analysis" refers to the process of compiling collected data and simulation results into an easy-to-understand format and extracting the meaning and trends of the data using statistical analysis and machine learning techniques.

[0452] To specifically implement this invention, the system configuration and operating procedures are as follows: The system collects, preprocesses, and integrates massive amounts of data, trains a generative AI model based on the data, and uses the generated model to run a simulation in a virtual environment similar to the real world. Furthermore, the system analyzes the user's emotional state in real time and dynamically adjusts the simulation results to support more accurate decision-making.

[0453] Components and their operation

[0454] Data collection and preprocessing

[0455] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data, using techniques such as APIs, database queries, and web scraping. The collected data undergoes preprocessing, such as noise removal, missing value filling, and data standardization, to ensure consistency.

[0456] Training generative artificial intelligence models

[0457] The server trains generative AI models using the preprocessed and integrated data. Machine learning algorithms are implemented using Scikit-learn, TensorFlow, Keras, etc., and datasets are divided into training, validation, and testing phases. Model performance is also evaluated and hyperparameters are adjusted.

[0458] Creating a virtual environment and inputting a scenario

[0459] Using a trained generative artificial intelligence model, a virtual environment similar to the real world is generated. The user inputs the specific scenario they want to simulate through their device. The device then sends the input scenario to the server, which then starts the simulation in the virtual environment.

[0460] Analysis by emotion engine

[0461] The device collects emotional data from the user's facial expressions, voice, input, etc. The emotion engine, built using Keras and other tools, analyzes the collected emotional data, recognizes the user's emotional state in real time, and transmits the data to the server.

[0462] Running a simulation and displaying the results

[0463] The server runs a simulation in a virtual environment based on the received scenario and emotional data. It dynamically adjusts the simulation results taking into account the emotional data. The server analyzes the simulation results and visually displays them using statistical analysis and machine learning techniques. The terminal presents the analysis results to the user in the form of graphs and reports.

[0464] Specific examples

[0465] As a concrete example, consider a case where a user of an autonomous vehicle feels anxious during their morning commute due to a traffic jam. The emotion engine recognizes this anxiety and adjusts the autonomous vehicle to slow down a little and select a safe route.

[0466] Prompt Sentence Examples

[0467] An example of a prompt to enter an example into the system is:

[0468] "A user is using an autonomous vehicle during their morning commute. They encounter a traffic jam along the way, which makes them feel anxious. The emotion engine recognizes this feeling of anxiety, and the autonomous vehicle adjusts to slow down and choose a safer route."

[0469] A system configured in this way realizes highly accurate simulation and decision-making support that takes into account the user's emotional state.

[0470] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0471] Step 1: Data collection

[0472] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data, using techniques such as APIs, database queries, and web scraping. The input is data provided by each source, and the output is a huge amount of raw data.

[0473] Step 2: Data preprocessing and integration

[0474] The server performs preprocessing on the collected data, such as noise removal, missing value filling, and data standardization, to make it consistent. Specifically, it uses data cleaning algorithms, interpolates missing values, and standardizes the data. The input is raw data, and the output is preprocessed and integrated data.

[0475] Step 3: Training the generative artificial intelligence model

[0476] The server uses the preprocessed and integrated data to train a generative AI model. Specifically, it uses machine learning frameworks such as Scikit-learn, TensorFlow, and Keras to divide the dataset into training, validation, and testing sections, and selects an appropriate algorithm for training. The input is the preprocessed data, and the output is a trained AI model.

[0477] Step 4: Create a virtual environment

[0478] The server uses a trained artificial intelligence model to generate a virtual environment similar to the real world. Specifically, it uses 3D simulation software to build the environment based on the information the model has learned. The input is the trained model, and the output is the virtual environment.

[0479] Step 5: Entering the scenario

[0480] The user inputs the specific scenario they want to simulate through their device. The input scenario can include forecasts of market reactions to new products or carrier selection. The input is the scenario information provided by the user, and the output is the data sent from the device that receives it to the server.

[0481] Step 6: Collecting Emotion Data with the Emotion Engine

[0482] The device collects emotion data from the user's facial expressions, voice, input content, etc. The emotion engine analyzes emotions using models such as Keras. The input is the user's facial expressions and voice data, and the output is analyzed emotional state data.

[0483] Step 7: Run the simulation

[0484] The server executes a simulation in the virtual environment based on the received scenario and emotional data. Specifically, it dynamically adjusts simulation parameters based on the scenario and emotional data to generate results. The input is the user's scenario and emotional state data, and the output is the simulation results.

[0485] Step 8: Analyze and display simulation results

[0486] The server visually analyzes the simulation results and displays them in the form of graphs and reports. Specifically, it uses statistical analysis and machine learning techniques to summarize the results in an easy-to-understand format. The input is the simulation results, and the output is the analyzed results in graphs and reports.

[0487] Step 9: Support decision making and gather feedback

[0488] The user makes optimal decisions based on the displayed simulation results and sends feedback from the device to the server, which then performs further simulations and adjusts the model based on the feedback. The input is the user's feedback, and the output is the adjusted simulation results and model.

[0489] 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.

[0490] 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.

[0491] 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.

[0492] [Second embodiment]

[0493] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0494] 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.

[0495] 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).

[0496] 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.

[0497] 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.

[0498] 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).

[0499] 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.

[0500] 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.

[0501] 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.

[0502] 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.

[0503] 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.

[0504] 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."

[0505] This invention provides a system that uses generative artificial intelligence to build a virtual environment based on a huge amount of data and performs simulations within that virtual environment. This system aims to minimize risk by simulating real-world failures in the virtual environment in advance in order to improve the accuracy of decision-making.

[0506] System configuration and operation

[0507] 1. Data collection and integration

[0508] The servers collect vast amounts of data from various sources, including corporate, personal, and public data, using techniques such as APIs, database queries, and web scraping.

[0509] The server pre-processes the collected data, cleaning, normalizing, and consolidating it, transforming data from different sources into a unified format and making it consistent.

[0510] 2. Training a generative AI model

[0511] The server uses the preprocessed and integrated data to train a generative artificial intelligence model, selecting appropriate machine learning algorithms and training the model to learn patterns and trends in the data.

[0512] The training dataset is split into training, validation, and testing sets and is used to evaluate the performance of the model.

[0513] 3. Building a Virtual Environment

[0514] The server uses trained generative artificial intelligence models to generate virtual environments similar to the real world, allowing users to freely configure the scenarios they want to try out, recreating a wide range of scenarios and variables.

[0515] 4. Running the Simulation

[0516] The user inputs the specific scenario they want to simulate through the terminal, including specific settings such as new product features, pricing, and carrier selection.

[0517] The terminal transmits the user's input to the server, which runs the simulation within the virtual environment.

[0518] 5. Analyzing and displaying simulation results

[0519] The server analyzes the simulation results, finding trends and patterns in the data, and then uses specific statistical analysis and machine learning techniques to summarize the results in an easy-to-understand format.

[0520] The terminal visually displays the analysis results and provides them to the user, including graphs, charts, reports, etc.

[0521] 6. Decision support

[0522] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback and performs further simulations or re-adjusts the model as needed.

[0523] Specific examples

[0524] Case 1: A company decides to launch a new product

[0525] The user (corporate decision maker) inputs scenarios such as the characteristics and pricing of new products into the terminal.

[0526] The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions.

[0527] The terminal presents the analysis results to the user in the form of graphs and reports.

[0528] The user decides on a new product launch strategy based on the simulation results.

[0529] Case 2: Individual career choices

[0530] The user inputs a scenario regarding career choices (school choice, occupation choice, etc.) into the terminal.

[0531] The server simulates scenarios after career choices in a virtual environment and analyzes future annual income and job satisfaction.

[0532] The terminal presents the analysis results to the user as a report.

[0533] The user selects a carrier based on the simulation results.

[0534] In this way, the system of the present invention can support user decision-making and minimize risks through highly accurate simulations based on a variety of data.

[0535] The processing flow will be explained below.

[0536] Step 1: Data collection

[0537] The server retrieves data from a variety of sources, including corporate databases, public data services, and public data sources on the internet, using techniques such as APIs, database queries, and web scraping.

[0538] The server periodically collects data and keeps the necessary information up to date.

[0539] Step 2: Data Preprocessing

[0540] The server removes noise from the acquired data and fills in missing values. For example, missing data is interpolated using the average value or deleted.

[0541] The server performs data standardization and normalization, converting data from various formats into a unified format.

[0542] Step 3: Data Integration

[0543] The server consolidates the pre-processed data into one unified database.

[0544] The server maps and cross-references data schemas so that data from different sources is consistent.

[0545] Step 4: Prepare the dataset

[0546] The server extracts the training dataset from the integrated database and splits it into training, validation, and test datasets.

[0547] The server labels the data and performs feature engineering to prepare it in the optimal format for model training.

[0548] Step 5: Training the generative AI model

[0549] The server selects an appropriate generative AI algorithm and trains the model using a training dataset.

[0550] The server monitors the training process and tunes hyperparameters as needed.

[0551] Step 6: Evaluate the generative AI model

[0552] The server evaluates the model's performance using a validation dataset, using metrics such as precision, recall, and F1 score.

[0553] The server improves and retrains the model based on the evaluation results.

[0554] Step 7: Generate a Virtual Environment

[0555] The server uses a trained generative AI model to generate a virtual environment similar to the real world.

[0556] The server sets variables and parameters within the virtual environment to recreate realistic scenarios.

[0557] Step 8: Entering the Scenario

[0558] The user inputs the specific scenario he or she wants to simulate through the terminal.

[0559] The terminal collects the user's input and sends it to the server.

[0560] Step 9: Run the simulation

[0561] The server executes a simulation in a virtual environment based on the received specific scenario.

[0562] The server simulates multiple scenarios in parallel and collects data.

[0563] Step 10: Analyze the simulation results

[0564] The server analyzes the simulation results and finds trends and patterns in the data.

[0565] The server uses statistical analysis and machine learning techniques to compile the results into an easy-to-understand format.

[0566] Step 11: View the results

[0567] The terminal visually displays the simulation results sent from the server, including graphs, charts, and reports.

[0568] The user checks the displayed results and decides on the next action to take.

[0569] Step 12: Gather feedback

[0570] The user provides feedback on the simulation results through the terminal.

[0571] The terminal sends the collected feedback to the server.

[0572] Step 13: Further simulation

[0573] The server simulates additional scenarios or retunes the model based on user feedback.

[0574] The user inputs a new scenario as necessary and runs the simulation again.

[0575] In this way, the systems work together to help users make better decisions.

[0576] Example 1

[0577] 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."

[0578] Decisions in the real world involve many risks and uncertainties, so prior simulations are required to minimize risk. However, with conventional systems, the process from collecting and integrating massive amounts of data to training AI models, running simulations, and analyzing the results is often complex and inefficient, resulting in insufficient accuracy. It is also difficult for users to efficiently input specific scenarios and quickly make decisions based on the analysis results.

[0579] 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.

[0580] In this invention, the server includes means for collecting a huge amount of data, means for preprocessing and integrating the collected data, means for training a generative AI model using the preprocessed and integrated data, means for generating a virtual environment similar to the real world using the trained generative AI model, means for executing a simulation in the generated virtual environment, means for analyzing and displaying the results of the simulation, means for a user to input a specific scenario they wish to simulate, and means for automatically generating and analyzing multiple scenarios in the virtual environment. This enables users to make quick and accurate decisions through highly accurate simulations based on a huge amount of data.

[0581] "Big data" refers to data collected in large quantities from various sources, including corporate data, personal data, and public data.

[0582] "Preprocessing" refers to the process of cleaning (filling in missing values, removing outliers, etc.) and normalizing (formatting the data) the collected data.

[0583] "Integration" is the process of transforming data collected from different sources into a consistent format so that it can be treated as a single data set.

[0584] A "generative artificial intelligence model" is a model trained using machine learning algorithms to learn patterns and trends in data and perform simulations and predictions.

[0585] "Training" refers to the process of training a generative artificial intelligence model using preprocessed and integrated data.

[0586] "Virtual environment" refers to a simulated space that resembles the real world and is generated using a trained generative artificial intelligence model.

[0587] "Simulation" refers to the process of conducting experiments and verifications in a virtual environment based on scenarios and conditions specified by the user.

[0588] "Analysis" refers to the process of finding trends and patterns in the data from the simulation results, specifically using statistical analysis and machine learning techniques.

[0589] "Display" refers to providing the analyzed simulation results to the user in a visual format (graphs, charts, reports, etc.).

[0590] A "specific scenario" refers to the specific conditions or settings that the user wants to simulate (e.g., new product characteristics, pricing, carrier selection, etc.).

[0591] "Automatically generating and analyzing multiple scenarios" refers to the process of automatically generating multiple scenarios within a virtual environment based on user input and analyzing the results of each scenario.

[0592] MODE FOR CARRYING OUT THE INVENTION

[0593] This invention relates to a system that uses a generative artificial intelligence model based on a huge amount of data to create a virtual environment and perform simulations within that virtual environment. The system aims to minimize risk by simulating real-world failures in the virtual environment in advance in order to improve the accuracy of user decision-making.

[0594] System configuration

[0595] Hardware and Software Configuration

[0596] server:

[0597] Data collection is done using APIs (e.g., Twitter API, Google Analytics API), databases (e.g., MySQL, PostgreSQL), and web scraping tools (e.g., BeautifulSoup, Scrapy).

[0598] Use data cleansing tools (e.g., Pandas) for data preprocessing and integration.

[0599] Use machine learning libraries (e.g., TensorFlow, PyTorch) to train generative artificial intelligence models.

[0600] Statistical analysis tools (e.g., R, SciPy) and data analysis platforms (e.g., Jupyter Notebook, MATLAB) will be used for simulation and result analysis.

[0601] Device:

[0602] Use visualization tools (e.g., Matplotlib, Tableau) to display the results.

[0603] Data collection and preprocessing

[0604] The servers collect vast amounts of data, including corporate, personal, and public data, from various sources, including APIs, databases, and web scraping.

[0605] Example: A server retrieves the latest tweets from the Twitter API at regular intervals and stores them in a MySQL database.

[0606] The server cleans, normalizes, and consolidates the collected data, transforming it from different sources into a consistent format.

[0607] For example, the server uses Pandas to impute missing values, remove outliers, and convert each data set into a standard format.

[0608] Training generative artificial intelligence models

[0609] Based on the preprocessed data, the server selects an appropriate machine learning algorithm and trains a generative artificial intelligence model.

[0610] Example: The server uses TensorFlow to split the dataset into training, validation, and test data, and train and evaluate the model.

[0611] Building a virtual environment

[0612] The server uses a trained generative AI model to generate a virtual environment similar to the real world, which can be flexibly changed depending on the scenario or conditions the user wants to try.

[0613] Example: The server uses a generative AI model to generate various market scenarios based on the price range and market conditions set by the user.

[0614] Running and analyzing the simulation

[0615] Users input the specific scenario they want to simulate through the device, including the features and pricing of new products, carrier selection, and so on.

[0616] Example: A user enters prices and feature settings into the device interface and sends them to the server.

[0617] The server receives the scenario sent by the user and executes the simulation in the virtual environment.

[0618] Example: The server uses a generative AI model to run parallel simulations of scenarios based on input conditions.

[0619] The server analyzes the simulation results and finds trends and patterns in the data, using statistical analysis and machine learning techniques.

[0620] Example: The server uses SciPy to perform statistical analysis of the simulation results and graph sales forecasts and market reactions.

[0621] Results presentation and decision support

[0622] The terminal presents the analysis results to the user in a visual format, for example in the form of graphs, charts, or reports.

[0623] Example: The terminal receives the analysis results from the server and displays the results as a graph using Matplotlib.

[0624] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback from the user and performs further simulations or readjusts the model as needed.

[0625] Example: The user decides on a new product launch strategy based on the results and sends that feedback to the server, which uses the feedback to recalibrate the model.

[0626] Examples of concrete examples and prompts

[0627] Case 1: A company decides to launch a new product

[0628] Example of prompt text entered by the user (corporate decision maker):

[0629] Please define the following characteristics for your new product:

[0630] Function A

[0631] Price B

[0632] Target Market C

[0633] Simulate market reactions based on these conditions and analyze projected sales and market share.

[0634] The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions.

[0635] The terminal presents the analysis results to the user in the form of graphs and reports.

[0636] The user decides on a new product launch strategy based on the simulation results.

[0637] Case 2: Individual career choices

[0638] An example of a prompt that the user might enter:

[0639] "Enter the following career choice scenario:

[0640] Occupation X

[0641] University Y

[0642] Based on these, please analyze your future annual income and job satisfaction.

[0643] The server simulates scenarios after career choices in a virtual environment and analyzes future annual income and job satisfaction.

[0644] The terminal presents the analysis results to the user as a report.

[0645] The user makes a carrier selection based on the simulation results.

[0646] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0647] Step 1: Data collection

[0648] The server uses APIs (e.g., Twitter API, Google Analytics API), databases (e.g., MySQL, PostgreSQL), and web scraping tools (e.g., BeautifulSoup, Scrapy) to collect huge amounts of data, including corporate data, personal data, and public data.

[0649] Input: API endpoint, database query, website URL.

[0650] What it does: The server periodically fetches data from the API and stores it in a MySQL database. It also uses a web scraping tool to crawl websites and retrieve the required data.

[0651] Output: The raw data collected.

