System

Abstract

To provide a system.SOLUTION: A system includes a means for constructing a virtual brain model that imitates a decision-making style of a manager by using a trained generative AI, means for enhancing the precision of the model by repeating a step of a plan, execution, evaluation and improvement, means for training the model by using past decision-making data of the manager, means for recognizing feelings of the manager, support means for supporting decision-making by the model on the basis of the feelings, and means for improving the model by using evaluation of the manager to proposals and the decision-making of the model and repeating.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] As business becomes more diverse, the number of decision-making opportunities for managers of large corporations increases, and the burden on them grows. However, it is difficult, both in terms of time and mentally, for managers to make all decisions themselves. [Means for solving the problem]

[0005] The system disclosed herein includes: a means for using a trained generative AI to construct a virtual brain model that mimics the decision-making style of a manager; a means for improving the accuracy of the model by repeating steps of planning, executing, evaluating, and improving; a means for training the model using past decision-making data of the manager; a means for recognizing the manager's emotions; a support means for supporting decision-making by the model based on the emotions; and a means for improving the model using the manager's evaluation and reaction to the model's proposals and decision-making. [Brief explanation of the drawings]

[0006] [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. 2 is a sequence diagram showing a flow of processing in the data processing system according to the first embodiment of the first form example. [Figure 12] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Embodiment 1. [Figure 13] FIG. 10 is a sequence diagram showing a processing flow of a data processing system in a second embodiment of the second form example. [Figure 14] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Embodiment Example 2. [Figure 15] FIG. 10 is a sequence diagram showing the flow of processing in a data processing system according to a third embodiment of the third embodiment. [Figure 16] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Embodiment 3. [Figure 17] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the first embodiment of the first form example when an emotion engine is combined. [Figure 18] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 1 of Form Example 1 when an emotion engine is combined. [Figure 19] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the second embodiment of the second form example when an emotion engine is combined. [Figure 20] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 2 of Form Example 2 when an emotion engine is combined. [Figure 21] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in the third embodiment of the third form example when an emotion engine is combined. [Figure 22] FIG. 10 is a sequence diagram showing the flow of processing in the data processing system in Application Example 3 of Form Example 3 when an emotion engine is combined. DETAILED DESCRIPTION OF THE INVENTION

[0007] 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.

[0008] First, the terms used in the following description will be explained.

[0009] 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, the 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), an APU (Accelerated Processing Unit), or a TPU (TENSOR PROCESSING UNIT (registered trademark)).

[0010] 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.

[0011] 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.

[0012] 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.

[0013] 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."

[0014] [First embodiment]

[0015] FIG. 1 shows an example of the configuration of a data processing system 10 according to the first embodiment.

[0016] 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.

[0017] 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).

[0018] 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.

[0019] 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.

[0020] 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.

[0021] 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.

[0022] FIG. 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0023] 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.

[0024] 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.

[0025] 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.

[0026] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0027] "Example 1"

[0028] In one embodiment of the present invention, the means for constructing a virtual brain model using generative AI includes an AI algorithm for learning the decision-making patterns of managers. Specifically, data on past decisions made by managers is used as learning data, and the AI ​​learns those patterns.

[0029] "Example 2"

[0030] Furthermore, to speed up the PDCA cycle and improve the accuracy of the model, AI periodically updates its learning data and learns new decision-making patterns, enabling it to make decisions that respond to changes in the business environment.

[0031] "Example 3"

[0032] Furthermore, as a means of learning and applying data obtained from an individual's brain, a personalized virtual brain is constructed by acquiring the manager's brainwave data and behavioral data and using this as learning data for the AI. This makes it possible to construct a virtual brain that reflects the individual decision-making style of each manager. As a specific example, the AI ​​can learn data on how a manager made decisions in specific business situations, and the virtual brain can make decisions based on that data.

[0033] The processing flow of each embodiment will be described below.

[0034] "Example 1"

[0035] Step 1: Collect data on management's past decisions. This includes information on the results of decisions made by management in the past and the circumstances at the time.

[0036] Step 2: The collected data is input into the generative AI, which then learns the manager's decision-making patterns. This learning process builds a model for the AI ​​to imitate the manager's decision-making style.

[0037] Step 3: Use the constructed model to predict managerial decisions for new business situations.

[0038] "Example 2"

[0039] Step 1: The AI ​​periodically updates its learning data by adding newly acquired managerial decision-making data to the learning data.

[0040] Step 2: Using the updated learning data, the AI ​​learns new decision-making patterns. Step 3: Using the new decision-making patterns it has learned, it predicts decisions that respond to changes in the business environment.

[0041] "Example 3"

[0042] Step 1: Obtain the manager's brainwave data, behavioral data, etc. These data are used to reflect the manager's individual decision-making style.

[0043] Step 2: The acquired data is used as training data for the AI, which then learns the decision-making styles of individual managers.

[0044] Step 3: Build a personalized virtual brain using the learned decision-making styles of individual managers.

[0045] Example 1

[0046] Next, a description will be given of Example 1 of Form 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."

[0047] Conventional decision support systems lacked personalization and were unable to fully reflect the individual decision-making patterns of managers. Furthermore, data collection, preprocessing, model training, and evaluation were often performed manually, resulting in inefficiency and low accuracy. Furthermore, generating and inputting prompts was cumbersome, and the output of prediction results was not intuitive, making it difficult for managers to make quick decisions.

[0048] 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.

[0049] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for collecting past decision-making data and storing it in a database, means for preprocessing the collected data, means for training the AI ​​model using the preprocessed data, means for evaluating the trained model, means for generating prompt sentences and inputting them into the AI ​​model, and means for outputting prediction results.This allows for the construction of a personalized virtual brain that reflects the individual decision-making patterns of a manager, enabling efficient and highly accurate decision-making support.

[0050] The server may include means for using a trained generative AI to construct a virtual brain model that imitates the decision-making style of the manager, means for improving the accuracy of the model by repeating steps of planning, executing, evaluating, and improving, means for training the model using past decision-making data of the manager, means for recognizing the manager's emotions, support means for supporting decision-making by the model based on the emotions, and means for improving the model using the manager's evaluation and reaction to the model's proposals and decision-making.

[0051] When a machine failure occurs in the factory, the support means may propose a repair procedure for the failed machine based on data relating to past failure cases.

[0052] The support means may optimize the factory production line using the model by learning the decision-making patterns of the manager.

[0053] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new data and models.

[0054] A "virtual brain" is a system that uses artificial intelligence technology to mimic the functions of the human brain and make decisions.

[0055] "PDCA" refers to the cycle of planning (Plan), doing (Do), checking (Check), and improving (Act), and is a method for making continuous improvements.

[0056] "Personalization" refers to optimizing systems and services based on the characteristics and behavioral patterns of individual users.

[0057] A "database" is a system for efficiently storing, managing, and retrieving data.

[0058] "Preprocessing" is the process of cleaning and transforming data before data analysis or machine learning.

[0059] "Training" is the process by which a machine learning model learns patterns from data.

[0060] "Evaluation" is the process of measuring the performance of a trained model using validation data.

[0061] A "prompt" is an instruction given to a generative AI model, containing instructions for performing a specific task.

[0062] "Prediction result" refers to the result that a generative AI model outputs based on a prompt, including predictions of future events or data.

[0063] This invention is a system for constructing a virtual brain model that uses generative AI to support managerial decision-making. Specific embodiments of this system are described below.

[0064] First, the server collects data on past decisions made by management, including sales data, market trends, economic indicators, competitor trends, etc. The server stores this data in a database management system (e.g., MySQL (registered trademark)).

[0065] The server then preprocesses the collected data, specifically using the Python Pandas library to impute missing values, normalize the data, and detect and remove outliers, thereby improving the quality of the data.

[0066] The server then uses the preprocessed data to train an AI model, using a machine learning framework (e.g., TENSORFLOW®) to learn the decision-making patterns of managers. Training uses past decision-making data as input so the model can recognize patterns.

[0067] After training is complete, the server evaluates the trained model using a validation dataset to measure the model's precision and recall, allowing the server to verify the model's performance and make adjustments if necessary.

[0068] The user inputs a prompt to the generative AI model. For example, the user inputs a prompt such as, "Please predict sales for the next quarter based on sales data from the past five years and market trends." The input prompt is sent to the server.

[0069] The server runs the AI ​​model based on the input prompt and generates a prediction result. The prediction result is displayed to the user. The user can use this prediction result as a reference to develop future management strategies.

[0070] For example, data on a manager's past decisions may include the following:

[0071] Sales Data

[0072] market trends

[0073] Economic indicators

[0074] Competitor trends

[0075] The server collects this data and inputs it into the AI ​​model. For example, the following prompt sentence is input into the generative AI model:

[0076] "Please provide a sales forecast for the next quarter based on sales data from the past five years and market trends."

[0077] Based on this prompt, the AI ​​model analyzes past data and provides a sales forecast for the next quarter, which users can use to plan future business strategies.

[0078] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0079] Step 1: Data collection

[0080] The server collects data on past decisions made by management. Specifically, it obtains data such as sales data, market trends, economic indicators, and competitor trends using APIs and web scraping. This data is saved as a CSV file and imported into a database management system (e.g., MySQL). The input is raw data from various data sources, and the output is structured data stored in the database.

[0081] Step 2: Data Preprocessing

[0082] The server preprocesses the collected data. Specifically, it uses the Python Pandas library to impute missing values, normalize the data, and detect and remove outliers. For example, missing values ​​are imputed with the mean value, and data normalization uses min-max scaling. The input is the raw data stored in the database, and the output is the preprocessed, clean data.

[0083] Step 3: Train the model

[0084] The server trains an AI model using the preprocessed data. It uses a machine learning framework (e.g., TensorFlow) to build a neural network model. It processes the training data in batches and updates the model weights every epoch. The input is the preprocessed data, and the output is the trained AI model.

[0085] Step 4: Evaluate the model

[0086] The server evaluates the trained model. For evaluation, a validation dataset is used to measure the model's precision and recall. Evaluation metrics include mean absolute error (MAE) and coefficient of determination (R²). The input is the trained AI model and the validation dataset, and the output is the evaluation result.

[0087] Step 5: Generate and enter a prompt

[0088] The user inputs a prompt to the generative AI model. For example, the prompt might be, "Please predict sales for the next quarter based on sales data from the past five years and market trends." The input is the prompt entered by the user, and the output is the prompt sent to the server.