[0652] Step 2: Data Preprocessing

[0653] The server pre-processes the collected data, cleaning, normalizing, and integrating it, transforming data from different sources into a consistent format.

[0654] Input: Raw data collected.

[0655] Specific operation: The server uses Pandas to impute missing values, remove outliers, and convert each data into a standard format.

[0656] Output: The preprocessed dataset.

[0657] Step 3: Training the generative artificial intelligence model

[0658] Based on the preprocessed data, the server selects an appropriate machine learning algorithm and trains a generative artificial intelligence model.

[0659] Input: The preprocessed dataset.

[0660] Specific operation: The server uses TensorFlow to split the dataset into training data, validation data, and test data, and trains and evaluates the model.

[0661] Output: A trained generative AI model.

[0662] Step 4: Build a virtual environment

[0663] The server uses a trained generative AI model to generate a virtual environment similar to the real world, which can be flexibly changed depending on the scenario or conditions the user wants to try.

[0664] Input: A trained generative AI model.

[0665] Specific operation: The server uses the generative AI model to set various parameters within the virtual environment and provide scenario templates that can be customized by the user.

[0666] Output: A customizable virtual environment.

[0667] Step 5: Enter the simulation scenario

[0668] Users input the specific scenario they want to simulate through the device, including new product features, pricing, carrier selection, etc.

[0669] Input: Scenario conditions set by the user (e.g., new product characteristics, pricing).

[0670] Specific operation: The user inputs the required information into the terminal interface and sends it to the server.

[0671] Output: User's scenario configuration data.

[0672] Step 6: Run the simulation

[0673] The server receives the scenario sent by the user and executes the simulation in the virtual environment. The simulation automatically generates and analyzes multiple scenarios based on the specified conditions.

[0674] Input: User-submitted scenario configuration data, customizable virtual environment.

[0675] Specific operation: The server uses the generative AI model to simulate multiple scenarios in parallel and record the results.

[0676] Output: Simulation result data.

[0677] Step 7: Analyze the simulation results

[0678] The server analyzes the simulation results and finds trends and patterns in the data, using statistical analysis and machine learning techniques.

[0679] Input: Simulation result data.

[0680] How it works: The server uses SciPy to statistically analyze the data and visualize the results, such as sales forecasts and market reactions.

[0681] Output: Analyzed simulation results.

[0682] Step 8: View the results

[0683] The terminal visually displays the analysis results and provides them to the user, specifically in the form of graphs, charts, and reports.

[0684] Input: Analyzed simulation results.

[0685] Specific operation: The terminal receives the analysis results sent from the server and visually displays the results using Matplotlib.

[0686] Output: The visualization data that is presented to the user.

[0687] Step 9: Decision making and feedback

[0688] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback from the user and performs further simulations or readjusts the model as needed.

[0689] Input: User decision results, feedback data.

[0690] How it works: The user decides on a strategy based on the results and enters it into the device. The server receives the feedback and retrains the model as needed.

[0691] Output: Retuned generative AI model, further simulation results.

[0692] (Application example 1)

[0693] 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."

[0694] In content distribution services, accurately predicting user viewing trends and engagement and determining optimal distribution schedules based on the results is a very difficult challenge. Conventional methods only allow for limited predictions based on past data and experience, and are unable to effectively capture viewer reactions. Therefore, new technologies are needed to optimize content distribution schedules and improve engagement.

[0695] 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.

[0696] In this invention, the server includes means for collecting a huge amount of data, means for preprocessing and integrating the collected data, means for training a generative artificial intelligence model using the preprocessed and integrated data, means for generating a virtual environment similar to the real world using the trained generative artificial intelligence model, means for running a simulation in the generated virtual environment, means for analyzing and displaying the results of the simulation, means for predicting viewer responses based on the collected data, means for predicting an engagement score for adjusting a content delivery schedule, and means for visualizing the simulation results and providing them to a user. This makes it possible to predict viewer responses with high accuracy and to formulate an optimal delivery schedule based on the predictions.

[0697] "Big data" refers to large amounts of digital information collected from a wide variety of sources.

[0698] "Preprocessing" refers to a series of steps that transform collected data into a form that is applicable for analysis and modeling.

[0699] "Integration" refers to the act of bringing together data collected from different sources into a consistent format.

[0700] A "generative artificial intelligence model" refers to a machine learning model trained on collected and preprocessed data.

[0701] A "virtual environment" refers to a digital space created to simulate real-world actions or events.

[0702] "Simulation" refers to the process of experimenting with behaviors and outcomes under specific conditions within a virtual environment.

[0703] "Analysis" refers to the process of summarizing simulation results and data trends in an easy-to-understand format.

[0704] "Display" refers to the act of visually presenting the analysis results to the user.

[0705] "Viewer response" refers to the viewer's behavior and reaction to content.

[0706] "Engagement score" refers to a numerical indicator of the level of engagement that viewers show with content.

[0707] "Content delivery schedule" refers to a plan that determines when particular content should be delivered.

[0708] A "scenario" refers to the specific conditions and settings that a user inputs to perform a simulation.

[0709] "Optimization" refers to adjusting a system or process to its best state according to a specific goal.

[0710] The system of the present invention provides a means to collect, preprocess, and integrate massive amounts of data, train generative artificial intelligence models to generate virtual environments that resemble the real world, and run simulations within those virtual environments to determine optimal content delivery schedules.

[0711] The system includes a server, a terminal, and a user.

[0712] Data collection and preprocessing

[0713] The server collects a huge amount of data using techniques such as APIs, database queries, and web scraping. The collected data includes information such as viewer viewing history, viewing times, and behavioral patterns. The server then cleans and normalizes the data, integrating it for consistency.

[0714] Training generative artificial intelligence models

[0715] Using the preprocessed and integrated data, the server trains a generative artificial intelligence model. The training dataset is divided into training, validation, and testing datasets, and the target machine learning algorithm predicts viewer viewing habits. The model is designed to predict viewer behavior and reactions, and the algorithm used for training can be a linear regression model.

[0716] Virtual environment generation and simulation

[0717] Using a trained generative artificial intelligence model, the server generates a virtual environment similar to the real world. Users can input specific scenarios via their devices, specifying specific dates and times, the number of views, likes, shares, etc. The server then runs a simulation based on these scenarios and predicts viewer reactions.

[0718] Analyzing and displaying results

[0719] The results of the simulation are analyzed by the server. The analysis results include viewer reactions and engagement scores. The device visualizes these results and provides them to the user. The user can view the results in the form of graphs, charts, and reports.

[0720] Example

[0721] For example, if a user wants to stream at 8pm with the hope of getting 5000 views, 300 likes, and 50 shares, they can enter a specific scenario. This scenario can be expressed with a prompt like this:

[0722] Based on your viewing data, predict your viewer engagement score for the following stream schedule:

[0723] Views: 5000

[0724] Likes: 300

[0725] Shares: 50

[0726] Time: 8 PM

[0727] Based on this prompt, the server runs a simulation in the virtual environment to predict the viewer's reaction. The prediction results are provided to the user via their device, and the user can then decide on the optimal distribution schedule.

[0728] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0729] Step 1:

[0730] The server collects a huge amount of data. The types of data include viewer viewing history, viewing times, and behavioral patterns. This data is obtained using APIs, database queries, and web scraping techniques. Specifically, data is obtained by issuing queries to a viewing history database. The input in this step is the viewer data source, and the output is raw viewer data.

[0731] Step 2:

[0732] The server preprocesses and integrates the collected data. It removes noise from the data, imputes missing values, and converts various data sources into a consistent format. This step includes cleaning and normalizing the data. Specifically, it uses data cleaning tools to handle missing values. The input to this step is the collected data from step 1, and the output is the preprocessed and integrated data.

[0733] Step 3:

[0734] The server trains a generative artificial intelligence model using the preprocessed and integrated data. Specifically, it splits the dataset into training, validation, and test sets, and trains the model using a machine learning algorithm such as a linear regression model. The input in this step is the preprocessed data from step 2, and the output is a trained AI model.

[0735] Step 4:

[0736] The server generates a virtual environment using a trained generative AI model. The virtual environment is a digital space for simulating viewer behavior. Various scenarios and variables can be set, enabling various simulations. The input in this step is the trained AI model, and the output is the virtual environment.

[0737] Step 5:

[0738] Through the terminal, the user inputs the specific scenario they want to simulate. The scenario includes parameters such as a specific date and time, expected number of viewers, number of likes, number of shares, etc. The input in this step is the user's scenario information, and the output is the scenario data sent to the server.

[0739] Step 6:

[0740] The server runs a simulation in the virtual environment based on the input scenario. Specifically, it uses the model to generate data to predict viewer reactions and then runs the simulation. The input in this step is the scenario data from step 5, and the output is the simulation results.

[0741] Step 7:

[0742] The server analyzes the simulation results. The analysis aims to predict viewer responses and engagement scores. This procedure includes analyzing data trends using statistical analysis and machine learning techniques. The input in this step is the simulation results, and the output is the analysis results.

[0743] Step 8:

[0744] The device visualizes the analysis results and provides them to the user. Specifically, the analysis results are displayed in the form of graphs, charts, and reports. The user views these results and determines the optimal content distribution schedule based on the engagement scores. The input in this step is the analysis results, and the output is the visualized analysis results provided to the user.

[0745] By following the steps above, it becomes possible to predict viewer reactions with high accuracy and formulate an optimal distribution schedule.

[0746] 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.

[0747] The present invention is a system that collects massive amounts of data, integrates and preprocesses them, trains a generative artificial intelligence model, and runs simulations in a virtual environment using the generated model. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to support more accurate decision-making that takes into account the user's emotional state. The present invention is implemented using the following processing steps and configuration.

[0748] System configuration and operation

[0749] 1. Data collection and integration

[0750] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data, and uses techniques such as APIs, database queries, and web scraping to efficiently retrieve the data.

[0751] The server performs preprocessing on the collected data, such as noise removal, missing value filling, and data standardization, to ensure consistency.

[0752] 2. Training a generative AI model

[0753] The server uses the preprocessed and integrated data to train a generative artificial intelligence model, selecting an appropriate machine learning algorithm and splitting the dataset into training, validation, and testing sections.

[0754] Evaluate the performance of the model and adjust the hyperparameters as necessary.

[0755] 3. Creating a virtual environment

[0756] The server uses a trained generative artificial intelligence model to generate a virtual environment similar to the real world, with the flexibility to recreate a variety of scenarios.

[0757] 4. Enter the scenario

[0758] The user inputs the specific scenario they want to simulate through the terminal, including, for example, the characteristics and pricing of a new product, carrier selection, etc.

[0759] The terminal transmits the input scenario to the server.

[0760] 5. Collecting Emotion Data Using an Emotion Engine

[0761] The device collects emotional data from the user's facial expressions, voice, input, etc.

[0762] The emotion engine analyzes the collected emotion data and recognizes the user's emotional state in real time.

[0763] 6. Running the Simulation

[0764] The server runs a simulation in the virtual environment based on the received scenario and emotion data, dynamically adjusting the scenario based on the emotion data to provide more realistic results.

[0765] 7. Analyzing and displaying simulation results

[0766] The server analyzes the simulation results and uses statistical analysis and machine learning techniques to summarize the results in an easy-to-understand format.

[0767] The terminal provides the user with a visual display of the analysis results, including graphs, charts, and reports.

[0768] 8. Decision support and feedback gathering

[0769] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback from the user and performs additional simulations or refines the model.

[0770] Specific examples

[0771] Case 1: A company decides to launch a new product

[0772] Users (corporate decision makers) input scenarios such as new product characteristics and pricing into the terminal, and their emotional state (e.g., excitement, anxiety, etc.) is collected at the same time.

[0773] The emotion engine analyzes the user's emotional state in real time and transmits it to the server.

[0774] The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions taking into account emotional data.

[0775] The terminal presents the analysis results to the user in the form of graphs and reports.

[0776] The user decides on a new product launch strategy based on the simulation results.

[0777] Case 2: Individual career choices

[0778] The user inputs a scenario regarding career choices (such as the school to attend, the occupation to choose, etc.) into the terminal, and their emotional state (such as anxiety, expectations, etc.) is collected at the same time.

[0779] The emotion engine analyzes the user's emotional state and transmits it to the server.

[0780] The server simulates scenarios after career selection and analyzes future annual income and job satisfaction taking into account emotional data.

[0781] The terminal presents the analysis results to the user as a report.

[0782] The user selects a carrier based on the simulation results.

[0783] In this way, the system of the present invention, which is combined with an emotion engine, provides highly accurate simulation and decision support that takes into account the user's emotional state, helping the user make optimal choices.

[0784] The processing flow will be explained below.

[0785] Step 1: Data collection

[0786] The server retrieves data from corporate databases, public data services, and public data sources on the Internet, using techniques such as APIs, database queries, and web scraping to efficiently gather data.

[0787] The server periodically updates the data to keep it up to date.

[0788] Step 2: Data Preprocessing

[0789] The server removes noise from the acquired data and fills in missing values. Specifically, it interpolates the average value of missing data and removes outliers.

[0790] The server performs data standardization (eg, normalization, scaling) and converts data from different sources into a consistent format.

[0791] Step 3: Data Integration

[0792] The server consolidates the pre-processed data into one unified database.

[0793] The server maps the schema of the data from different sources and makes it consistent.

[0794] Step 4: Prepare the dataset

[0795] The server extracts training, validation, and test datasets from the integrated database.

[0796] The server labels the data and performs feature engineering to prepare it in a format that is easy for the model to learn.

[0797] Step 5: Training the generative AI model

[0798] The server selects an appropriate generative AI algorithm (e.g., a deep learning model) and trains the model using a training dataset.

[0799] The server monitors the training process and optimizes hyperparameters as needed.

[0800] Step 6: Evaluate the generative AI model

[0801] The server evaluates the model's performance using a validation dataset, using metrics such as precision, recall, and F1 score.

[0802] The server will make any necessary improvements based on the evaluation results.

[0803] Step 7: Generate a Virtual Environment

[0804] The server uses a trained generative AI model to generate a virtual environment that resembles the real world.

[0805] The server sets the variables and parameters to be simulated within the virtual environment.

[0806] Step 8: Entering the Scenario

[0807] The user inputs the specific scenario he or she wants to simulate (for example, the characteristics and pricing of a new product, carrier selection, etc.) through the terminal.

[0808] The terminal transmits the user's input to the server.

[0809] Step 9: Collect emotion data

[0810] The device collects emotional data from the user's facial expressions, voice, input, etc. This is typically done using devices such as a camera and microphone.

[0811] The emotion engine analyzes the collected data and recognizes the user's emotional state in real time.

[0812] Step 10: Run the simulation

[0813] The server executes a simulation in the virtual environment based on the scenario received from the user and the emotion data from the emotion engine.

[0814] The server dynamically adjusts the simulation to produce results that reflect real-world emotional states.

[0815] Step 11: Analyze the simulation results

[0816] The server collects and analyzes the simulation results, using statistical analysis and machine learning techniques to extract trends and patterns from the results.

[0817] The server summarizes the analysis results in an easy-to-understand format (e.g., graphs, charts, reports).

[0818] Step 12: View the results

[0819] The terminal visually displays the simulation results sent from the server, presenting the results using visual elements (e.g., graphs, charts, reports) so that the user can easily understand them.

[0820] The user checks the results displayed on the terminal and makes a decision.

[0821] Step 13: Gather feedback

[0822] The user provides feedback on the simulation results and the displayed content through the terminal.

[0823] The terminal sends the collected feedback to the server.

[0824] Step 14: Further simulation

[0825] The server simulates additional scenarios or retunes the model based on user feedback.

[0826] The user inputs a new scenario as needed, and the server executes a re-simulation based on that.

[0827] In this way, the systems work together to help users make better, more emotionally informed decisions.

[0828] Example 2

[0829] 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."

[0830] Currently, there are systems that utilize a large amount of data to provide highly accurate decision-making support, but few of them take into account the user's emotional state. As a result, conventional systems are unable to fully reflect the impact of the user's emotions on decision-making, making it difficult to make optimal decisions.

[0831] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[0832] In this invention, the server includes means for collecting a huge amount of data, means for preprocessing and integrating the collected data, means for training a generative AI model using the preprocessed and integrated data, means for generating a virtual environment similar to the real world using the trained generative AI model, means for analyzing the collected emotional data and recognizing a user's emotional state in real time, means for dynamically adjusting a simulation scenario based on the emotional data, and means for analyzing and displaying the results of the simulation, thereby enabling highly accurate simulation and decision-making support that takes the user's emotional state into account.

[0833] "Big data" refers to a large and diverse collection of information, collected from sources such as corporate data, personal data, and public data.

[0834] "Preprocessing" refers to the process of carrying out procedures such as noise removal, missing value filling, and data standardization on collected data to make the data consistent.

[0835] "Synthesis" is the process of combining pre-processed data into one unified data set.

[0836] "Training a generative artificial intelligence model" is the process of applying machine learning algorithms to preprocessed and integrated data to train the model for optimal performance.

[0837] A "virtual environment" is an artificial environment that uses a generative artificial intelligence model to create a simulated environment similar to the real world.

[0838] "Emotional data" refers to information collected from the user's facial expressions, voice, input content, etc., that indicates the user's emotional state in real time.

[0839] "Analyzing emotional data" refers to the process of recognizing the user's emotional state based on the collected emotional data, and detecting classifications or specific emotional states as needed.

[0840] "Dynamic adjustment of the simulation scenario" refers to the process of appropriately changing the content of the scenario based on the user's emotional state during the simulation, in order to bring the results closer to reality.

[0841] "Displaying the simulation results" refers to the process of visually expressing the data obtained from the simulation and presenting it in a form that is easy for the user to understand.

[0842] MODE FOR CARRYING OUT THE INVENTION

[0843] The present invention is a system that collects massive amounts of data, integrates and preprocesses them, trains a generative artificial intelligence model, and runs simulations in a virtual environment using the generated model. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to support more accurate decision-making that takes into account the user's emotional state. This system is configured as follows:

[0844] 1. Data collection and integration

[0845] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data. Collection methods include APIs, database queries, and web scraping tools (e.g., Beautiful Soup and Scrapy). The collected data is preprocessed using the Python Pandas library, which removes noise, fills in missing values, and standardizes the data to ensure consistency.

[0846] 2. Training a generative AI model

[0847] The server trains a generative artificial intelligence model using the preprocessed and integrated data. It uses machine learning algorithms such as TensorFlow and PyTorch to split the dataset into training, validation, and testing sets. It also uses Optuna and Grid Search to tune the model's hyperparameters and optimize accuracy.

[0848] 3. Creating a virtual environment

[0849] The server uses the trained generative AI model to generate a virtual environment similar to the real world, using a virtual environment generation tool (e.g., Unity or Unreal Engine) to recreate the scenario required for the simulation.

[0850] 4. Enter the scenario

[0851] The user inputs a specific simulation scenario through the terminal. This input process involves using a web application form or an interactive chat window to enter details such as new product features, pricing, and carrier selection. The terminal then sends the input scenario data to the server via an HTTP request.

[0852] 5. Emotional Data Collection and Analysis Using an Emotional Engine

[0853] The device collects emotion data from the user's facial expressions, voice, and input. For example, it captures facial expressions with a camera, analyzes them using OpenCV, and converts the speech into text using a speech recognition engine (e.g., Google Speech-to-Text). The emotion engine analyzes the collected emotion data and uses a deep learning model to classify and recognize emotions into multiple categories, such as "happiness," "sadness," and "excitement."

[0854] 6. Running the Simulation

[0855] The server executes a simulation based on the received scenario and emotion data. The scenario is dynamically adjusted based on the emotion data, and the simulation is performed in real time within the virtual environment. For example, if a user wants to predict market reaction to a new product, they can input the characteristics and pricing of the new product, and if the emotion engine interprets this as "excitement," it will affect the simulation results.

[0856] 7. Analyzing and displaying simulation results

[0857] The server analyzes the simulation results and uses statistical analysis and machine learning techniques to compile the results into easy-to-understand formats, such as Pandas and Matplotlib, to convert the data into graphs and charts. The terminal visually displays these results and provides them to the user in an easy-to-understand format.

[0858] 8. Decision support and feedback gathering

[0859] The user makes optimal decisions based on the displayed simulation results. The server collects feedback from the user and uses it to conduct additional simulations and refine the model, thereby enabling more accurate decision-making support.

[0860] Specific examples

[0861] Case 1: A company decides to launch a new product

[0862] 1. The user (corporate decision maker) inputs scenarios such as new product characteristics and pricing into the terminal, and their emotional state (e.g., excitement) is collected at the same time.

[0863] 2. The emotion engine analyzes the user's emotional state in real time and sends it to the server.

[0864] 3. The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions taking into account emotional data.

[0865] 4. The terminal presents the analysis results to the user in the form of graphs and reports.

[0866] 5. The user decides on a new product launch strategy based on the simulation results.

[0867] Case 2: Individual career choices

[0868] 1. The user inputs a scenario about career choices (such as the school they will attend, the occupation they will choose, etc.) into the terminal, and their emotional state (e.g., anxiety, expectation) is collected at the same time.

[0869] 2. The emotion engine analyzes the user's emotional state and sends it to the server.

[0870] 3. The server simulates scenarios after career selection and analyzes future annual income and job satisfaction taking into account emotional data.

[0871] 4. The device presents the analysis results to the user as a report.

[0872] 5. The user makes a carrier selection based on the simulation results.

[0873] Prompt Sentence Examples

[0874] If you want to predict market reaction to a new product:

[0875] Predict the market reaction to a newly developed smartphone. The characteristics are 5G, a 6.5-inch display, a triple camera, and a price of $800. The emotional state is "Excited."

[0876] If you would like to discuss your career options:

[0877] I'm looking for help choosing a future career. My options are to go to college for engineering or to go to design school. My current emotional state is a mixture of anxiety and excitement.

[0878] In this way, the system of the present invention, which is combined with an emotion engine, provides highly accurate simulation and decision support that takes into account the user's emotional state, helping the user make optimal choices.

[0879] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0880] Step 1:

[0881] The server collects data from various sources, including corporate data, personal data, and public data. It uses APIs, database queries, and web scraping tools (such as Beautiful Soup and Scrapy) as input. It saves the collected data in JSON or CSV format as output. Specifically, it periodically accesses a specified API endpoint, retrieves the latest data, and saves it in a local database.

[0882] Step 2:

[0883] The server preprocesses and standardizes the collected data. It receives the data collected in the previous step as input. It denoises the data, fills missing values, and standardizes the data. For example, it uses the Python Pandas library to generate a data frame, removes inaccurate values, and fills missing values ​​with the median. As output, it obtains a preprocessed, clean dataset.

[0884] Step 3:

[0885] The server trains a generative AI model using the preprocessed data. It receives the preprocessed data as input and trains the model using a machine learning algorithm (e.g., TensorFlow or PyTorch). It splits the dataset into training, validation, and test datasets and feeds each dataset into the model. It uses Optuna or Grid Search to tune the hyperparameters. As output, it obtains a trained generative AI model.

[0886] Step 4:

[0887] The server generates a virtual environment using a trained generative AI model. It receives the trained model and scenario data as input. It uses a virtual environment generation tool (such as Unity or Unreal Engine) to create an environment similar to the real world. As output, it obtains a virtual environment for simulation.

[0888] Step 5:

[0889] The user inputs a specific simulation scenario through a terminal. For input, the scenario (such as the characteristics and pricing of a new product) is entered using a form or chat window in the web application. For output, the scenario data is sent to the server.

[0890] Step 6:

[0891] The device collects the user's emotional data. As input, the user's facial expressions and voice are captured using a camera and microphone. OpenCV and Google Speech-to-Text are used to analyze the emotional data. The analyzed emotional data is obtained as output.

[0892] Step 7:

[0893] The emotion engine analyzes collected emotion data in real time to recognize the user's emotional state. It receives data from the camera and microphone as input and applies emotion recognition algorithms. The output is classified emotion data (e.g., joy, sadness, excitement).

[0894] Step 8:

[0895] The server executes a simulation using the received scenario and emotion data. It receives the scenario data and emotion data as input. It performs a simulation in the virtual environment while dynamically adjusting the scenario based on the emotion data. It obtains the simulation results as output.

[0896] Step 9:

[0897] The server analyzes the simulation results and generates data for visual display. It receives the simulation results as input and organizes them into an easy-to-understand format using statistical analysis and machine learning techniques. The output is analysis data, graphs, and charts.

[0898] Step 10:

[0899] The terminal visually displays the analysis results and provides them to the user. It receives the analysis data as input. It uses a web interface to display the results in the form of interactive graphs and reports. It outputs the results in a format that is easy for the user to understand.

[0900] Step 11:

[0901] The user makes optimal decisions based on the displayed simulation results. The analysis results are used as input. The information necessary for decision-making is collected and reflected in actual actions. For example, decisions are made regarding new product launch strategies or career choices. The user's decision is obtained as output.

[0902] Step 12:

[0903] The server collects user feedback and uses it to refine the model and conduct additional simulations. It receives user feedback as input, updates the model based on it, and retrains it to improve its accuracy. The output is an improved generative AI model.

[0904] (Application example 2)

[0905] 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."

[0906] Conventional autonomous driving systems control operation based on external environmental data, but do not take into account the emotional state of passengers, which means they are unable to fully alleviate the anxiety and stress felt by passengers. Furthermore, due to the lack of technology to collect and analyze emotional data and dynamically adjust driving modes, there is an issue of not being able to improve passenger safety and comfort.

[0907] 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 means for collecting a huge amount of data, means for preprocessing and integrating the collected data, and means for training a generative artificial intelligence model using the preprocessed and integrated data. This enables the operation of an autonomous vehicle that improves safety and comfort by analyzing passenger emotional states in real time and reflecting this in driving control.

[0908] "Big data" refers to data collected in large quantities from various sources, including corporate data, personal data, and public data.

[0909] "Preprocessing" refers to the process of performing operations such as noise removal, missing value filling, and data standardization on collected data to improve consistency and quality.

[0910] "Integration" is the process of bringing together data collected from different data sources, making it consistent and easier to process and analyze.

[0911] A "generative artificial intelligence model" is a model trained using machine learning algorithms that is generated to perform specific tasks based on large amounts of data.

[0912] A "virtual environment" is a simulation environment that imitates the real world, and is an environment that reproduces various situations based on scenarios and conditions specified by the user.

[0913] "Simulation" is a technique for reproducing phenomena based on specific scenarios and conditions in a virtual environment and analyzing the results.

[0914] The "emotion engine" is a system that recognizes a user's emotional state in real time by collecting and analyzing emotional data from the user's facial expressions, voice, input content, etc.

[0915] "Dynamic adjustment" refers to the act of changing and optimizing system behavior and settings on the fly based on real-time information such as emotional data.

[0916] "Analysis" refers to the process of compiling collected data and simulation results into an easy-to-understand format and extracting the meaning and trends of the data using statistical analysis and machine learning techniques.

[0917] To specifically implement this invention, the system configuration and operating procedures are as follows: The system collects, preprocesses, and integrates massive amounts of data, trains a generative AI model based on the data, and uses the generated model to run a simulation in a virtual environment similar to the real world. Furthermore, the system analyzes the user's emotional state in real time and dynamically adjusts the simulation results to support more accurate decision-making.

[0918] Components and their operation

[0919] Data collection and preprocessing

[0920] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data, using techniques such as APIs, database queries, and web scraping. The collected data undergoes preprocessing, such as noise removal, missing value filling, and data standardization, to ensure consistency.

[0921] Training generative artificial intelligence models

[0922] The server trains generative AI models using the preprocessed and integrated data. Machine learning algorithms are implemented using Scikit-learn, TensorFlow, Keras, etc., and datasets are divided into training, validation, and testing phases. Model performance is also evaluated and hyperparameters are adjusted.

[0923] Creating a virtual environment and inputting a scenario

[0924] Using a trained generative artificial intelligence model, a virtual environment similar to the real world is generated. The user inputs the specific scenario they want to simulate through their device. The device then sends the input scenario to the server, which then starts the simulation in the virtual environment.

[0925] Analysis by emotion engine

[0926] The device collects emotional data from the user's facial expressions, voice, input, etc. The emotion engine, built using Keras and other tools, analyzes the collected emotional data, recognizes the user's emotional state in real time, and transmits the data to the server.

[0927] Running a simulation and displaying the results

[0928] The server runs a simulation in a virtual environment based on the received scenario and emotional data. It dynamically adjusts the simulation results taking into account the emotional data. The server analyzes the simulation results and visually displays them using statistical analysis and machine learning techniques. The terminal presents the analysis results to the user in the form of graphs and reports.

[0929] Specific examples

[0930] As a concrete example, consider a case where a user of an autonomous vehicle feels anxious during their morning commute due to a traffic jam. The emotion engine recognizes this anxiety and adjusts the autonomous vehicle to slow down a little and select a safe route.

[0931] Prompt Sentence Examples

[0932] An example of a prompt to enter an example into the system is:

[0933] "A user is using an autonomous vehicle during their morning commute. They encounter a traffic jam along the way, which makes them feel anxious. The emotion engine recognizes this feeling of anxiety, and the autonomous vehicle adjusts to slow down and choose a safer route."

[0934] A system configured in this way realizes highly accurate simulation and decision-making support that takes into account the user's emotional state.

[0935] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0936] Step 1: Data collection

[0937] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data, using techniques such as APIs, database queries, and web scraping. The input is data provided by each source, and the output is a huge amount of raw data.

[0938] Step 2: Data preprocessing and integration

[0939] The server performs preprocessing on the collected data, such as noise removal, missing value filling, and data standardization, to make it consistent. Specifically, it uses data cleaning algorithms, interpolates missing values, and standardizes the data. The input is raw data, and the output is preprocessed and integrated data.

[0940] Step 3: Training the generative artificial intelligence model

[0941] The server uses the preprocessed and integrated data to train a generative AI model. Specifically, it uses machine learning frameworks such as Scikit-learn, TensorFlow, and Keras to divide the dataset into training, validation, and testing sections, and selects an appropriate algorithm for training. The input is the preprocessed data, and the output is a trained AI model.

[0942] Step 4: Create a virtual environment

[0943] The server uses a trained artificial intelligence model to generate a virtual environment similar to the real world. Specifically, it uses 3D simulation software to build the environment based on the information the model has learned. The input is the trained model, and the output is the virtual environment.

[0944] Step 5: Entering the scenario

[0945] The user inputs the specific scenario they want to simulate through their device. The input scenario can include forecasts of market reactions to new products or carrier selection. The input is the scenario information provided by the user, and the output is the data sent from the device that receives it to the server.

[0946] Step 6: Collecting Emotion Data with the Emotion Engine

[0947] The device collects emotion data from the user's facial expressions, voice, input content, etc. The emotion engine analyzes emotions using models such as Keras. The input is the user's facial expressions and voice data, and the output is analyzed emotional state data.

[0948] Step 7: Run the simulation

[0949] The server executes a simulation in the virtual environment based on the received scenario and emotional data. Specifically, it dynamically adjusts simulation parameters based on the scenario and emotional data to generate results. The input is the user's scenario and emotional state data, and the output is the simulation results.

[0950] Step 8: Analyze and display simulation results

[0951] The server visually analyzes the simulation results and displays them in the form of graphs and reports. Specifically, it uses statistical analysis and machine learning techniques to summarize the results in an easy-to-understand format. The input is the simulation results, and the output is the analyzed results in graphs and reports.

[0952] Step 9: Support decision making and gather feedback

[0953] The user makes optimal decisions based on the displayed simulation results and sends feedback from the device to the server, which then performs further simulations and adjusts the model based on the feedback. The input is the user's feedback, and the output is the adjusted simulation results and model.

[0954] 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.

[0955] 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.

[0956] 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.

[0957] [Third embodiment]

[0958] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[0959] 5, the data processing system 310 includes the data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0960] 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).

[0961] 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.

[0962] 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.

[0963] 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).

[0964] 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.

[0965] 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.

[0966] 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.

[0967] 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.

[0968] 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.

[0969] 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."

[0970] This invention provides a system that uses generative artificial intelligence to build a virtual environment based on a huge amount of data and performs simulations within that virtual environment. This system aims to minimize risk by simulating real-world failures in the virtual environment in advance in order to improve the accuracy of decision-making.

[0971] System configuration and operation

[0972] 1. Data collection and integration

[0973] The servers collect vast amounts of data from various sources, including corporate, personal, and public data, using techniques such as APIs, database queries, and web scraping.

[0974] The server pre-processes the collected data, cleaning, normalizing, and consolidating it, transforming data from different sources into a unified format and making it consistent.

[0975] 2. Training a generative AI model

[0976] The server uses the preprocessed and integrated data to train a generative artificial intelligence model, selecting appropriate machine learning algorithms and training the model to learn patterns and trends in the data.

[0977] The training dataset is split into training, validation, and testing sets and is used to evaluate the performance of the model.

[0978] 3. Building a Virtual Environment

[0979] The server uses trained generative artificial intelligence models to generate virtual environments similar to the real world, allowing users to freely configure the scenarios they want to try out, recreating a wide range of scenarios and variables.

[0980] 4. Running the Simulation

[0981] The user inputs the specific scenario they want to simulate through the terminal, including specific settings such as new product features, pricing, and carrier selection.

[0982] The terminal transmits the user's input to the server, which runs the simulation within the virtual environment.

[0983] 5. Analyzing and displaying simulation results

[0984] The server analyzes the simulation results, finding trends and patterns in the data, and then uses specific statistical analysis and machine learning techniques to summarize the results in an easy-to-understand format.

[0985] The terminal visually displays the analysis results and provides them to the user, including graphs, charts, reports, etc.

[0986] 6. Decision support

[0987] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback and performs further simulations or re-adjusts the model as needed.

[0988] Specific examples

[0989] Case 1: A company decides to launch a new product

[0990] The user (corporate decision maker) inputs scenarios such as the characteristics and pricing of new products into the terminal.

[0991] The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions.

[0992] The terminal presents the analysis results to the user in the form of graphs and reports.

[0993] The user decides on a new product launch strategy based on the simulation results.

[0994] Case 2: Individual career choices

[0995] The user inputs a scenario regarding career choices (school choice, occupation choice, etc.) into the terminal.

[0996] The server simulates scenarios after career choices in a virtual environment and analyzes future annual income and job satisfaction.

[0997] The terminal presents the analysis results to the user as a report.

[0998] The user selects a carrier based on the simulation results.

[0999] In this way, the system of the present invention can support user decision-making and minimize risks through highly accurate simulations based on a variety of data.

[1000] The processing flow will be explained below.