[0089] Step 6: Output prediction results

[0090] The server runs the AI ​​model based on the input prompt and generates a prediction result. The prediction result is displayed in graph or table format on a web interface. The input is the prompt and the trained AI model, and the output is the prediction result. Users can use this prediction result as a reference to develop future management strategies.

[0091] (Application example 1)

[0092] Next, a description will be given of Application Example 1 of Embodiment 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."

[0093] With conventional production management systems, it was difficult to mimic management decisions in real time and perform optimal production management. Furthermore, there was a lack of means to effectively utilize management's past decision-making data, making it impossible to achieve efficient production management. Furthermore, systems for collecting and analyzing data in real time were complex, resulting in significant costs for implementation and operation.

[0094] 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.

[0095] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from individual brains to construct a personalized virtual brain, means for collecting and storing data from production lines in factories in real time, means for preprocessing the collected data and training the AI ​​model, and means for analyzing production line data in real time using the trained AI model to make optimal decisions. This enables efficient production management by imitating management decision-making in real time.

[0096] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new data and models.

[0097] A "virtual brain" is a model that uses artificial intelligence technology to mimic the decision-making process of the human brain.

[0098] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0099] "Personalization" refers to the individual optimization of systems and services based on the characteristics and behavior of individual users.

[0100] A "production line" refers to a series of machines or work stations arranged to efficiently produce a product.

[0101] "Real-time data collection" refers to obtaining data immediately from ongoing processes or events.

[0102] "Data preprocessing" refers to processing collected data to convert it into a format suitable for analysis and model training.

[0103] "Training an AI model" refers to the learning process of using collected data to improve the performance of an artificial intelligence model.

[0104] "Making optimal decisions" refers to determining the most effective actions or choices based on collected data and trained AI models.

[0105] A system for implementing this invention has the following configuration. First, the server uses a generative AI to build a virtual brain model that makes decisions. Specifically, the manager's past decision-making data is used as learning data, and the generative AI learns the patterns. This builds a virtual brain that mimics the manager's decision-making.

[0106] Next, the server has the means to speed up the PDCA cycle and improve the accuracy of the model, which allows the virtual brain model to be continuously improved and enable more accurate decision-making.

[0107] Furthermore, the server has the means to learn and apply data obtained from individual brains to build a personalized virtual brain, thereby providing optimized decision-making support for each manager.

[0108] The server also includes a means for collecting and storing real-time data from the production line in the factory. Specifically, sensors are used to obtain data on product quality, production speed, machine operation status, etc., and the data is stored in an SQL database.

[0109] The server also includes a means to preprocess the collected data and train the AI ​​model. Python is used to fill in missing values ​​and remove outliers, and TensorFlow is used to train the AI ​​model.

[0110] The server also includes a means to analyze production line data in real time using trained AI models to make optimal decisions, suggesting actions to adjust production speed or control quality.

[0111] For example, if product quality is declining, the AI ​​model will learn from past data what measures management has taken and suggest similar measures, such as temporarily slowing down production speed or performing maintenance on specific machines.

[0112] An example of a prompt is, "Please learn about the measures taken by management when product quality deteriorated in the past, and propose the optimal measures based on the current production line data."

[0113] In this way, the server can mimic the manager's decision-making in real time and achieve efficient production management.

[0114] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0115] Step 1:

[0116] The server collects data in real time from the production line in the factory. Specifically, sensors are used to obtain data on product quality, production speed, machine operation status, etc. The input is raw data from the sensors, and the output is data stored in an SQL database.

[0117] Step 2:

[0118] The server preprocesses the collected data. Specifically, it uses Python to fill in missing values ​​and remove outliers. The input is raw data stored in an SQL database, and the output is the preprocessed, clean data.

[0119] Step 3:

[0120] The server uses the preprocessed data to train the AI ​​model. Specifically, it uses TensorFlow to learn from managers' past decision-making data and build a generative AI model. The input is the preprocessed data, and the output is the trained AI model.

[0121] Step 4:

[0122] The server uses the trained AI model to analyze production line data in real time. Specifically, it inputs current production line data into the AI ​​model to make optimal decisions. The input is production line data collected in real time, and the output is a proposal for optimal decisions.

[0123] Step 5:

[0124] The server adjusts the production line based on optimal decision-making. Specifically, it proposes actions for adjusting production speed and quality control. The input is the decision-making proposal from the AI ​​model, and the output is the adjustment results for the production line.

[0125] Step 6:

[0126] The user checks the suggestions from the server and manually adjusts them as necessary. Specifically, based on the suggestions from the server, the user can take actions such as temporarily slowing down the production speed or performing maintenance on a specific machine. The input is the suggestion from the server, and the output is the adjustment results made by the user.

[0127] Step 7:

[0128] The server collects the adjustment results as data again and saves them as training data for the next run. This completes the PDCA cycle, improving the accuracy of the model. The input is the adjustment results made by the user, and the output is the training data for the next run.

[0129] Example 2

[0130] Next, a description will be given of Example 2 of Form 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."

[0131] Today's business environment is rapidly changing, requiring managers to make quick and accurate decisions. However, traditional decision support systems often require manual data updates and model retraining, making it difficult to quickly implement the PDCA cycle. They also lack personalized support that reflects the decision-making patterns of individual managers. This makes it difficult for managers to make quick decisions to adapt to a changing environment.

[0132] 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.