[1001] Step 1: Data collection

[1002] The server retrieves data from a variety of sources, including corporate databases, public data services, and public data sources on the internet, using techniques such as APIs, database queries, and web scraping.

[1003] The server periodically collects data and keeps the necessary information up to date.

[1004] Step 2: Data Preprocessing

[1005] The server removes noise from the acquired data and fills in missing values. For example, missing data is interpolated using the average value or deleted.

[1006] The server performs data standardization and normalization, converting data from various formats into a unified format.

[1007] Step 3: Data Integration

[1008] The server consolidates the pre-processed data into one unified database.

[1009] The server maps and cross-references data schemas so that data from different sources is consistent.

[1010] Step 4: Prepare the dataset

[1011] The server extracts the training dataset from the integrated database and splits it into training, validation, and test datasets.

[1012] The server labels the data and performs feature engineering to prepare it in the optimal format for model training.

[1013] Step 5: Training the generative AI model

[1014] The server selects an appropriate generative AI algorithm and trains the model using a training dataset.

[1015] The server monitors the training process and tunes hyperparameters as needed.

[1016] Step 6: Evaluate the generative AI model

[1017] The server evaluates the model's performance using a validation dataset, using metrics such as precision, recall, and F1 score.

[1018] The server improves and retrains the model based on the evaluation results.

[1019] Step 7: Generate a Virtual Environment

[1020] The server uses a trained generative AI model to generate a virtual environment similar to the real world.

[1021] The server sets variables and parameters within the virtual environment to recreate realistic scenarios.

[1022] Step 8: Entering the Scenario

[1023] The user inputs the specific scenario he or she wants to simulate through the terminal.

[1024] The terminal collects the user's input and sends it to the server.

[1025] Step 9: Run the simulation

[1026] The server executes a simulation in a virtual environment based on the received specific scenario.

[1027] The server simulates multiple scenarios in parallel and collects data.

[1028] Step 10: Analyze the simulation results

[1029] The server analyzes the simulation results and finds trends and patterns in the data.

[1030] The server uses statistical analysis and machine learning techniques to compile the results into an easy-to-understand format.

[1031] Step 11: View the results

[1032] The terminal visually displays the simulation results sent from the server, including graphs, charts, and reports.

[1033] The user checks the displayed results and decides on the next action to take.

[1034] Step 12: Gather feedback

[1035] The user provides feedback on the simulation results through the terminal.

[1036] The terminal sends the collected feedback to the server.

[1037] Step 13: Further simulation

[1038] The server simulates additional scenarios or retunes the model based on user feedback.

[1039] The user inputs a new scenario as necessary and runs the simulation again.

[1040] In this way, the systems work together to help users make better decisions.

[1041] Example 1

[1042] 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."

[1043] Decisions in the real world involve many risks and uncertainties, so prior simulations are required to minimize risk. However, with conventional systems, the process from collecting and integrating massive amounts of data to training AI models, running simulations, and analyzing the results is often complex and inefficient, resulting in insufficient accuracy. It is also difficult for users to efficiently input specific scenarios and quickly make decisions based on the analysis results.

[1044] 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.

[1045] In this invention, the server includes means for collecting a huge amount of data, means for preprocessing and integrating the collected data, means for training a generative AI model using the preprocessed and integrated data, means for generating a virtual environment similar to the real world using the trained generative AI model, means for executing a simulation in the generated virtual environment, means for analyzing and displaying the results of the simulation, means for a user to input a specific scenario they wish to simulate, and means for automatically generating and analyzing multiple scenarios in the virtual environment. This enables users to make quick and accurate decisions through highly accurate simulations based on a huge amount of data.

[1046] "Big data" refers to data collected in large quantities from various sources, including corporate data, personal data, and public data.

[1047] "Preprocessing" refers to the process of cleaning (filling in missing values, removing outliers, etc.) and normalizing (formatting the data) the collected data.

[1048] "Integration" is the process of transforming data collected from different sources into a consistent format so that it can be treated as a single data set.

[1049] A "generative artificial intelligence model" is a model trained using machine learning algorithms to learn patterns and trends in data and perform simulations and predictions.

[1050] "Training" refers to the process of training a generative artificial intelligence model using preprocessed and integrated data.

[1051] "Virtual environment" refers to a simulated space that resembles the real world and is generated using a trained generative artificial intelligence model.

[1052] "Simulation" refers to the process of conducting experiments and verifications in a virtual environment based on scenarios and conditions specified by the user.

[1053] "Analysis" refers to the process of finding trends and patterns in the data from the simulation results, specifically using statistical analysis and machine learning techniques.

[1054] "Display" refers to providing the analyzed simulation results to the user in a visual format (graphs, charts, reports, etc.).

[1055] A "specific scenario" refers to the specific conditions or settings that the user wants to simulate (e.g., new product characteristics, pricing, carrier selection, etc.).

[1056] "Automatically generating and analyzing multiple scenarios" refers to the process of automatically generating multiple scenarios within a virtual environment based on user input and analyzing the results of each scenario.

[1057] MODE FOR CARRYING OUT THE INVENTION

[1058] This invention relates to a system that uses a generative artificial intelligence model based on a huge amount of data to create a virtual environment and perform simulations within that virtual environment. The system aims to minimize risk by simulating real-world failures in the virtual environment in advance in order to improve the accuracy of user decision-making.

[1059] System configuration

[1060] Hardware and Software Configuration

[1061] server:

[1062] Data collection is done using APIs (e.g., Twitter API, Google Analytics API), databases (e.g., MySQL, PostgreSQL), and web scraping tools (e.g., BeautifulSoup, Scrapy).

[1063] Use data cleansing tools (e.g., Pandas) for data preprocessing and integration.

[1064] Use machine learning libraries (e.g., TensorFlow, PyTorch) to train generative artificial intelligence models.

[1065] Statistical analysis tools (e.g., R, SciPy) and data analysis platforms (e.g., Jupyter Notebook, MATLAB) will be used for simulation and result analysis.

[1066] Device:

[1067] Use visualization tools (e.g., Matplotlib, Tableau) to display the results.

[1068] Data collection and preprocessing

[1069] The servers collect vast amounts of data, including corporate, personal, and public data, from various sources, including APIs, databases, and web scraping.

[1070] Example: A server retrieves the latest tweets from the Twitter API at regular intervals and stores them in a MySQL database.

[1071] The server cleans, normalizes, and consolidates the collected data, transforming it from different sources into a consistent format.

[1072] For example, the server uses Pandas to impute missing values, remove outliers, and convert each data set into a standard format.

[1073] Training generative artificial intelligence models

[1074] Based on the preprocessed data, the server selects an appropriate machine learning algorithm and trains a generative artificial intelligence model.

[1075] Example: The server uses TensorFlow to split the dataset into training, validation, and test data, and train and evaluate the model.

[1076] Building a virtual environment

[1077] The server uses a trained generative AI model to generate a virtual environment similar to the real world, which can be flexibly changed depending on the scenario or conditions the user wants to try.

[1078] Example: The server uses a generative AI model to generate various market scenarios based on the price range and market conditions set by the user.

[1079] Running and analyzing the simulation

[1080] Users input the specific scenario they want to simulate through the device, including the features and pricing of new products, carrier selection, and so on.

[1081] Example: A user enters prices and feature settings into the device interface and sends them to the server.

[1082] The server receives the scenario sent by the user and executes the simulation in the virtual environment.

[1083] Example: The server uses a generative AI model to run parallel simulations of scenarios based on input conditions.

[1084] The server analyzes the simulation results and finds trends and patterns in the data, using statistical analysis and machine learning techniques.

[1085] Example: The server uses SciPy to perform statistical analysis of the simulation results and graph sales forecasts and market reactions.

[1086] Results presentation and decision support

[1087] The terminal presents the analysis results to the user in a visual format, for example in the form of graphs, charts, or reports.

[1088] Example: The terminal receives the analysis results from the server and displays the results as a graph using Matplotlib.

[1089] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback from the user and performs further simulations or readjusts the model as needed.

[1090] Example: The user decides on a new product launch strategy based on the results and sends that feedback to the server, which uses the feedback to recalibrate the model.

[1091] Examples of concrete examples and prompts

[1092] Case 1: A company decides to launch a new product

[1093] Example of prompt text entered by the user (corporate decision maker):

[1094] Please define the following characteristics for your new product:

[1095] Function A

[1096] Price B

[1097] Target Market C

[1098] Simulate market reactions based on these conditions and analyze projected sales and market share.

[1099] The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions.

[1100] The terminal presents the analysis results to the user in the form of graphs and reports.

[1101] The user decides on a new product launch strategy based on the simulation results.

[1102] Case 2: Individual career choices

[1103] An example of a prompt that the user might enter:

[1104] "Enter the following career choice scenario:

[1105] Occupation X

[1106] University Y

[1107] Based on these, please analyze your future annual income and job satisfaction.

[1108] The server simulates scenarios after career choices in a virtual environment and analyzes future annual income and job satisfaction.

[1109] The terminal presents the analysis results to the user as a report.

[1110] The user makes a carrier selection based on the simulation results.

[1111] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1112] Step 1: Data collection

[1113] The server uses APIs (e.g., Twitter API, Google Analytics API), databases (e.g., MySQL, PostgreSQL), and web scraping tools (e.g., BeautifulSoup, Scrapy) to collect huge amounts of data, including corporate data, personal data, and public data.

[1114] Input: API endpoint, database query, website URL.

[1115] What it does: The server periodically fetches data from the API and stores it in a MySQL database. It also uses a web scraping tool to crawl websites and retrieve the required data.

[1116] Output: The raw data collected.

[1117] Step 2: Data Preprocessing

[1118] The server pre-processes the collected data, cleaning, normalizing, and integrating it, transforming data from different sources into a consistent format.

[1119] Input: Raw data collected.

[1120] Specific operation: The server uses Pandas to impute missing values, remove outliers, and convert each data into a standard format.

[1121] Output: The preprocessed dataset.

[1122] Step 3: Training the generative artificial intelligence model

[1123] Based on the preprocessed data, the server selects an appropriate machine learning algorithm and trains a generative artificial intelligence model.

[1124] Input: The preprocessed dataset.

[1125] Specific operation: The server uses TensorFlow to split the dataset into training data, validation data, and test data, and trains and evaluates the model.

[1126] Output: A trained generative AI model.

[1127] Step 4: Build a virtual environment

[1128] The server uses a trained generative AI model to generate a virtual environment similar to the real world, which can be flexibly changed depending on the scenario or conditions the user wants to try.

[1129] Input: A trained generative AI model.

[1130] Specific operation: The server uses the generative AI model to set various parameters within the virtual environment and provide scenario templates that can be customized by the user.

[1131] Output: A customizable virtual environment.

[1132] Step 5: Enter the simulation scenario

[1133] Users input the specific scenario they want to simulate through the device, including new product features, pricing, carrier selection, etc.

[1134] Input: Scenario conditions set by the user (e.g., new product characteristics, pricing).

[1135] Specific operation: The user inputs the required information into the terminal interface and sends it to the server.

[1136] Output: User's scenario configuration data.

[1137] Step 6: Run the simulation

[1138] The server receives the scenario sent by the user and executes the simulation in the virtual environment. The simulation automatically generates and analyzes multiple scenarios based on the specified conditions.

[1139] Input: User-submitted scenario configuration data, customizable virtual environment.

[1140] Specific operation: The server uses the generative AI model to simulate multiple scenarios in parallel and record the results.

[1141] Output: Simulation result data.

[1142] Step 7: Analyze the simulation results

[1143] The server analyzes the simulation results and finds trends and patterns in the data, using statistical analysis and machine learning techniques.

[1144] Input: Simulation result data.

[1145] How it works: The server uses SciPy to statistically analyze the data and visualize the results, such as sales forecasts and market reactions.

[1146] Output: Analyzed simulation results.

[1147] Step 8: View the results

[1148] The terminal visually displays the analysis results and provides them to the user, specifically in the form of graphs, charts, and reports.

[1149] Input: Analyzed simulation results.

[1150] Specific operation: The terminal receives the analysis results sent from the server and visually displays the results using Matplotlib.

[1151] Output: The visualization data that is presented to the user.

[1152] Step 9: Decision making and feedback

[1153] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback from the user and performs further simulations or readjusts the model as needed.

[1154] Input: User decision results, feedback data.

[1155] How it works: The user decides on a strategy based on the results and enters it into the device. The server receives the feedback and retrains the model as needed.

[1156] Output: Retuned generative AI model, further simulation results.

[1157] (Application example 1)

[1158] 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."

[1159] In content distribution services, accurately predicting user viewing trends and engagement and determining optimal distribution schedules based on the results is a very difficult challenge. Conventional methods only allow for limited predictions based on past data and experience, and are unable to effectively capture viewer reactions. Therefore, new technologies are needed to optimize content distribution schedules and improve engagement.

[1160] 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.

[1161] In this invention, the server includes means for collecting a huge amount of data, means for preprocessing and integrating the collected data, means for training a generative artificial intelligence model using the preprocessed and integrated data, means for generating a virtual environment similar to the real world using the trained generative artificial intelligence model, means for running a simulation in the generated virtual environment, means for analyzing and displaying the results of the simulation, means for predicting viewer responses based on the collected data, means for predicting an engagement score for adjusting a content delivery schedule, and means for visualizing the simulation results and providing them to a user. This makes it possible to predict viewer responses with high accuracy and to formulate an optimal delivery schedule based on the predictions.

[1162] "Big data" refers to large amounts of digital information collected from a wide variety of sources.

[1163] "Preprocessing" refers to a series of steps that transform collected data into a form that is applicable for analysis and modeling.

[1164] "Integration" refers to the act of bringing together data collected from different sources into a consistent format.

[1165] A "generative artificial intelligence model" refers to a machine learning model trained on collected and preprocessed data.

[1166] A "virtual environment" refers to a digital space created to simulate real-world actions or events.

[1167] "Simulation" refers to the process of experimenting with behaviors and outcomes under specific conditions within a virtual environment.

[1168] "Analysis" refers to the process of summarizing simulation results and data trends in an easy-to-understand format.

[1169] "Display" refers to the act of visually presenting the analysis results to the user.

[1170] "Viewer response" refers to the viewer's behavior and reaction to content.

[1171] "Engagement score" refers to a numerical indicator of the level of engagement that viewers show with content.

[1172] "Content delivery schedule" refers to a plan that determines when particular content should be delivered.

[1173] A "scenario" refers to the specific conditions and settings that a user inputs to perform a simulation.

[1174] "Optimization" refers to adjusting a system or process to its best state according to a specific goal.

[1175] The system of the present invention provides a means to collect, preprocess, and integrate massive amounts of data, train generative artificial intelligence models to generate virtual environments that resemble the real world, and run simulations within those virtual environments to determine optimal content delivery schedules.

[1176] The system includes a server, a terminal, and a user.

[1177] Data collection and preprocessing

[1178] The server collects a huge amount of data using techniques such as APIs, database queries, and web scraping. The collected data includes information such as viewer viewing history, viewing times, and behavioral patterns. The server then cleans and normalizes the data, integrating it for consistency.

[1179] Training generative artificial intelligence models

[1180] Using the preprocessed and integrated data, the server trains a generative artificial intelligence model. The training dataset is divided into training, validation, and testing datasets, and the target machine learning algorithm predicts viewer viewing habits. The model is designed to predict viewer behavior and reactions, and the algorithm used for training can be a linear regression model.

[1181] Virtual environment generation and simulation

[1182] Using a trained generative artificial intelligence model, the server generates a virtual environment similar to the real world. Users can input specific scenarios via their devices, specifying specific dates and times, the number of views, likes, shares, etc. The server then runs a simulation based on these scenarios and predicts viewer reactions.

[1183] Analyzing and displaying results

[1184] The results of the simulation are analyzed by the server. The analysis results include viewer reactions and engagement scores. The device visualizes these results and provides them to the user. The user can view the results in the form of graphs, charts, and reports.

[1185] Example

[1186] For example, if a user wants to stream at 8pm with the hope of getting 5000 views, 300 likes, and 50 shares, they can enter a specific scenario. This scenario can be expressed with a prompt like this:

[1187] Based on your viewing data, predict your viewer engagement score for the following stream schedule:

[1188] Views: 5000

[1189] Likes: 300

[1190] Shares: 50

[1191] Time: 8 PM

[1192] Based on this prompt, the server runs a simulation in the virtual environment to predict the viewer's reaction. The prediction results are provided to the user via their device, and the user can then decide on the optimal distribution schedule.

[1193] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1194] Step 1:

[1195] The server collects a huge amount of data. The types of data include viewer viewing history, viewing times, and behavioral patterns. This data is obtained using APIs, database queries, and web scraping techniques. Specifically, data is obtained by issuing queries to a viewing history database. The input in this step is the viewer data source, and the output is raw viewer data.

[1196] Step 2:

[1197] The server preprocesses and integrates the collected data. It removes noise from the data, imputes missing values, and converts various data sources into a consistent format. This step includes cleaning and normalizing the data. Specifically, it uses data cleaning tools to handle missing values. The input to this step is the collected data from step 1, and the output is the preprocessed and integrated data.

[1198] Step 3:

[1199] The server trains a generative artificial intelligence model using the preprocessed and integrated data. Specifically, it splits the dataset into training, validation, and test sets, and trains the model using a machine learning algorithm such as a linear regression model. The input in this step is the preprocessed data from step 2, and the output is a trained AI model.

[1200] Step 4:

[1201] The server generates a virtual environment using a trained generative AI model. The virtual environment is a digital space for simulating viewer behavior. Various scenarios and variables can be set, enabling various simulations. The input in this step is the trained AI model, and the output is the virtual environment.

[1202] Step 5:

[1203] Through the terminal, the user inputs the specific scenario they want to simulate. The scenario includes parameters such as a specific date and time, expected number of viewers, number of likes, number of shares, etc. The input in this step is the user's scenario information, and the output is the scenario data sent to the server.

[1204] Step 6:

[1205] The server runs a simulation in the virtual environment based on the input scenario. Specifically, it uses the model to generate data to predict viewer reactions and then runs the simulation. The input in this step is the scenario data from step 5, and the output is the simulation results.

[1206] Step 7:

[1207] The server analyzes the simulation results. The analysis aims to predict viewer responses and engagement scores. This procedure includes analyzing data trends using statistical analysis and machine learning techniques. The input in this step is the simulation results, and the output is the analysis results.

[1208] Step 8:

[1209] The device visualizes the analysis results and provides them to the user. Specifically, the analysis results are displayed in the form of graphs, charts, and reports. The user views these results and determines the optimal content distribution schedule based on the engagement scores. The input in this step is the analysis results, and the output is the visualized analysis results provided to the user.

[1210] By following the steps above, it becomes possible to predict viewer reactions with high accuracy and formulate an optimal distribution schedule.

[1211] 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.

[1212] The present invention is a system that collects massive amounts of data, integrates and preprocesses them, trains a generative artificial intelligence model, and runs simulations in a virtual environment using the generated model. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to support more accurate decision-making that takes into account the user's emotional state. The present invention is implemented using the following processing steps and configuration.

[1213] System configuration and operation

[1214] 1. Data collection and integration

[1215] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data, and uses techniques such as APIs, database queries, and web scraping to efficiently retrieve the data.

[1216] The server performs preprocessing on the collected data, such as noise removal, missing value filling, and data standardization, to ensure consistency.

[1217] 2. Training a generative AI model

[1218] The server uses the preprocessed and integrated data to train a generative artificial intelligence model, selecting an appropriate machine learning algorithm and splitting the dataset into training, validation, and testing sections.

[1219] Evaluate the performance of the model and adjust the hyperparameters as necessary.

[1220] 3. Creating a virtual environment

[1221] The server uses a trained generative artificial intelligence model to generate a virtual environment similar to the real world, with the flexibility to recreate a variety of scenarios.

[1222] 4. Enter the scenario

[1223] The user inputs the specific scenario they want to simulate through the terminal, including, for example, the characteristics and pricing of a new product, carrier selection, etc.

[1224] The terminal transmits the input scenario to the server.

[1225] 5. Collecting Emotion Data Using an Emotion Engine

[1226] The device collects emotional data from the user's facial expressions, voice, input, etc.

[1227] The emotion engine analyzes the collected emotion data and recognizes the user's emotional state in real time.

[1228] 6. Running the Simulation

[1229] The server runs a simulation in the virtual environment based on the received scenario and emotion data, dynamically adjusting the scenario based on the emotion data to provide more realistic results.

[1230] 7. Analyzing and displaying simulation results

[1231] The server analyzes the simulation results and uses statistical analysis and machine learning techniques to summarize the results in an easy-to-understand format.

[1232] The terminal provides the user with a visual display of the analysis results, including graphs, charts, and reports.

[1233] 8. Decision support and feedback gathering

[1234] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback from the user and performs additional simulations or refines the model.

[1235] Specific examples

[1236] Case 1: A company decides to launch a new product

[1237] Users (corporate decision makers) input scenarios such as new product characteristics and pricing into the terminal, and their emotional state (e.g., excitement, anxiety, etc.) is collected at the same time.

[1238] The emotion engine analyzes the user's emotional state in real time and transmits it to the server.

[1239] The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions taking into account emotional data.

[1240] The terminal presents the analysis results to the user in the form of graphs and reports.

[1241] The user decides on a new product launch strategy based on the simulation results.

[1242] Case 2: Individual career choices

[1243] The user inputs a scenario regarding career choices (such as the school to attend, the occupation to choose, etc.) into the terminal, and their emotional state (such as anxiety, expectations, etc.) is collected at the same time.

[1244] The emotion engine analyzes the user's emotional state and transmits it to the server.

[1245] The server simulates scenarios after career selection and analyzes future annual income and job satisfaction taking into account emotional data.

[1246] The terminal presents the analysis results to the user as a report.

[1247] The user selects a carrier based on the simulation results.

[1248] In this way, the system of the present invention, which is combined with an emotion engine, provides highly accurate simulation and decision support that takes into account the user's emotional state, helping the user make optimal choices.

[1249] The processing flow will be explained below.

[1250] Step 1: Data collection

[1251] The server retrieves data from corporate databases, public data services, and public data sources on the Internet, using techniques such as APIs, database queries, and web scraping to efficiently gather data.

[1252] The server periodically updates the data to keep it up to date.

[1253] Step 2: Data Preprocessing

[1254] The server removes noise from the acquired data and fills in missing values. Specifically, it interpolates the average value of missing data and removes outliers.

[1255] The server performs data standardization (eg, normalization, scaling) and converts data from different sources into a consistent format.

[1256] Step 3: Data Integration

[1257] The server consolidates the pre-processed data into one unified database.

[1258] The server maps the schema of the data from different sources and makes it consistent.

[1259] Step 4: Prepare the dataset

[1260] The server extracts training, validation, and test datasets from the integrated database.

[1261] The server labels the data and performs feature engineering to prepare it in a format that is easy for the model to learn.

[1262] Step 5: Training the generative AI model

[1263] The server selects an appropriate generative AI algorithm (e.g., a deep learning model) and trains the model using a training dataset.

[1264] The server monitors the training process and optimizes hyperparameters as needed.

[1265] Step 6: Evaluate the generative AI model

[1266] The server evaluates the model's performance using a validation dataset, using metrics such as precision, recall, and F1 score.

[1267] The server will make any necessary improvements based on the evaluation results.

[1268] Step 7: Generate a Virtual Environment

[1269] The server uses a trained generative AI model to generate a virtual environment that resembles the real world.

[1270] The server sets the variables and parameters to be simulated within the virtual environment.

[1271] Step 8: Entering the Scenario

[1272] The user inputs the specific scenario he or she wants to simulate (for example, the characteristics and pricing of a new product, carrier selection, etc.) through the terminal.

[1273] The terminal transmits the user's input to the server.

[1274] Step 9: Collect emotion data

[1275] The device collects emotional data from the user's facial expressions, voice, input, etc. This is typically done using devices such as a camera and microphone.

[1276] The emotion engine analyzes the collected data and recognizes the user's emotional state in real time.

[1277] Step 10: Run the simulation

[1278] The server executes a simulation in the virtual environment based on the scenario received from the user and the emotion data from the emotion engine.

[1279] The server dynamically adjusts the simulation to produce results that reflect real-world emotional states.

[1280] Step 11: Analyze the simulation results

[1281] The server collects and analyzes the simulation results, using statistical analysis and machine learning techniques to extract trends and patterns from the results.

[1282] The server summarizes the analysis results in an easy-to-understand format (e.g., graphs, charts, reports).

[1283] Step 12: View the results

[1284] The terminal visually displays the simulation results sent from the server, presenting the results using visual elements (e.g., graphs, charts, reports) so that the user can easily understand them.

[1285] The user checks the results displayed on the terminal and makes a decision.

[1286] Step 13: Gather feedback

[1287] The user provides feedback on the simulation results and the displayed content through the terminal.

[1288] The terminal sends the collected feedback to the server.

[1289] Step 14: Further simulation

[1290] The server simulates additional scenarios or retunes the model based on user feedback.

[1291] The user inputs a new scenario as needed, and the server executes a re-simulation based on that.

[1292] In this way, the systems work together to help users make better, more emotionally informed decisions.

[1293] Example 2

[1294] 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."

[1295] Currently, there are systems that utilize a large amount of data to provide highly accurate decision-making support, but few of them take into account the user's emotional state. As a result, conventional systems are unable to fully reflect the impact of the user's emotions on decision-making, making it difficult to make optimal decisions.

[1296] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1297] In this invention, the server includes means for collecting a huge amount of data, means for preprocessing and integrating the collected data, means for training a generative AI model using the preprocessed and integrated data, means for generating a virtual environment similar to the real world using the trained generative AI model, means for analyzing the collected emotional data and recognizing a user's emotional state in real time, means for dynamically adjusting a simulation scenario based on the emotional data, and means for analyzing and displaying the results of the simulation, thereby enabling highly accurate simulation and decision-making support that takes the user's emotional state into account.

[1298] "Big data" refers to a large and diverse collection of information, collected from sources such as corporate data, personal data, and public data.

[1299] "Preprocessing" refers to the process of carrying out procedures such as noise removal, missing value filling, and data standardization on collected data to make the data consistent.

[1300] "Synthesis" is the process of combining pre-processed data into one unified data set.

[1301] "Training a generative artificial intelligence model" is the process of applying machine learning algorithms to preprocessed and integrated data to train the model for optimal performance.

[1302] A "virtual environment" is an artificial environment that uses a generative artificial intelligence model to create a simulated environment similar to the real world.

[1303] "Emotional data" refers to information collected from the user's facial expressions, voice, input content, etc., that indicates the user's emotional state in real time.

[1304] "Analyzing emotional data" refers to the process of recognizing the user's emotional state based on the collected emotional data, and detecting classifications or specific emotional states as needed.

[1305] "Dynamic adjustment of the simulation scenario" refers to the process of appropriately changing the content of the scenario based on the user's emotional state during the simulation, in order to bring the results closer to reality.

[1306] "Displaying the simulation results" refers to the process of visually expressing the data obtained from the simulation and presenting it in a form that is easy for the user to understand.

[1307] MODE FOR CARRYING OUT THE INVENTION

[1308] The present invention is a system that collects massive amounts of data, integrates and preprocesses them, trains a generative artificial intelligence model, and runs simulations in a virtual environment using the generated model. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to support more accurate decision-making that takes into account the user's emotional state. This system is configured as follows:

[1309] 1. Data collection and integration

[1310] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data. Collection methods include APIs, database queries, and web scraping tools (e.g., Beautiful Soup and Scrapy). The collected data is preprocessed using the Python Pandas library, which removes noise, fills in missing values, and standardizes the data to ensure consistency.

[1311] 2. Training a generative AI model

[1312] The server trains a generative artificial intelligence model using the preprocessed and integrated data. It uses machine learning algorithms such as TensorFlow and PyTorch to split the dataset into training, validation, and testing sets. It also uses Optuna and Grid Search to tune the model's hyperparameters and optimize accuracy.

[1313] 3. Creating a virtual environment

[1314] The server uses the trained generative AI model to generate a virtual environment similar to the real world, using a virtual environment generation tool (e.g., Unity or Unreal Engine) to recreate the scenario required for the simulation.

[1315] 4. Enter the scenario

[1316] The user inputs a specific simulation scenario through the terminal. This input process involves using a web application form or an interactive chat window to enter details such as new product features, pricing, and carrier selection. The terminal then sends the input scenario data to the server via an HTTP request.

[1317] 5. Emotional Data Collection and Analysis Using an Emotional Engine

[1318] The device collects emotion data from the user's facial expressions, voice, and input. For example, it captures facial expressions with a camera, analyzes them using OpenCV, and converts the speech into text using a speech recognition engine (e.g., Google Speech-to-Text). The emotion engine analyzes the collected emotion data and uses a deep learning model to classify and recognize emotions into multiple categories, such as "happiness," "sadness," and "excitement."

[1319] 6. Running the Simulation

[1320] The server executes a simulation based on the received scenario and emotion data. The scenario is dynamically adjusted based on the emotion data, and the simulation is performed in real time within the virtual environment. For example, if a user wants to predict market reaction to a new product, they can input the characteristics and pricing of the new product, and if the emotion engine interprets this as "excitement," it will affect the simulation results.

[1321] 7. Analyzing and displaying simulation results

[1322] The server analyzes the simulation results and uses statistical analysis and machine learning techniques to compile the results into easy-to-understand formats, such as Pandas and Matplotlib, to convert the data into graphs and charts. The terminal visually displays these results and provides them to the user in an easy-to-understand format.

[1323] 8. Decision support and feedback gathering

[1324] The user makes optimal decisions based on the displayed simulation results. The server collects feedback from the user and uses it to conduct additional simulations and refine the model, thereby enabling more accurate decision-making support.

[1325] Specific examples

[1326] Case 1: A company decides to launch a new product

[1327] 1. The user (corporate decision maker) inputs scenarios such as new product characteristics and pricing into the terminal, and their emotional state (e.g., excitement) is collected at the same time.

[1328] 2. The emotion engine analyzes the user's emotional state in real time and sends it to the server.

[1329] 3. The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions taking into account emotional data.

[1330] 4. The terminal presents the analysis results to the user in the form of graphs and reports.

[1331] 5. The user decides on a new product launch strategy based on the simulation results.

[1332] Case 2: Individual career choices

[1333] 1. The user inputs a scenario about career choices (such as the school they will attend, the occupation they will choose, etc.) into the terminal, and their emotional state (e.g., anxiety, expectation) is collected at the same time.

[1334] 2. The emotion engine analyzes the user's emotional state and sends it to the server.

[1335] 3. The server simulates scenarios after career selection and analyzes future annual income and job satisfaction taking into account emotional data.

[1336] 4. The device presents the analysis results to the user as a report.

[1337] 5. The user makes a carrier selection based on the simulation results.

[1338] Prompt Sentence Examples

[1339] If you want to predict market reaction to a new product:

[1340] Predict the market reaction to a newly developed smartphone. The characteristics are 5G, a 6.5-inch display, a triple camera, and a price of $800. The emotional state is "Excited."

[1341] If you would like to discuss your career options:

[1342] I'm looking for help choosing a future career. My options are to go to college for engineering or to go to design school. My current emotional state is a mixture of anxiety and excitement.

[1343] In this way, the system of the present invention, which is combined with an emotion engine, provides highly accurate simulation and decision support that takes into account the user's emotional state, helping the user make optimal choices.

[1344] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1345] Step 1:

[1346] The server collects data from various sources, including corporate data, personal data, and public data. It uses APIs, database queries, and web scraping tools (such as Beautiful Soup and Scrapy) as input. It saves the collected data in JSON or CSV format as output. Specifically, it periodically accesses a specified API endpoint, retrieves the latest data, and saves it in a local database.

[1347] Step 2:

[1348] The server preprocesses and standardizes the collected data. It receives the data collected in the previous step as input. It denoises the data, fills missing values, and standardizes the data. For example, it uses the Python Pandas library to generate a data frame, removes inaccurate values, and fills missing values ​​with the median. As output, it obtains a preprocessed, clean dataset.

[1349] Step 3:

[1350] The server trains a generative AI model using the preprocessed data. It receives the preprocessed data as input and trains the model using a machine learning algorithm (e.g., TensorFlow or PyTorch). It splits the dataset into training, validation, and test datasets and feeds each dataset into the model. It uses Optuna or Grid Search to tune the hyperparameters. As output, it obtains a trained generative AI model.

[1351] Step 4:

[1352] The server generates a virtual environment using a trained generative AI model. It receives the trained model and scenario data as input. It uses a virtual environment generation tool (such as Unity or Unreal Engine) to create an environment similar to the real world. As output, it obtains a virtual environment for simulation.

[1353] Step 5:

[1354] The user inputs a specific simulation scenario through a terminal. For input, the scenario (such as the characteristics and pricing of a new product) is entered using a form or chat window in the web application. For output, the scenario data is sent to the server.

[1355] Step 6:

[1356] The device collects the user's emotional data. As input, the user's facial expressions and voice are captured using a camera and microphone. OpenCV and Google Speech-to-Text are used to analyze the emotional data. The analyzed emotional data is obtained as output.

[1357] Step 7:

[1358] The emotion engine analyzes collected emotion data in real time to recognize the user's emotional state. It receives data from the camera and microphone as input and applies emotion recognition algorithms. The output is classified emotion data (e.g., joy, sadness, excitement).

[1359] Step 8:

[1360] The server executes a simulation using the received scenario and emotion data. It receives the scenario data and emotion data as input. It performs a simulation in the virtual environment while dynamically adjusting the scenario based on the emotion data. It obtains the simulation results as output.

[1361] Step 9:

[1362] The server analyzes the simulation results and generates data for visual display. It receives the simulation results as input and organizes them into an easy-to-understand format using statistical analysis and machine learning techniques. The output is analysis data, graphs, and charts.

[1363] Step 10:

[1364] The terminal visually displays the analysis results and provides them to the user. It receives the analysis data as input. It uses a web interface to display the results in the form of interactive graphs and reports. It outputs the results in a format that is easy for the user to understand.

[1365] Step 11:

[1366] The user makes optimal decisions based on the displayed simulation results. The analysis results are used as input. The information necessary for decision-making is collected and reflected in actual actions. For example, decisions are made regarding new product launch strategies or career choices. The user's decision is obtained as output.

[1367] Step 12:

[1368] The server collects user feedback and uses it to refine the model and conduct additional simulations. It receives user feedback as input, updates the model based on it, and retrains it to improve its accuracy. The output is an improved generative AI model.

[1369] (Application example 2)

[1370] 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."