[0133] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for periodically updating the learning data and learning new decision-making patterns, means for retraining and evaluating the generative AI model, means for inputting prompt sentences to request specific predictions or proposals, and means for providing the generated predictions or proposals, thereby enabling managers to make quick and accurate decisions.

[0134] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and patterns from data.

[0135] A "virtual brain" is a model that uses artificial intelligence technology to mimic the decision-making process of the human brain.

[0136] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0137] "Personalization" refers to optimizing systems and services to suit the characteristics and needs of individual users.

[0138] "Training data" is a data set used to train an artificial intelligence model.

[0139] "Retraining" is the process of adding new data to an existing artificial intelligence model and training it again.

[0140] "Evaluation" is the process of measuring the performance of an artificial intelligence model and confirming its accuracy and effectiveness.

[0141] A "prompt sentence" is an instruction sentence that a user enters to request a specific prediction or suggestion from the system.

[0142] "Prediction" refers to an artificial intelligence model's estimation of future events or outcomes based on past data.

[0143] A "suggestion" is a specific recommendation of action or strategy that an AI model provides to a user.

[0144] This invention is a system that uses generative AI to build a virtual brain model for decision-making and improves the accuracy of the model by quickly implementing the PDCA cycle. Furthermore, it aims to build a personalized virtual brain by learning and applying data obtained from individual brains.

[0145] Hardware and software used

[0146] The server uses a computer equipped with a high-performance GPU (e.g., NVIDIA Tesla V100), and the software uses a machine learning framework (e.g., TensorFlow, PyTorch), which enables rapid processing of large amounts of data and efficient retraining of generative AI models.

[0147] Program processing

[0148] The server first collects data provided by the user, including past sales data, seasonal fluctuation data, competitor trends, etc. This data is uploaded in CSV or Excel file format.

[0149] The server then preprocesses the collected data, specifically by completing missing values, detecting and correcting outliers, and normalizing the data, thereby improving the quality of the data and the accuracy of the model.

[0150] The server retrains the generative AI model using the preprocessed data. For retraining, a dataset divided into training data and test data is used. The server initializes the generative AI model and retrains the model using the training data. After retraining, the performance of the model is evaluated using the test data.

[0151] Once the evaluation is complete, the model is used to assist the user in making decisions. The user inputs a prompt to request a specific prediction or suggestion. The server uses the generative AI model to generate the requested prediction or suggestion and provides it to the user.

[0152] Specific examples

[0153] Consider a company that wants to forecast market demand. Users upload sales data from the past three years, seasonal fluctuation data, and competitor trends to a server. The server uses this data to retrain a generative AI model and generate demand forecasts for the next quarter.

[0154] Example prompt sentence:

[0155] "Retrain your generative AI model using the past three years of sales data, seasonal fluctuation data, and competitor trends data to forecast demand for product A for the next quarter."

[0156] By entering this prompt, the server automatically collects data, retrains the model, and provides demand forecasts, allowing businesses to make fast and accurate decisions.

[0157] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0158] Step 1: Data collection

[0159] Users upload past sales data, seasonal fluctuation data, and competitor trends data to the server. Data in CSV or Excel file format is used as input. The server receives these files and stores them in the database. Specifically, the server validates the file format and stores them appropriately in the database.

[0160] Step 2: Data Preprocessing

[0161] The server pre-processes the collected data. As input, it uses the raw data stored in the database. Specifically, the server:

[0162] Imputing missing values: Imputing missing values ​​with the mean or median.

[0163] Outlier detection and correction: Statistical methods are used to detect outliers and correct them appropriately.

[0164] Data normalization: Standardizing or normalizing data.

[0165] As an output, pre-processed data is obtained.

[0166] Step 3: Retrain the model

[0167] The server retrains the generative AI model using the preprocessed data. The preprocessed data is used as input. Specifically, the server performs the following operations:

[0168] Data Split: Split the preprocessed data into training data and test data.

[0169] Initialize model: Initialize the generative AI model.

[0170] Train the model: Retrain the model using the training data.

[0171] The output is a retrained generative AI model.

[0172] Step 4: Model evaluation

[0173] The server evaluates the performance of the retrained model. The test data and the retrained model are used as input. Specifically, the server performs the following operations:

[0174] Run prediction: Make a prediction for the model using the test data.

[0175] Calculate evaluation metrics: Calculate evaluation metrics such as precision, recall, and F1 score.

[0176] As an output, the evaluation result is obtained.

[0177] Step 5: Decision support

[0178] The user inputs a prompt to request a specific prediction or suggestion. The prompt and the retrained model are used as input. Specifically, the server performs the following actions:

[0179] Parsing prompts: Parse the prompts entered by the user.

[0180] Generate predictions and recommendations: Use generative AI models to generate the requested predictions and recommendations.

[0181] As an output, the generated predictions and suggestions are provided to the user.

[0182] Example prompt sentence:

[0183] "Retrain your generative AI model using the past three years of sales data, seasonal fluctuation data, and competitor trends data to forecast demand for product A for the next quarter."

[0184] (Application example 2)

[0185] Next, a description will be given of Application Example 2 of Form 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."

[0186] Inventory management and delivery route optimization are important issues for modern logistics centers. However, traditional manual management methods have limitations when it comes to quickly responding to changes in the business environment. In particular, delays in updating inventory and delivery data make efficient operations difficult, resulting in increased costs and decreased customer satisfaction. To solve these issues, an automated decision-making support system using AI is needed.