[1371] Conventional autonomous driving systems control operation based on external environmental data, but do not take into account the emotional state of passengers, which means they are unable to fully alleviate the anxiety and stress felt by passengers. Furthermore, due to the lack of technology to collect and analyze emotional data and dynamically adjust driving modes, there is an issue of not being able to improve passenger safety and comfort.

[1372] 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 means for collecting a huge amount of data, means for preprocessing and integrating the collected data, and means for training a generative artificial intelligence model using the preprocessed and integrated data. This enables the operation of an autonomous vehicle that improves safety and comfort by analyzing passenger emotional states in real time and reflecting this in driving control.

[1373] "Big data" refers to data collected in large quantities from various sources, including corporate data, personal data, and public data.

[1374] "Preprocessing" refers to the process of performing operations such as noise removal, missing value filling, and data standardization on collected data to improve consistency and quality.

[1375] "Integration" is the process of bringing together data collected from different data sources, making it consistent and easier to process and analyze.

[1376] A "generative artificial intelligence model" is a model trained using machine learning algorithms that is generated to perform specific tasks based on large amounts of data.

[1377] A "virtual environment" is a simulation environment that imitates the real world, and is an environment that reproduces various situations based on scenarios and conditions specified by the user.

[1378] "Simulation" is a technique for reproducing phenomena based on specific scenarios and conditions in a virtual environment and analyzing the results.

[1379] The "emotion engine" is a system that recognizes a user's emotional state in real time by collecting and analyzing emotional data from the user's facial expressions, voice, input content, etc.

[1380] "Dynamic adjustment" refers to the act of changing and optimizing system behavior and settings on the fly based on real-time information such as emotional data.

[1381] "Analysis" refers to the process of compiling collected data and simulation results into an easy-to-understand format and extracting the meaning and trends of the data using statistical analysis and machine learning techniques.

[1382] To specifically implement this invention, the system configuration and operating procedures are as follows: The system collects, preprocesses, and integrates massive amounts of data, trains a generative AI model based on the data, and uses the generated model to run a simulation in a virtual environment similar to the real world. Furthermore, the system analyzes the user's emotional state in real time and dynamically adjusts the simulation results to support more accurate decision-making.

[1383] Components and their operation

[1384] Data collection and preprocessing

[1385] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data, using techniques such as APIs, database queries, and web scraping. The collected data undergoes preprocessing, such as noise removal, missing value filling, and data standardization, to ensure consistency.

[1386] Training generative artificial intelligence models

[1387] The server trains generative AI models using the preprocessed and integrated data. Machine learning algorithms are implemented using Scikit-learn, TensorFlow, Keras, etc., and datasets are divided into training, validation, and testing phases. Model performance is also evaluated and hyperparameters are adjusted.

[1388] Creating a virtual environment and inputting a scenario

[1389] Using a trained generative artificial intelligence model, a virtual environment similar to the real world is generated. The user inputs the specific scenario they want to simulate through their device. The device then sends the input scenario to the server, which then starts the simulation in the virtual environment.

[1390] Analysis by emotion engine

[1391] The device collects emotional data from the user's facial expressions, voice, input, etc. The emotion engine, built using Keras and other tools, analyzes the collected emotional data, recognizes the user's emotional state in real time, and transmits the data to the server.

[1392] Running a simulation and displaying the results

[1393] The server runs a simulation in a virtual environment based on the received scenario and emotional data. It dynamically adjusts the simulation results taking into account the emotional data. The server analyzes the simulation results and visually displays them using statistical analysis and machine learning techniques. The terminal presents the analysis results to the user in the form of graphs and reports.

[1394] Specific examples

[1395] As a concrete example, consider a case where a user of an autonomous vehicle feels anxious during their morning commute due to a traffic jam. The emotion engine recognizes this anxiety and adjusts the autonomous vehicle to slow down a little and select a safe route.

[1396] Prompt Sentence Examples

[1397] An example of a prompt to enter an example into the system is:

[1398] "A user is using an autonomous vehicle during their morning commute. They encounter a traffic jam along the way, which makes them feel anxious. The emotion engine recognizes this feeling of anxiety, and the autonomous vehicle adjusts to slow down and choose a safer route."

[1399] A system configured in this way realizes highly accurate simulation and decision-making support that takes into account the user's emotional state.

[1400] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1401] Step 1: Data collection

[1402] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data, using techniques such as APIs, database queries, and web scraping. The input is data provided by each source, and the output is a huge amount of raw data.

[1403] Step 2: Data preprocessing and integration

[1404] The server performs preprocessing on the collected data, such as noise removal, missing value filling, and data standardization, to make it consistent. Specifically, it uses data cleaning algorithms, interpolates missing values, and standardizes the data. The input is raw data, and the output is preprocessed and integrated data.

[1405] Step 3: Training the generative artificial intelligence model

[1406] The server uses the preprocessed and integrated data to train a generative AI model. Specifically, it uses machine learning frameworks such as Scikit-learn, TensorFlow, and Keras to divide the dataset into training, validation, and testing sections, and selects an appropriate algorithm for training. The input is the preprocessed data, and the output is a trained AI model.

[1407] Step 4: Create a virtual environment

[1408] The server uses a trained artificial intelligence model to generate a virtual environment similar to the real world. Specifically, it uses 3D simulation software to build the environment based on the information the model has learned. The input is the trained model, and the output is the virtual environment.

[1409] Step 5: Entering the scenario

[1410] The user inputs the specific scenario they want to simulate through their device. The input scenario can include forecasts of market reactions to new products or carrier selection. The input is the scenario information provided by the user, and the output is the data sent from the device that receives it to the server.

[1411] Step 6: Collecting Emotion Data with the Emotion Engine

[1412] The device collects emotion data from the user's facial expressions, voice, input content, etc. The emotion engine analyzes emotions using models such as Keras. The input is the user's facial expressions and voice data, and the output is analyzed emotional state data.

[1413] Step 7: Run the simulation

[1414] The server executes a simulation in the virtual environment based on the received scenario and emotional data. Specifically, it dynamically adjusts simulation parameters based on the scenario and emotional data to generate results. The input is the user's scenario and emotional state data, and the output is the simulation results.

[1415] Step 8: Analyze and display simulation results

[1416] The server visually analyzes the simulation results and displays them in the form of graphs and reports. Specifically, it uses statistical analysis and machine learning techniques to summarize the results in an easy-to-understand format. The input is the simulation results, and the output is the analyzed results in graphs and reports.

[1417] Step 9: Support decision making and gather feedback

[1418] The user makes optimal decisions based on the displayed simulation results and sends feedback from the device to the server, which then performs further simulations and adjusts the model based on the feedback. The input is the user's feedback, and the output is the adjusted simulation results and model.

[1419] 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.

[1420] 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.

[1421] 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.

[1422] [Fourth embodiment]

[1423] FIG. 7 shows an example of the configuration of a data processing system 410 according to the fourth embodiment.

[1424] 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.

[1425] 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).

[1426] 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.

[1427] 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.

[1428] 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).

[1429] 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.

[1430] 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.

[1431] 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.

[1432] 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.

[1433] 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.

[1434] 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.

[1435] 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."

[1436] This invention provides a system that uses generative artificial intelligence to build a virtual environment based on a huge amount of data and performs simulations within that virtual environment. This system aims to minimize risk by simulating real-world failures in the virtual environment in advance in order to improve the accuracy of decision-making.

[1437] System configuration and operation

[1438] 1. Data collection and integration

[1439] The servers collect vast amounts of data from various sources, including corporate, personal, and public data, using techniques such as APIs, database queries, and web scraping.

[1440] The server pre-processes the collected data, cleaning, normalizing, and consolidating it, transforming data from different sources into a unified format and making it consistent.

[1441] 2. Training a generative AI model

[1442] The server uses the preprocessed and integrated data to train a generative artificial intelligence model, selecting appropriate machine learning algorithms and training the model to learn patterns and trends in the data.

[1443] The training dataset is split into training, validation, and testing sets and is used to evaluate the performance of the model.

[1444] 3. Building a Virtual Environment

[1445] The server uses trained generative artificial intelligence models to generate virtual environments similar to the real world, allowing users to freely configure the scenarios they want to try out, recreating a wide range of scenarios and variables.

[1446] 4. Running the Simulation

[1447] The user inputs the specific scenario they want to simulate through the terminal, including specific settings such as new product features, pricing, and carrier selection.

[1448] The terminal transmits the user's input to the server, which runs the simulation within the virtual environment.

[1449] 5. Analyzing and displaying simulation results

[1450] The server analyzes the simulation results, finding trends and patterns in the data, and then uses specific statistical analysis and machine learning techniques to summarize the results in an easy-to-understand format.

[1451] The terminal visually displays the analysis results and provides them to the user, including graphs, charts, reports, etc.

[1452] 6. Decision support

[1453] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback and performs further simulations or re-adjusts the model as needed.

[1454] Specific examples

[1455] Case 1: A company decides to launch a new product

[1456] The user (corporate decision maker) inputs scenarios such as the characteristics and pricing of new products into the terminal.

[1457] The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions.

[1458] The terminal presents the analysis results to the user in the form of graphs and reports.

[1459] The user decides on a new product launch strategy based on the simulation results.

[1460] Case 2: Individual career choices

[1461] The user inputs a scenario regarding career choices (school choice, occupation choice, etc.) into the terminal.

[1462] The server simulates scenarios after career choices in a virtual environment and analyzes future annual income and job satisfaction.

[1463] The terminal presents the analysis results to the user as a report.

[1464] The user selects a carrier based on the simulation results.

[1465] In this way, the system of the present invention can support user decision-making and minimize risks through highly accurate simulations based on a variety of data.

[1466] The processing flow will be explained below.

[1467] Step 1: Data collection

[1468] The server retrieves data from a variety of sources, including corporate databases, public data services, and public data sources on the internet, using techniques such as APIs, database queries, and web scraping.

[1469] The server periodically collects data and keeps the necessary information up to date.

[1470] Step 2: Data Preprocessing

[1471] The server removes noise from the acquired data and fills in missing values. For example, missing data is interpolated using the average value or deleted.

[1472] The server performs data standardization and normalization, converting data from various formats into a unified format.

[1473] Step 3: Data Integration

[1474] The server consolidates the pre-processed data into one unified database.

[1475] The server maps and cross-references data schemas so that data from different sources is consistent.

[1476] Step 4: Prepare the dataset

[1477] The server extracts the training dataset from the integrated database and splits it into training, validation, and test datasets.

[1478] The server labels the data and performs feature engineering to prepare it in the optimal format for model training.

[1479] Step 5: Training the generative AI model

[1480] The server selects an appropriate generative AI algorithm and trains the model using a training dataset.

[1481] The server monitors the training process and tunes hyperparameters as needed.

[1482] Step 6: Evaluate the generative AI model

[1483] The server evaluates the model's performance using a validation dataset, using metrics such as precision, recall, and F1 score.

[1484] The server improves and retrains the model based on the evaluation results.

[1485] Step 7: Generate a Virtual Environment

[1486] The server uses a trained generative AI model to generate a virtual environment similar to the real world.

[1487] The server sets variables and parameters within the virtual environment to recreate realistic scenarios.

[1488] Step 8: Entering the Scenario

[1489] The user inputs the specific scenario he or she wants to simulate through the terminal.

[1490] The terminal collects the user's input and sends it to the server.

[1491] Step 9: Run the simulation

[1492] The server executes a simulation in a virtual environment based on the received specific scenario.

[1493] The server simulates multiple scenarios in parallel and collects data.

[1494] Step 10: Analyze the simulation results

[1495] The server analyzes the simulation results and finds trends and patterns in the data.

[1496] The server uses statistical analysis and machine learning techniques to compile the results into an easy-to-understand format.

[1497] Step 11: View the results

[1498] The terminal visually displays the simulation results sent from the server, including graphs, charts, and reports.

[1499] The user checks the displayed results and decides on the next action to take.

[1500] Step 12: Gather feedback

[1501] The user provides feedback on the simulation results through the terminal.

[1502] The terminal sends the collected feedback to the server.

[1503] Step 13: Further simulation

[1504] The server simulates additional scenarios or retunes the model based on user feedback.

[1505] The user inputs a new scenario as necessary and runs the simulation again.

[1506] In this way, the systems work together to help users make better decisions.

[1507] Example 1

[1508] 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."

[1509] Decisions in the real world involve many risks and uncertainties, so prior simulations are required to minimize risk. However, with conventional systems, the process from collecting and integrating massive amounts of data to training AI models, running simulations, and analyzing the results is often complex and inefficient, resulting in insufficient accuracy. It is also difficult for users to efficiently input specific scenarios and quickly make decisions based on the analysis results.

[1510] 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.

[1511] In this invention, the server includes means for collecting a huge amount of data, means for preprocessing and integrating the collected data, means for training a generative AI model using the preprocessed and integrated data, means for generating a virtual environment similar to the real world using the trained generative AI model, means for executing a simulation in the generated virtual environment, means for analyzing and displaying the results of the simulation, means for a user to input a specific scenario they wish to simulate, and means for automatically generating and analyzing multiple scenarios in the virtual environment. This enables users to make quick and accurate decisions through highly accurate simulations based on a huge amount of data.

[1512] "Big data" refers to data collected in large quantities from various sources, including corporate data, personal data, and public data.

[1513] "Preprocessing" refers to the process of cleaning (filling in missing values, removing outliers, etc.) and normalizing (formatting the data) the collected data.

[1514] "Integration" is the process of transforming data collected from different sources into a consistent format so that it can be treated as a single data set.

[1515] A "generative artificial intelligence model" is a model trained using machine learning algorithms to learn patterns and trends in data and perform simulations and predictions.

[1516] "Training" refers to the process of training a generative artificial intelligence model using preprocessed and integrated data.

[1517] "Virtual environment" refers to a simulated space that resembles the real world and is generated using a trained generative artificial intelligence model.

[1518] "Simulation" refers to the process of conducting experiments and verifications in a virtual environment based on scenarios and conditions specified by the user.

[1519] "Analysis" refers to the process of finding trends and patterns in the data from the simulation results, specifically using statistical analysis and machine learning techniques.

[1520] "Display" refers to providing the analyzed simulation results to the user in a visual format (graphs, charts, reports, etc.).

[1521] A "specific scenario" refers to the specific conditions or settings that the user wants to simulate (e.g., new product characteristics, pricing, carrier selection, etc.).

[1522] "Automatically generating and analyzing multiple scenarios" refers to the process of automatically generating multiple scenarios within a virtual environment based on user input and analyzing the results of each scenario.

[1523] MODE FOR CARRYING OUT THE INVENTION

[1524] This invention relates to a system that uses a generative artificial intelligence model based on a huge amount of data to create a virtual environment and perform simulations within that virtual environment. The system aims to minimize risk by simulating real-world failures in the virtual environment in advance in order to improve the accuracy of user decision-making.

[1525] System configuration

[1526] Hardware and Software Configuration

[1527] server:

[1528] Data collection is done using APIs (e.g., Twitter API, Google Analytics API), databases (e.g., MySQL, PostgreSQL), and web scraping tools (e.g., BeautifulSoup, Scrapy).

[1529] Use data cleansing tools (e.g., Pandas) for data preprocessing and integration.

[1530] Use machine learning libraries (e.g., TensorFlow, PyTorch) to train generative artificial intelligence models.

[1531] Statistical analysis tools (e.g., R, SciPy) and data analysis platforms (e.g., Jupyter Notebook, MATLAB) will be used for simulation and result analysis.

[1532] Device:

[1533] Use visualization tools (e.g., Matplotlib, Tableau) to display the results.

[1534] Data collection and preprocessing

[1535] The servers collect vast amounts of data, including corporate, personal, and public data, from various sources, including APIs, databases, and web scraping.

[1536] Example: A server retrieves the latest tweets from the Twitter API at regular intervals and stores them in a MySQL database.

[1537] The server cleans, normalizes, and consolidates the collected data, transforming it from different sources into a consistent format.

[1538] For example, the server uses Pandas to impute missing values, remove outliers, and convert each data set into a standard format.

[1539] Training generative artificial intelligence models

[1540] Based on the preprocessed data, the server selects an appropriate machine learning algorithm and trains a generative artificial intelligence model.

[1541] Example: The server uses TensorFlow to split the dataset into training, validation, and test data, and train and evaluate the model.

[1542] Building a virtual environment

[1543] The server uses a trained generative AI model to generate a virtual environment similar to the real world, which can be flexibly changed depending on the scenario or conditions the user wants to try.

[1544] Example: The server uses a generative AI model to generate various market scenarios based on the price range and market conditions set by the user.

[1545] Running and analyzing the simulation

[1546] Users input the specific scenario they want to simulate through the device, including the features and pricing of new products, carrier selection, and so on.

[1547] Example: A user enters prices and feature settings into the device interface and sends them to the server.

[1548] The server receives the scenario sent by the user and executes the simulation in the virtual environment.

[1549] Example: The server uses a generative AI model to run parallel simulations of scenarios based on input conditions.

[1550] The server analyzes the simulation results and finds trends and patterns in the data, using statistical analysis and machine learning techniques.

[1551] Example: The server uses SciPy to perform statistical analysis of the simulation results and graph sales forecasts and market reactions.

[1552] Results presentation and decision support

[1553] The terminal presents the analysis results to the user in a visual format, for example in the form of graphs, charts, or reports.

[1554] Example: The terminal receives the analysis results from the server and displays the results as a graph using Matplotlib.

[1555] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback from the user and performs further simulations or readjusts the model as needed.