[0187] 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 constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for periodically updating inventory data and delivery data of the logistics center and proposing optimal inventory management and delivery routes, and means for periodically updating data and relearning the model to make decisions in response to changes in the business environment. This makes it possible to optimize inventory management and delivery routes at the logistics center, enabling rapid response to changes in the business environment.

[0188] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and patterns from data.

[0189] A "virtual brain" is a model that uses artificial intelligence technology to mimic the decision-making process of the human brain.

[0190] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0191] "Personalization" refers to the individual optimization of services and content based on the characteristics and behavior of individual users.

[0192] "Inventory data" refers to data that includes information such as the quantity of products at the logistics center, arrival date, and shipping date.

[0193] "Delivery data" refers to data that includes information such as the delivery route, delivery time, and destination of the product departing from the logistics center.

[0194] The "business environment" refers to the external and internal factors that affect a company's business operations, including market trends, competitive conditions, and regulations.

[0195] A "decision support system" is a system that uses data analysis and artificial intelligence technology to provide information and suggestions to managers and executives to help them make optimal decisions.

[0196] The system for implementing this invention aims to optimize inventory management and delivery routes in logistics centers. The system uses generative AI to build a virtual brain model for decision-making, speeding up the PDCA cycle and improving the accuracy of the model. It also learns and applies data obtained from individual brains to build a personalized virtual brain.

[0197] The server uses the following hardware and software:

[0198] Hardware: Server or cloud environment (e.g., AWS, Google Cloud)

[0199] Software: Python, pandas (data manipulation), scikit-learn (machine learning), schedule (regular execution)

[0200] The server first periodically updates inventory and delivery data, allowing it to propose optimal inventory management and delivery routes based on the latest data. Specifically, inventory data includes information such as product ID, stock quantity, arrival date, and shipping date, while delivery data includes information such as delivery ID, departure point, destination, and delivery time.

[0201] The server then uses generative AI to build a virtual brain model. This model is then incorporated into the "Smart Logistics Manager" smartphone application used by logistics center operators. The application provides optimized inventory management, delivery routes, real-time data updates and analysis, and decision-making support in response to changes in the business environment.

[0202] The server periodically updates the data and retrains the model to quickly respond to changes in the business environment, enabling optimization of inventory management and delivery routes at logistics centers and achieving efficient operations.

[0203] As a concrete example, the following prompt sentence can be used:

[0204] "Use inventory and delivery data from your distribution center to create an AI model that suggests optimal inventory management and delivery routes. The model will update the data regularly and make decisions that respond to changes in the business environment."

[0205] In this way, the server can provide the logistics center operator with information to help them make optimal decisions.

[0206] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0207] Step 1:

[0208] The server reads inventory data and delivery data. Specifically, the server obtains inventory data (product ID, stock quantity, arrival date, shipping date, etc.) and delivery data (delivery ID, departure point, destination, delivery time, etc.) from a CSV file or database. The input is inventory data and delivery data, and the output is a data frame containing these data.

[0209] Step 2:

[0210] The server preprocesses the acquired data, specifically by imputing missing values, normalizing data, encoding categorical data, etc. The input is the data frame acquired in step 1, and the output is the preprocessed data frame.

[0211] Step 3:

[0212] The server trains a generative AI model using the preprocessed data. Specifically, it splits the data into training data and test data, and trains it using a random forest regression model. The input is the preprocessed data frame, and the output is the trained generative AI model.

[0213] Step 4:

[0214] The server uses a trained generative AI model to optimize inventory management and delivery routes. Specifically, it inputs new data into the model and predicts optimal inventory levels and delivery routes. The input is new inventory data and delivery data, and the output is a proposal for optimal inventory levels and delivery routes.

[0215] Step 5:

[0216] The server evaluates the accuracy of the generative AI model. Specifically, it evaluates the model's predictive accuracy using test data and adjusts the model's parameters as necessary. The input is the test data and the trained generative AI model, and the output is the model's accuracy evaluation result.

[0217] Step 6:

[0218] The server periodically updates the data and retrains the model. Specifically, it uses a schedule library to perform data updates and model retraining at midnight every day. The input is the latest inventory and delivery data, and the output is an updated generative AI model.

[0219] Step 7:

[0220] The user uses the smartphone application "Smart Logistics Manager" to check the optimal inventory management and delivery route proposals provided by the server. Specifically, the application retrieves data from the server and displays it to the user. The input is the proposal data from the server, and the output is the optimization proposal displayed to the user.

[0221] Example 3

[0222] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0223] Conventional decision support systems have difficulty reflecting the decision-making styles of individual managers, and can only provide support based on general algorithms. Therefore, personalized support that takes into account the individual decision-making criteria and styles of managers is required. In addition, efficient and accurate methods are required for preprocessing collected data and model training.

[0224] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0225] In this invention, the server includes means for collecting individual electroencephalogram data, means for collecting behavioral data, means for preprocessing the collected electroencephalogram data and behavioral data, means for training a generative AI model using the preprocessed data, means for constructing a personalized virtual brain using the trained generative AI model, and means for displaying the decision-making results of the virtual brain, thereby enabling personalized support that reflects the individual decision-making style of the manager.

[0226] "Individual electroencephalogram data" refers to data of electrical signals acquired to measure an individual's brain activity.

[0227] "Behavioral data" refers to data that records an individual's behavior, operation history, activity log, etc.

[0228] "Preprocessing" is the process of performing processes such as noise removal and normalization on collected data to convert it into a format suitable for machine learning models.

[0229] A "generative AI model" is an artificial intelligence model that is generated using machine learning algorithms to learn from data and perform specific tasks.

[0230] A "personalized virtual brain" is a virtual brain model designed to reflect the decision-making style of an individual manager.

[0231] "Decision-making results" are the results of judgments and choices made by the virtual brain in a specific business situation.

[0232] "Display means" refers to devices or software that visually present the decision-making results of the virtual brain to the user.

[0233] This invention is a system that collects individual electroencephalogram (EEG) data and behavioral data, trains a generative AI model based on the data, and constructs a personalized virtual brain. Specific embodiments of this system are described below.

[0234] First, the user wears an EEG device to collect the manager's brainwave data. The device acquires the brainwave data from the EEG device in real time and sends it to a server. For example, a manager wears the EEG device during a meeting, collecting brainwave data during that time. The user also installs dedicated software on their PC or smartphone to collect the manager's behavioral data. The device records the manager's operation history and activity log and sends it to the server. For example, it collects the operation history when the manager checks email or edits documents.

[0235] The server then stores the received EEG and behavioral data in a database. The server then preprocesses the data, specifically removing noise and normalizing the data to convert it into a format suitable for machine learning models. For example, the server filters the EEG data and checks the consistency of the timestamps of the behavioral data. The server uses a filtering algorithm to remove noise from the EEG data and standardize the timestamps of the behavioral data.

[0236] The server then uses the preprocessed data to train a generative AI model. Specifically, it uses machine learning libraries such as Python's TensorFlow and PyTorch. The server trains the model based on past data to learn the manager's decision-making style. For example, it trains the model using brainwave and behavioral data from when a manager decides whether to enter a new market. The server generates a training dataset and optimizes the model's parameters.

[0237] Next, the server uses the trained generative AI model to construct a virtual brain that reflects the manager's decision-making style. The server sets the virtual brain's parameters and simulates the manager's decision-making in specific business situations. For example, when the virtual brain decides whether to enter a new market, it reproduces the decision-making process based on past data. The server saves the results of the virtual brain's simulation and provides them to the user.

[0238] Finally, the user applies the constructed virtual brain to a real business situation. The terminal displays the virtual brain's decision-making results and provides them to the user. For example, the user can refer to the virtual brain's judgment when considering whether to enter a new market. The terminal displays the virtual brain's decision-making results in graphs and text format, making them easy for the user to understand.

[0239] Examples of prompts include:

[0240] "When a manager decides whether to enter a new market, the virtual brain learns what factors he or she considers based on past brainwave and behavioral data, and then makes a decision in a similar situation."

[0241] In this way, the server collects the manager's EEG data and behavioral data and uses the generative AI model to build a personalized virtual brain. This enables the virtual brain, which reflects the manager's decision-making style, to make decisions in specific business situations. The flow of the specific processing in Example 3 will be explained using Figure 15.

[0242] Step 1:

[0243] The user wears an EEG device to collect the executive's brainwave data. The terminal acquires the brainwave data from the EEG device in real time and transmits it to the server. The input is the brainwave data from the EEG device, and the output is the brainwave data transmitted to the server. Specifically, the executive wears the EEG device during a meeting, and brainwave data is collected during that time.

[0244] Step 2:

[0245] To collect behavioral data on managers, users install dedicated software on their PCs or smartphones. The devices record the manager's operation history and activity logs and send them to a server. The input is the operation history and activity logs from the PC or smartphone, and the output is the behavioral data sent to the server. Specifically, the software collects operation history when the manager checks email or edits documents.

[0246] Step 3:

[0247] The server stores the received EEG data and behavioral data in a database. The input is the EEG data and behavioral data sent to the server, and the output is the data stored in the database. Specifically, the server creates a new entry in the database and stores the received data.

[0248] Step 4:

[0249] The server preprocesses the data. Specifically, it removes noise and normalizes the data, converting it into a format suitable for machine learning models. The input is EEG data and behavioral data stored in a database, and the output is the preprocessed data. Specifically, the server uses a filtering algorithm to remove noise from the EEG data and standardize the timestamps of the behavioral data.

[0250] Step 5:

[0251] The server uses the preprocessed data to train the generative AI model. Specifically, it uses machine learning libraries such as Python's TensorFlow and PyTorch. The input is the preprocessed EEG data and behavioral data, and the output is a trained generative AI model. Specifically, the server generates a training dataset and optimizes the model parameters.

[0252] Step 6:

[0253] The server uses a trained generative AI model to build a virtual brain that reflects the manager's decision-making style. The input is the trained generative AI model, and the output is the virtual brain. Specifically, the server sets the parameters of the virtual brain and simulates the manager's decision-making in specific business situations.

[0254] Step 7:

[0255] The user applies the constructed virtual brain to actual business situations. The terminal displays the decision-making results of the virtual brain and provides them to the user. The input is the decision-making results of the virtual brain, and the output is the results displayed on the terminal. In concrete terms, the terminal displays the decision-making results of the virtual brain in graphs and text format, making them easy for the user to understand.