[1556] Example: The user decides on a new product launch strategy based on the results and sends that feedback to the server, which uses the feedback to recalibrate the model.

[1557] Examples of concrete examples and prompts

[1558] Case 1: A company decides to launch a new product

[1559] Example of prompt text entered by the user (corporate decision maker):

[1560] Please define the following characteristics for your new product:

[1561] Function A

[1562] Price B

[1563] Target Market C

[1564] Simulate market reactions based on these conditions and analyze projected sales and market share.

[1565] The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions.

[1566] The terminal presents the analysis results to the user in the form of graphs and reports.

[1567] The user decides on a new product launch strategy based on the simulation results.

[1568] Case 2: Individual career choices

[1569] An example of a prompt that the user might enter:

[1570] "Enter the following career choice scenario:

[1571] Occupation X

[1572] University Y

[1573] Based on these, please analyze your future annual income and job satisfaction.

[1574] The server simulates scenarios after career choices in a virtual environment and analyzes future annual income and job satisfaction.

[1575] The terminal presents the analysis results to the user as a report.

[1576] The user makes a carrier selection based on the simulation results.

[1577] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1578] Step 1: Data collection

[1579] The server uses APIs (e.g., Twitter API, Google Analytics API), databases (e.g., MySQL, PostgreSQL), and web scraping tools (e.g., BeautifulSoup, Scrapy) to collect huge amounts of data, including corporate data, personal data, and public data.

[1580] Input: API endpoint, database query, website URL.

[1581] What it does: The server periodically fetches data from the API and stores it in a MySQL database. It also uses a web scraping tool to crawl websites and retrieve the required data.

[1582] Output: The raw data collected.

[1583] Step 2: Data Preprocessing

[1584] The server pre-processes the collected data, cleaning, normalizing, and integrating it, transforming data from different sources into a consistent format.

[1585] Input: Raw data collected.

[1586] Specific operation: The server uses Pandas to impute missing values, remove outliers, and convert each data into a standard format.

[1587] Output: The preprocessed dataset.

[1588] Step 3: Training the generative artificial intelligence model

[1589] Based on the preprocessed data, the server selects an appropriate machine learning algorithm and trains a generative artificial intelligence model.

[1590] Input: The preprocessed dataset.

[1591] Specific operation: The server uses TensorFlow to split the dataset into training data, validation data, and test data, and trains and evaluates the model.

[1592] Output: A trained generative AI model.

[1593] Step 4: Build a virtual environment

[1594] The server uses a trained generative AI model to generate a virtual environment similar to the real world, which can be flexibly changed depending on the scenario or conditions the user wants to try.

[1595] Input: A trained generative AI model.

[1596] Specific operation: The server uses the generative AI model to set various parameters within the virtual environment and provide scenario templates that can be customized by the user.

[1597] Output: A customizable virtual environment.

[1598] Step 5: Enter the simulation scenario

[1599] Users input the specific scenario they want to simulate through the device, including new product features, pricing, carrier selection, etc.

[1600] Input: Scenario conditions set by the user (e.g., new product characteristics, pricing).

[1601] Specific operation: The user inputs the required information into the terminal interface and sends it to the server.

[1602] Output: User's scenario configuration data.

[1603] Step 6: Run the simulation

[1604] The server receives the scenario sent by the user and executes the simulation in the virtual environment. The simulation automatically generates and analyzes multiple scenarios based on the specified conditions.

[1605] Input: User-submitted scenario configuration data, customizable virtual environment.

[1606] Specific operation: The server uses the generative AI model to simulate multiple scenarios in parallel and record the results.

[1607] Output: Simulation result data.

[1608] Step 7: Analyze the simulation results

[1609] The server analyzes the simulation results and finds trends and patterns in the data, using statistical analysis and machine learning techniques.

[1610] Input: Simulation result data.

[1611] How it works: The server uses SciPy to statistically analyze the data and visualize the results, such as sales forecasts and market reactions.

[1612] Output: Analyzed simulation results.

[1613] Step 8: View the results

[1614] The terminal visually displays the analysis results and provides them to the user, specifically in the form of graphs, charts, and reports.

[1615] Input: Analyzed simulation results.

[1616] Specific operation: The terminal receives the analysis results sent from the server and visually displays the results using Matplotlib.

[1617] Output: The visualization data that is presented to the user.

[1618] Step 9: Decision making and feedback

[1619] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback from the user and performs further simulations or readjusts the model as needed.

[1620] Input: User decision results, feedback data.

[1621] How it works: The user decides on a strategy based on the results and enters it into the device. The server receives the feedback and retrains the model as needed.

[1622] Output: Retuned generative AI model, further simulation results.

[1623] (Application example 1)

[1624] 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."

[1625] In content distribution services, accurately predicting user viewing trends and engagement and determining optimal distribution schedules based on the results is a very difficult challenge. Conventional methods only allow for limited predictions based on past data and experience, and are unable to effectively capture viewer reactions. Therefore, new technologies are needed to optimize content distribution schedules and improve engagement.

[1626] 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.

[1627] In this invention, the server includes means for collecting a huge amount of data, means for preprocessing and integrating the collected data, means for training a generative artificial intelligence model using the preprocessed and integrated data, means for generating a virtual environment similar to the real world using the trained generative artificial intelligence model, means for running a simulation in the generated virtual environment, means for analyzing and displaying the results of the simulation, means for predicting viewer responses based on the collected data, means for predicting an engagement score for adjusting a content delivery schedule, and means for visualizing the simulation results and providing them to a user. This makes it possible to predict viewer responses with high accuracy and to formulate an optimal delivery schedule based on the predictions.

[1628] "Big data" refers to large amounts of digital information collected from a wide variety of sources.

[1629] "Preprocessing" refers to a series of steps that transform collected data into a form that is applicable for analysis and modeling.

[1630] "Integration" refers to the act of bringing together data collected from different sources into a consistent format.

[1631] A "generative artificial intelligence model" refers to a machine learning model trained on collected and preprocessed data.

[1632] A "virtual environment" refers to a digital space created to simulate real-world actions or events.

[1633] "Simulation" refers to the process of experimenting with behaviors and outcomes under specific conditions within a virtual environment.

[1634] "Analysis" refers to the process of summarizing simulation results and data trends in an easy-to-understand format.

[1635] "Display" refers to the act of visually presenting the analysis results to the user.

[1636] "Viewer response" refers to the viewer's behavior and reaction to content.

[1637] "Engagement score" refers to a numerical indicator of the level of engagement that viewers show with content.

[1638] "Content delivery schedule" refers to a plan that determines when particular content should be delivered.

[1639] A "scenario" refers to the specific conditions and settings that a user inputs to perform a simulation.

[1640] "Optimization" refers to adjusting a system or process to its best state according to a specific goal.

[1641] The system of the present invention provides a means to collect, preprocess, and integrate massive amounts of data, train generative artificial intelligence models to generate virtual environments that resemble the real world, and run simulations within those virtual environments to determine optimal content delivery schedules.

[1642] The system includes a server, a terminal, and a user.

[1643] Data collection and preprocessing

[1644] The server collects a huge amount of data using techniques such as APIs, database queries, and web scraping. The collected data includes information such as viewer viewing history, viewing times, and behavioral patterns. The server then cleans and normalizes the data, integrating it for consistency.

[1645] Training generative artificial intelligence models

[1646] Using the preprocessed and integrated data, the server trains a generative artificial intelligence model. The training dataset is divided into training, validation, and testing datasets, and the target machine learning algorithm predicts viewer viewing habits. The model is designed to predict viewer behavior and reactions, and the algorithm used for training can be a linear regression model.

[1647] Virtual environment generation and simulation

[1648] Using a trained generative artificial intelligence model, the server generates a virtual environment similar to the real world. Users can input specific scenarios via their devices, specifying specific dates and times, the number of views, likes, shares, etc. The server then runs a simulation based on these scenarios and predicts viewer reactions.

[1649] Analyzing and displaying results

[1650] The results of the simulation are analyzed by the server. The analysis results include viewer reactions and engagement scores. The device visualizes these results and provides them to the user. The user can view the results in the form of graphs, charts, and reports.

[1651] Example

[1652] For example, if a user wants to stream at 8pm with the hope of getting 5000 views, 300 likes, and 50 shares, they can enter a specific scenario. This scenario can be expressed with a prompt like this:

[1653] Based on your viewing data, predict your viewer engagement score for the following stream schedule:

[1654] Views: 5000

[1655] Likes: 300

[1656] Shares: 50

[1657] Time: 8 PM

[1658] Based on this prompt, the server runs a simulation in the virtual environment to predict the viewer's reaction. The prediction results are provided to the user via their device, and the user can then decide on the optimal distribution schedule.

[1659] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1660] Step 1:

[1661] The server collects a huge amount of data. The types of data include viewer viewing history, viewing times, and behavioral patterns. This data is obtained using APIs, database queries, and web scraping techniques. Specifically, data is obtained by issuing queries to a viewing history database. The input in this step is the viewer data source, and the output is raw viewer data.

[1662] Step 2:

[1663] The server preprocesses and integrates the collected data. It removes noise from the data, imputes missing values, and converts various data sources into a consistent format. This step includes cleaning and normalizing the data. Specifically, it uses data cleaning tools to handle missing values. The input to this step is the collected data from step 1, and the output is the preprocessed and integrated data.

[1664] Step 3:

[1665] The server trains a generative artificial intelligence model using the preprocessed and integrated data. Specifically, it splits the dataset into training, validation, and test sets, and trains the model using a machine learning algorithm such as a linear regression model. The input in this step is the preprocessed data from step 2, and the output is a trained AI model.

[1666] Step 4:

[1667] The server generates a virtual environment using a trained generative AI model. The virtual environment is a digital space for simulating viewer behavior. Various scenarios and variables can be set, enabling various simulations. The input in this step is the trained AI model, and the output is the virtual environment.

[1668] Step 5:

[1669] Through the terminal, the user inputs the specific scenario they want to simulate. The scenario includes parameters such as a specific date and time, expected number of viewers, number of likes, number of shares, etc. The input in this step is the user's scenario information, and the output is the scenario data sent to the server.

[1670] Step 6:

[1671] The server runs a simulation in the virtual environment based on the input scenario. Specifically, it uses the model to generate data to predict viewer reactions and then runs the simulation. The input in this step is the scenario data from step 5, and the output is the simulation results.

[1672] Step 7:

[1673] The server analyzes the simulation results. The analysis aims to predict viewer responses and engagement scores. This procedure includes analyzing data trends using statistical analysis and machine learning techniques. The input in this step is the simulation results, and the output is the analysis results.

[1674] Step 8:

[1675] The device visualizes the analysis results and provides them to the user. Specifically, the analysis results are displayed in the form of graphs, charts, and reports. The user views these results and determines the optimal content distribution schedule based on the engagement scores. The input in this step is the analysis results, and the output is the visualized analysis results provided to the user.

[1676] By following the steps above, it becomes possible to predict viewer reactions with high accuracy and formulate an optimal distribution schedule.

[1677] 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.

[1678] The present invention is a system that collects massive amounts of data, integrates and preprocesses them, trains a generative artificial intelligence model, and runs simulations in a virtual environment using the generated model. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to support more accurate decision-making that takes into account the user's emotional state. The present invention is implemented using the following processing steps and configuration.

[1679] System configuration and operation

[1680] 1. Data collection and integration

[1681] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data, and uses techniques such as APIs, database queries, and web scraping to efficiently retrieve the data.

[1682] The server performs preprocessing on the collected data, such as noise removal, missing value filling, and data standardization, to ensure consistency.

[1683] 2. Training a generative AI model

[1684] The server uses the preprocessed and integrated data to train a generative artificial intelligence model, selecting an appropriate machine learning algorithm and splitting the dataset into training, validation, and testing sections.

[1685] Evaluate the performance of the model and adjust the hyperparameters as necessary.

[1686] 3. Creating a virtual environment

[1687] The server uses a trained generative artificial intelligence model to generate a virtual environment similar to the real world, with the flexibility to recreate a variety of scenarios.

[1688] 4. Enter the scenario

[1689] The user inputs the specific scenario they want to simulate through the terminal, including, for example, the characteristics and pricing of a new product, carrier selection, etc.

[1690] The terminal transmits the input scenario to the server.

[1691] 5. Collecting Emotion Data Using an Emotion Engine

[1692] The device collects emotional data from the user's facial expressions, voice, input, etc.

[1693] The emotion engine analyzes the collected emotion data and recognizes the user's emotional state in real time.

[1694] 6. Running the Simulation

[1695] The server runs a simulation in the virtual environment based on the received scenario and emotion data, dynamically adjusting the scenario based on the emotion data to provide more realistic results.

[1696] 7. Analyzing and displaying simulation results

[1697] The server analyzes the simulation results and uses statistical analysis and machine learning techniques to summarize the results in an easy-to-understand format.

[1698] The terminal provides the user with a visual display of the analysis results, including graphs, charts, and reports.

[1699] 8. Decision support and feedback gathering

[1700] The user makes optimal decisions based on the displayed simulation results, and the server collects feedback from the user and performs additional simulations or refines the model.

[1701] Specific examples

[1702] Case 1: A company decides to launch a new product

[1703] Users (corporate decision makers) input scenarios such as new product characteristics and pricing into the terminal, and their emotional state (e.g., excitement, anxiety, etc.) is collected at the same time.

[1704] The emotion engine analyzes the user's emotional state in real time and transmits it to the server.

[1705] The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions taking into account emotional data.

[1706] The terminal presents the analysis results to the user in the form of graphs and reports.

[1707] The user decides on a new product launch strategy based on the simulation results.

[1708] Case 2: Individual career choices

[1709] The user inputs a scenario regarding career choices (such as the school to attend, the occupation to choose, etc.) into the terminal, and their emotional state (such as anxiety, expectations, etc.) is collected at the same time.

[1710] The emotion engine analyzes the user's emotional state and transmits it to the server.

[1711] The server simulates scenarios after career selection and analyzes future annual income and job satisfaction taking into account emotional data.

[1712] The terminal presents the analysis results to the user as a report.

[1713] The user selects a carrier based on the simulation results.

[1714] In this way, the system of the present invention, which is combined with an emotion engine, provides highly accurate simulation and decision support that takes into account the user's emotional state, helping the user make optimal choices.

[1715] The processing flow will be explained below.

[1716] Step 1: Data collection

[1717] The server retrieves data from corporate databases, public data services, and public data sources on the Internet, using techniques such as APIs, database queries, and web scraping to efficiently gather data.

[1718] The server periodically updates the data to keep it up to date.

[1719] Step 2: Data Preprocessing

[1720] The server removes noise from the acquired data and fills in missing values. Specifically, it interpolates the average value of missing data and removes outliers.

[1721] The server performs data standardization (eg, normalization, scaling) and converts data from different sources into a consistent format.

[1722] Step 3: Data Integration

[1723] The server consolidates the pre-processed data into one unified database.

[1724] The server maps the schema of the data from different sources and makes it consistent.

[1725] Step 4: Prepare the dataset

[1726] The server extracts training, validation, and test datasets from the integrated database.

[1727] The server labels the data and performs feature engineering to prepare it in a format that is easy for the model to learn.

[1728] Step 5: Training the generative AI model

[1729] The server selects an appropriate generative AI algorithm (e.g., a deep learning model) and trains the model using a training dataset.

[1730] The server monitors the training process and optimizes hyperparameters as needed.

[1731] Step 6: Evaluate the generative AI model

[1732] The server evaluates the model's performance using a validation dataset, using metrics such as precision, recall, and F1 score.

[1733] The server will make any necessary improvements based on the evaluation results.

[1734] Step 7: Generate a Virtual Environment

[1735] The server uses a trained generative AI model to generate a virtual environment that resembles the real world.

[1736] The server sets the variables and parameters to be simulated within the virtual environment.

[1737] Step 8: Entering the Scenario

[1738] The user inputs the specific scenario he or she wants to simulate (for example, the characteristics and pricing of a new product, carrier selection, etc.) through the terminal.

[1739] The terminal transmits the user's input to the server.

[1740] Step 9: Collect emotion data

[1741] The device collects emotional data from the user's facial expressions, voice, input, etc. This is typically done using devices such as a camera and microphone.

[1742] The emotion engine analyzes the collected data and recognizes the user's emotional state in real time.

[1743] Step 10: Run the simulation

[1744] The server executes a simulation in the virtual environment based on the scenario received from the user and the emotion data from the emotion engine.

[1745] The server dynamically adjusts the simulation to produce results that reflect real-world emotional states.

[1746] Step 11: Analyze the simulation results

[1747] The server collects and analyzes the simulation results, using statistical analysis and machine learning techniques to extract trends and patterns from the results.

[1748] The server summarizes the analysis results in an easy-to-understand format (e.g., graphs, charts, reports).

[1749] Step 12: View the results

[1750] The terminal visually displays the simulation results sent from the server, presenting the results using visual elements (e.g., graphs, charts, reports) so that the user can easily understand them.

[1751] The user checks the results displayed on the terminal and makes a decision.

[1752] Step 13: Gather feedback

[1753] The user provides feedback on the simulation results and the displayed content through the terminal.

[1754] The terminal sends the collected feedback to the server.

[1755] Step 14: Further simulation

[1756] The server simulates additional scenarios or retunes the model based on user feedback.

[1757] The user inputs a new scenario as needed, and the server executes a re-simulation based on that.

[1758] In this way, the systems work together to help users make better, more emotionally informed decisions.