[0256] (Application example 3)

[0257] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0258] In conventional factory operations, it is difficult to optimize production lines and troubleshoot problems that reflect the manager's decision-making style, which hinders efficient operations. Additionally, because there is no system that can mimic the manager's decision-making, it is difficult to respond quickly when the manager is absent.

[0259] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0260] In this invention, the server includes a means for constructing a virtual brain model that makes decisions using generative AI, a means for speeding up the PDCA cycle and improving the accuracy of the model, a means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, and a means for using the virtual brain that reflects the manager's decision-making style to optimize and troubleshoot production lines in a factory, thereby enabling efficient factory operation that reflects the manager's decision-making style.

[0261] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and models from data.

[0262] A "virtual brain" is an artificial intelligence model built based on an individual's brainwave and behavioral data, which mimics a specific decision-making style.

[0263] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0264] "Personalization" refers to individually optimizing systems and services based on individual characteristics and behavior.

[0265] "Production line optimization" refers to making adjustments and improvements to streamline production processes within a factory and achieve maximum productivity.

[0266] "Troubleshooting" refers to procedures and methods for identifying and resolving problems or failures that occur in a system or process.

[0267] "Managerial decision-making style" refers to the tendency or pattern of how managers make judgments and choices in specific situations.

[0268] To implement the present invention, the following system configuration and program are required.

[0269] System Configuration

[0270] 1. Hardware:

[0271] Brainwave sensor: A device for acquiring brainwave data from business managers.

[0272] Behavioral tracking system: A device for capturing behavioral data of business managers.

[0273] Server: A computer system for processing data and constructing a virtual brain.

[0274] Factory robots: Robots for production line optimization and troubleshooting.

[0275] 2. Software:

[0276] Data collection software: Programs for collecting data from EEG sensors and behavioral tracking systems.

[0277] Generative AI model: An algorithm for constructing a virtual brain based on collected data. Specifically, it uses a multi-layer perceptron (MLPClassifier) ​​using the scikit-learn library.

[0278] Optimization software: A program that uses a virtual brain to optimize and troubleshoot production lines.

[0279] Program processing explanation

[0280] The server first collects the manager's brainwave and behavioral data from EEG sensors and a behavioral tracking system. This data is then sent to the server via data collection software. The server then uses a generative AI model to build a virtual brain based on the collected data. This virtual brain is designed to mimic the manager's decision-making style.

[0281] After the virtual brain is built, the server uses optimization software to optimize and troubleshoot the production line. Specifically, the virtual brain uses the current state of the production line in the factory as input and makes decisions to make optimal adjustments. When a problem occurs, the virtual brain uses the problem data as input and proposes the optimal solution.

[0282] Specific examples

[0283] For example, if a machine breaks down in a factory, the server will suggest the optimal repair procedure based on past data. The virtual brain, which has learned the decisions managers have made in the past in response to similar breakdowns, will suggest repair procedures in real time, enabling a fast and efficient response.

[0284] Prompt Sentence Examples

[0285] "A machine has broken down in the factory. Please suggest the optimal repair procedure based on past data."

[0286] In this way, by using a virtual brain that reflects the manager's decision-making style, production efficiency within the factory can be improved.

[0287] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0288] Step 1:

[0289] The server collects the manager's EEG data and behavioral data from the EEG sensor and behavior tracking system. These data are sent to the server through the data collection software. The input is EEG data and behavioral data, and the output is raw data stored on the server.

[0290] Step 2:

[0291] The server preprocesses the collected EEG and behavioral data, specifically by removing noise and normalizing the data. The input is raw data, and the output is preprocessed data.

[0292] Step 3:

[0293] The server uses the preprocessed data to train a generative AI model. Specifically, it uses the scikit-learn library to build a multilayer perceptron (MLPClassifier) ​​and trains it on the data. The input is the preprocessed data, and the output is the trained generative AI model.

[0294] Step 4:

[0295] The server uses a trained generative AI model to build a virtual brain, which is designed to mimic the decision-making style of a manager. The input is the trained generative AI model, and the output is the virtual brain.

[0296] Step 5:

[0297] The server inputs the current state of the production line in the factory and uses the virtual brain to make decisions for optimal adjustments. Specifically, production line data obtained from sensors is input into the virtual brain, which then outputs optimal adjustment proposals. The input is production line data, and the output is adjustment proposals.

[0298] Step 6:

[0299] When a problem occurs, the server inputs the problem data and uses the virtual brain to propose the optimal solution. Specifically, data on breakdowns and problems is input into the virtual brain, which then outputs the optimal repair procedures and solutions. The input is problem data, and the output is the solution.

[0300] Step 7:

[0301] The user adjusts the production line and performs troubleshooting based on the adjustment proposals and solutions provided by the server. Specifically, the user operates or repairs the machine according to instructions from the server. The input is the instruction from the server, and the output is the result of the adjustment or repair that was performed.

[0302] 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.

[0303] "Example 1"

[0304] In one embodiment of the present invention, the emotion engine recognizes the user's emotion by extracting the emotion from the user's facial expression, voice, text message, etc. For example, if the user shows an expression that expresses joy, the emotion engine recognizes the information as "joy."