[1759] Example 2

[1760] 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."

[1761] Currently, there are systems that utilize a large amount of data to provide highly accurate decision-making support, but few of them take into account the user's emotional state. As a result, conventional systems are unable to fully reflect the impact of the user's emotions on decision-making, making it difficult to make optimal decisions.

[1762] The specific processing by the specific processing unit 290 of the data processing device 12 in the second embodiment is realized by the following means.

[1763] In this invention, the server includes means for collecting a huge amount of data, means for preprocessing and integrating the collected data, means for training a generative AI model using the preprocessed and integrated data, means for generating a virtual environment similar to the real world using the trained generative AI model, means for analyzing the collected emotional data and recognizing a user's emotional state in real time, means for dynamically adjusting a simulation scenario based on the emotional data, and means for analyzing and displaying the results of the simulation, thereby enabling highly accurate simulation and decision-making support that takes the user's emotional state into account.

[1764] "Big data" refers to a large and diverse collection of information, collected from sources such as corporate data, personal data, and public data.

[1765] "Preprocessing" refers to the process of carrying out procedures such as noise removal, missing value filling, and data standardization on collected data to make the data consistent.

[1766] "Synthesis" is the process of combining pre-processed data into one unified data set.

[1767] "Training a generative artificial intelligence model" is the process of applying machine learning algorithms to preprocessed and integrated data to train the model for optimal performance.

[1768] A "virtual environment" is an artificial environment that uses a generative artificial intelligence model to create a simulated environment similar to the real world.

[1769] "Emotional data" refers to information collected from the user's facial expressions, voice, input content, etc., that indicates the user's emotional state in real time.

[1770] "Analyzing emotional data" refers to the process of recognizing the user's emotional state based on the collected emotional data, and detecting classifications or specific emotional states as needed.

[1771] "Dynamic adjustment of the simulation scenario" refers to the process of appropriately changing the content of the scenario based on the user's emotional state during the simulation, in order to bring the results closer to reality.

[1772] "Displaying the simulation results" refers to the process of visually expressing the data obtained from the simulation and presenting it in a form that is easy for the user to understand.

[1773] MODE FOR CARRYING OUT THE INVENTION

[1774] The present invention is a system that collects massive amounts of data, integrates and preprocesses them, trains a generative artificial intelligence model, and runs simulations in a virtual environment using the generated model. Furthermore, by combining it with an emotion engine that recognizes the user's emotions, it is possible to support more accurate decision-making that takes into account the user's emotional state. This system is configured as follows:

[1775] 1. Data collection and integration

[1776] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data. Collection methods include APIs, database queries, and web scraping tools (e.g., Beautiful Soup and Scrapy). The collected data is preprocessed using the Python Pandas library, which removes noise, fills in missing values, and standardizes the data to ensure consistency.

[1777] 2. Training a generative AI model

[1778] The server trains a generative artificial intelligence model using the preprocessed and integrated data. It uses machine learning algorithms such as TensorFlow and PyTorch to split the dataset into training, validation, and testing sets. It also uses Optuna and Grid Search to tune the model's hyperparameters and optimize accuracy.

[1779] 3. Creating a virtual environment

[1780] The server uses the trained generative AI model to generate a virtual environment similar to the real world, using a virtual environment generation tool (e.g., Unity or Unreal Engine) to recreate the scenario required for the simulation.

[1781] 4. Enter the scenario

[1782] The user inputs a specific simulation scenario through the terminal. This input process involves using a web application form or an interactive chat window to enter details such as new product features, pricing, and carrier selection. The terminal then sends the input scenario data to the server via an HTTP request.

[1783] 5. Emotional Data Collection and Analysis Using an Emotional Engine

[1784] The device collects emotion data from the user's facial expressions, voice, and input. For example, it captures facial expressions with a camera, analyzes them using OpenCV, and converts the speech into text using a speech recognition engine (e.g., Google Speech-to-Text). The emotion engine analyzes the collected emotion data and uses a deep learning model to classify and recognize emotions into multiple categories, such as "happiness," "sadness," and "excitement."

[1785] 6. Running the Simulation

[1786] The server executes a simulation based on the received scenario and emotion data. The scenario is dynamically adjusted based on the emotion data, and the simulation is performed in real time within the virtual environment. For example, if a user wants to predict market reaction to a new product, they can input the characteristics and pricing of the new product, and if the emotion engine interprets this as "excitement," it will affect the simulation results.

[1787] 7. Analyzing and displaying simulation results

[1788] The server analyzes the simulation results and uses statistical analysis and machine learning techniques to compile the results into easy-to-understand formats, such as Pandas and Matplotlib, to convert the data into graphs and charts. The terminal visually displays these results and provides them to the user in an easy-to-understand format.

[1789] 8. Decision support and feedback gathering

[1790] The user makes optimal decisions based on the displayed simulation results. The server collects feedback from the user and uses it to conduct additional simulations and refine the model, thereby enabling more accurate decision-making support.

[1791] Specific examples

[1792] Case 1: A company decides to launch a new product

[1793] 1. The user (corporate decision maker) inputs scenarios such as new product characteristics and pricing into the terminal, and their emotional state (e.g., excitement) is collected at the same time.

[1794] 2. The emotion engine analyzes the user's emotional state in real time and sends it to the server.

[1795] 3. The server simulates market reactions to new products in a virtual environment and analyzes sales forecasts and market reactions taking into account emotional data.

[1796] 4. The terminal presents the analysis results to the user in the form of graphs and reports.

[1797] 5. The user decides on a new product launch strategy based on the simulation results.

[1798] Case 2: Individual career choices

[1799] 1. The user inputs a scenario about career choices (such as the school they will attend, the occupation they will choose, etc.) into the terminal, and their emotional state (e.g., anxiety, expectation) is collected at the same time.

[1800] 2. The emotion engine analyzes the user's emotional state and sends it to the server.

[1801] 3. The server simulates scenarios after career selection and analyzes future annual income and job satisfaction taking into account emotional data.

[1802] 4. The device presents the analysis results to the user as a report.

[1803] 5. The user makes a carrier selection based on the simulation results.

[1804] Prompt Sentence Examples

[1805] If you want to predict market reaction to a new product:

[1806] Predict the market reaction to a newly developed smartphone. The characteristics are 5G, a 6.5-inch display, a triple camera, and a price of $800. The emotional state is "Excited."

[1807] If you would like to discuss your career options:

[1808] I'm looking for help choosing a future career. My options are to go to college for engineering or to go to design school. My current emotional state is a mixture of anxiety and excitement.

[1809] In this way, the system of the present invention, which is combined with an emotion engine, provides highly accurate simulation and decision support that takes into account the user's emotional state, helping the user make optimal choices.

[1810] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1811] Step 1:

[1812] The server collects data from various sources, including corporate data, personal data, and public data. It uses APIs, database queries, and web scraping tools (such as Beautiful Soup and Scrapy) as input. It saves the collected data in JSON or CSV format as output. Specifically, it periodically accesses a specified API endpoint, retrieves the latest data, and saves it in a local database.

[1813] Step 2:

[1814] The server preprocesses and standardizes the collected data. It receives the data collected in the previous step as input. It denoises the data, fills missing values, and standardizes the data. For example, it uses the Python Pandas library to generate a data frame, removes inaccurate values, and fills missing values ​​with the median. As output, it obtains a preprocessed, clean dataset.

[1815] Step 3:

[1816] The server trains a generative AI model using the preprocessed data. It receives the preprocessed data as input and trains the model using a machine learning algorithm (e.g., TensorFlow or PyTorch). It splits the dataset into training, validation, and test datasets and feeds each dataset into the model. It uses Optuna or Grid Search to tune the hyperparameters. As output, it obtains a trained generative AI model.

[1817] Step 4:

[1818] The server generates a virtual environment using a trained generative AI model. It receives the trained model and scenario data as input. It uses a virtual environment generation tool (such as Unity or Unreal Engine) to create an environment similar to the real world. As output, it obtains a virtual environment for simulation.

[1819] Step 5:

[1820] The user inputs a specific simulation scenario through a terminal. For input, the scenario (such as the characteristics and pricing of a new product) is entered using a form or chat window in the web application. For output, the scenario data is sent to the server.

[1821] Step 6:

[1822] The device collects the user's emotional data. As input, the user's facial expressions and voice are captured using a camera and microphone. OpenCV and Google Speech-to-Text are used to analyze the emotional data. The analyzed emotional data is obtained as output.

[1823] Step 7:

[1824] The emotion engine analyzes collected emotion data in real time to recognize the user's emotional state. It receives data from the camera and microphone as input and applies emotion recognition algorithms. The output is classified emotion data (e.g., joy, sadness, excitement).

[1825] Step 8:

[1826] The server executes a simulation using the received scenario and emotion data. It receives the scenario data and emotion data as input. It performs a simulation in the virtual environment while dynamically adjusting the scenario based on the emotion data. It obtains the simulation results as output.

[1827] Step 9:

[1828] The server analyzes the simulation results and generates data for visual display. It receives the simulation results as input and organizes them into an easy-to-understand format using statistical analysis and machine learning techniques. The output is analysis data, graphs, and charts.

[1829] Step 10:

[1830] The terminal visually displays the analysis results and provides them to the user. It receives the analysis data as input. It uses a web interface to display the results in the form of interactive graphs and reports. It outputs the results in a format that is easy for the user to understand.

[1831] Step 11:

[1832] The user makes optimal decisions based on the displayed simulation results. The analysis results are used as input. The information necessary for decision-making is collected and reflected in actual actions. For example, decisions are made regarding new product launch strategies or career choices. The user's decision is obtained as output.

[1833] Step 12:

[1834] The server collects user feedback and uses it to refine the model and conduct additional simulations. It receives user feedback as input, updates the model based on it, and retrains it to improve its accuracy. The output is an improved generative AI model.

[1835] (Application example 2)

[1836] 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."

[1837] Conventional autonomous driving systems control operation based on external environmental data, but do not take into account the emotional state of passengers, which means they are unable to fully alleviate the anxiety and stress felt by passengers. Furthermore, due to the lack of technology to collect and analyze emotional data and dynamically adjust driving modes, there is an issue of not being able to improve passenger safety and comfort.

[1838] 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 means for collecting a huge amount of data, means for preprocessing and integrating the collected data, and means for training a generative artificial intelligence model using the preprocessed and integrated data. This enables the operation of an autonomous vehicle that improves safety and comfort by analyzing passenger emotional states in real time and reflecting this in driving control.

[1839] "Big data" refers to data collected in large quantities from various sources, including corporate data, personal data, and public data.

[1840] "Preprocessing" refers to the process of performing operations such as noise removal, missing value filling, and data standardization on collected data to improve consistency and quality.

[1841] "Integration" is the process of bringing together data collected from different data sources, making it consistent and easier to process and analyze.

[1842] A "generative artificial intelligence model" is a model trained using machine learning algorithms that is generated to perform specific tasks based on large amounts of data.

[1843] A "virtual environment" is a simulation environment that imitates the real world, and is an environment that reproduces various situations based on scenarios and conditions specified by the user.

[1844] "Simulation" is a technique for reproducing phenomena based on specific scenarios and conditions in a virtual environment and analyzing the results.

[1845] The "emotion engine" is a system that recognizes a user's emotional state in real time by collecting and analyzing emotional data from the user's facial expressions, voice, input content, etc.

[1846] "Dynamic adjustment" refers to the act of changing and optimizing system behavior and settings on the fly based on real-time information such as emotional data.

[1847] "Analysis" refers to the process of compiling collected data and simulation results into an easy-to-understand format and extracting the meaning and trends of the data using statistical analysis and machine learning techniques.

[1848] To specifically implement this invention, the system configuration and operating procedures are as follows: The system collects, preprocesses, and integrates massive amounts of data, trains a generative AI model based on the data, and uses the generated model to run a simulation in a virtual environment similar to the real world. Furthermore, the system analyzes the user's emotional state in real time and dynamically adjusts the simulation results to support more accurate decision-making.

[1849] Components and their operation

[1850] Data collection and preprocessing

[1851] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data, using techniques such as APIs, database queries, and web scraping. The collected data undergoes preprocessing, such as noise removal, missing value filling, and data standardization, to ensure consistency.

[1852] Training generative artificial intelligence models

[1853] The server trains generative AI models using the preprocessed and integrated data. Machine learning algorithms are implemented using Scikit-learn, TensorFlow, Keras, etc., and datasets are divided into training, validation, and testing phases. Model performance is also evaluated and hyperparameters are adjusted.

[1854] Creating a virtual environment and inputting a scenario

[1855] Using a trained generative artificial intelligence model, a virtual environment similar to the real world is generated. The user inputs the specific scenario they want to simulate through their device. The device then sends the input scenario to the server, which then starts the simulation in the virtual environment.

[1856] Analysis by emotion engine

[1857] The device collects emotional data from the user's facial expressions, voice, input, etc. The emotion engine, built using Keras and other tools, analyzes the collected emotional data, recognizes the user's emotional state in real time, and transmits the data to the server.

[1858] Running a simulation and displaying the results

[1859] The server runs a simulation in a virtual environment based on the received scenario and emotional data. It dynamically adjusts the simulation results taking into account the emotional data. The server analyzes the simulation results and visually displays them using statistical analysis and machine learning techniques. The terminal presents the analysis results to the user in the form of graphs and reports.

[1860] Specific examples

[1861] As a concrete example, consider a case where a user of an autonomous vehicle feels anxious during their morning commute due to a traffic jam. The emotion engine recognizes this anxiety and adjusts the autonomous vehicle to slow down a little and select a safe route.

[1862] Prompt Sentence Examples

[1863] An example of a prompt to enter an example into the system is:

[1864] "A user is using an autonomous vehicle during their morning commute. They encounter a traffic jam along the way, which makes them feel anxious. The emotion engine recognizes this feeling of anxiety, and the autonomous vehicle adjusts to slow down and choose a safer route."

[1865] A system configured in this way realizes highly accurate simulation and decision-making support that takes into account the user's emotional state.

[1866] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1867] Step 1: Data collection

[1868] The server collects huge amounts of data from various sources, including corporate data, personal data, and public data, using techniques such as APIs, database queries, and web scraping. The input is data provided by each source, and the output is a huge amount of raw data.

[1869] Step 2: Data preprocessing and integration

[1870] The server performs preprocessing on the collected data, such as noise removal, missing value filling, and data standardization, to make it consistent. Specifically, it uses data cleaning algorithms, interpolates missing values, and standardizes the data. The input is raw data, and the output is preprocessed and integrated data.

[1871] Step 3: Training the generative artificial intelligence model

[1872] The server uses the preprocessed and integrated data to train a generative AI model. Specifically, it uses machine learning frameworks such as Scikit-learn, TensorFlow, and Keras to divide the dataset into training, validation, and testing sections, and selects an appropriate algorithm for training. The input is the preprocessed data, and the output is a trained AI model.

[1873] Step 4: Create a virtual environment

[1874] The server uses a trained artificial intelligence model to generate a virtual environment similar to the real world. Specifically, it uses 3D simulation software to build the environment based on the information the model has learned. The input is the trained model, and the output is the virtual environment.

[1875] Step 5: Entering the scenario

[1876] The user inputs the specific scenario they want to simulate through their device. The input scenario can include forecasts of market reactions to new products or carrier selection. The input is the scenario information provided by the user, and the output is the data sent from the device that receives it to the server.

[1877] Step 6: Collecting Emotion Data with the Emotion Engine

[1878] The device collects emotion data from the user's facial expressions, voice, input content, etc. The emotion engine analyzes emotions using models such as Keras. The input is the user's facial expressions and voice data, and the output is analyzed emotional state data.

[1879] Step 7: Run the simulation

[1880] The server executes a simulation in the virtual environment based on the received scenario and emotional data. Specifically, it dynamically adjusts simulation parameters based on the scenario and emotional data to generate results. The input is the user's scenario and emotional state data, and the output is the simulation results.

[1881] Step 8: Analyze and display simulation results

[1882] The server visually analyzes the simulation results and displays them in the form of graphs and reports. Specifically, it uses statistical analysis and machine learning techniques to summarize the results in an easy-to-understand format. The input is the simulation results, and the output is the analyzed results in graphs and reports.

[1883] Step 9: Support decision making and gather feedback

[1884] The user makes optimal decisions based on the displayed simulation results and sends feedback from the device to the server, which then performs further simulations and adjusts the model based on the feedback. The input is the user's feedback, and the output is the adjusted simulation results and model.

[1885] 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.

[1886] 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.

[1887] 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.

[1888] 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.

[1889] 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.

[1890] 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.

[1891] 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).

[1892] 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.

[1893] 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."

[1894] 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.

[1895] 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).

[1896] 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.

[1897] 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.

[1898] 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.

[1899] 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.

[1900] 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 ...

Claims

1. A means of collecting large amounts of data, a means of preprocessing and integrating the collected data; means for training a generative artificial intelligence model using the preprocessed and integrated data; means for generating a virtual environment similar to the real world using a trained generative artificial intelligence model; means for executing a simulation within the generated virtual environment; means for analyzing and displaying the results of the simulation; A system including:

2. a means for the user to input the specific scenario they wish to simulate; means for executing a simulation in a virtual environment based on the input specific scenario; The system of claim 1 , comprising:

3. A means of providing support for optimal decision-making based on the results of simulations; means for collecting feedback on said decision making; means for performing a further simulation based on said feedback; The system of claim 1 , comprising:

Images