[0305] "Example 2"

[0306] Furthermore, the emotion data obtained from the emotion engine is provided to the generative AI. The generative AI learns this emotion data and makes decisions that take into account the user's emotional state. For example, if the user expresses joy, the generative AI will take that information into account and provide decisions that are likely to make the user feel joyful.

[0307] "Example 3"

[0308] The personalized virtual brain also provides decision-making that takes into account the user's emotional data.

[0309] Specifically, the system adjusts the priority of decision-making depending on the user's emotional state. For example, if the user is feeling stressed, it will prioritize options that will reduce stress.

[0310] The processing flow of each embodiment will be described below.

[0311] "Example 1"

[0312] Step 1: The emotion engine extracts emotions from the user's facial expressions, voice, text messages, etc.

[0313] Step 2: The emotion engine converts the extracted emotions into emotion labels such as "joy," "sadness," or "anger."

[0314] Step 3: The emotion engine provides emotion labels to the generative AI.

[0315] "Example 2"

[0316] Step 1: The generative AI receives the emotion labels provided by the emotion engine.

[0317] Step 2: The generative AI uses the received emotion labels as training data to build a decision-making model that takes the user's emotional state into account.

[0318] Step 3: The generative AI uses the constructed decision-making model to provide a decision that takes into account the user's emotional state.

[0319] "Example 3"

[0320] Step 1: The personalized virtual brain receives the decisions provided by the generative AI.

[0321] Step 2: The personalized virtual brain adjusts the received decisions according to the user's emotional state.

[0322] Step 3: The personalized virtual brain provides tailored decision-making to the user.

[0323] Example 1

[0324] Next, a description will be given of Example 1 of Form 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."

[0325] Conventional decision support systems have the problem that it is difficult to imitate the decision-making patterns of managers and are unable to provide decision support that takes user emotions into account. Furthermore, there is a lack of means to effectively utilize user feedback to improve the model. For this reason, there is a need for a system that can more accurately imitate managerial decision-making and provide personalized decision support that takes user emotions into account.

[0326] 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.

[0327] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for collecting user emotion data, means for analyzing the collected emotion data and recognizing the user's emotion, means for supporting decision-making based on the recognized emotion, and means for collecting user feedback and improving the model. This enables personalized decision-making support that accurately mimics managerial decision-making and takes user emotion into consideration.

[0328] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and models from data.

[0329] A "virtual brain" is a system that uses artificial intelligence technology to mimic the functions of the human brain and perform decision-making and information processing.

[0330] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0331] "Personalization" refers to optimizing systems and services to suit the characteristics and needs of individual users.

[0332] "Emotion data" is information about emotions extracted from the user's facial expressions, voice, text messages, etc.

[0333] "Emotion recognition" is a technology that analyzes collected emotional data and identifies the user's emotional state.

[0334] "Decision support" refers to providing necessary information and suggestions to support users when making decisions.

[0335] "Feedback" refers to the evaluation and reaction information provided by the user regarding the system's suggestions and results.

[0336] This invention is a system that uses generative AI to construct a virtual brain model that makes decisions and provides personalized decision-making support that takes into account the user's emotions. Specific embodiments of this system are described below.

[0337] The server uses generative AI to build a virtual brain model that makes decisions. Specifically, the server connects to the company's database (e.g., MySQL) and collects data on past decisions made by managers. The collected data is preprocessed using the Pandas library to impute missing values, normalize the data, and encode categorical data. A neural network model is then built using TensorFlow, and the model is trained using the preprocessed data. This model is then used to predict the manager's decision-making patterns.

[0338] The device collects the user's emotional data by capturing the user's facial expressions with the device's camera, recording audio with the microphone, and capturing text messages. This data is then sent to the server in real time via WebSocket.

[0339] The server analyzes the received emotion data. Specifically, it uses OpenCV to analyze facial expressions and IBM Watson (registered trademark) to analyze voice and text messages. For example, if the user smiles, the server recognizes the information as "joy."

[0340] The server uses the trained AI model to assist users in decision-making. When a user asks a question about the launch of a new product, the server generates optimal suggestions based on past decision-making patterns. These suggestions are generated by the generative AI model.

[0341] As a concrete example, we use historical data on when a manager decides to launch a new product. The server provides this data to the AI, which then learns the patterns. When a user asks a question about launching a new product, the AI ​​suggests the optimal decision based on past patterns.

[0342] Example prompt sentence:

[0343] "Please provide optimal recommendations for the next new product launch based on past new product launch decision-making data."

[0344] The user receives suggestions from the server and provides feedback on the suggestions. Specifically, the user indicates their satisfaction with the suggestions through facial expressions and voice. The device collects this feedback and sends it to the server. The server uses this feedback to further improve the AI ​​model.

[0345] Example prompt sentence:

[0346] "Analyze how users feel when they receive a new product suggestion and tailor your next suggestion accordingly."

[0347] In this way, personalized decision support that accurately mimics the decision-making of managers and takes into account the user's emotions becomes possible.

[0348] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0349] Step 1: Data collection

[0350] The server connects to the company's database and collects data on past decisions made by management. Specifically, the server executes the query "SELECT FROM decision_data WHERE date > '2020-01-01'" on the MySQL database to retrieve decision data. This data includes information such as the type of decision, date and time, and outcome. The input is the database query, and the output is the retrieved decision data.

[0351] Step 2: Data Preprocessing

[0352] The server preprocesses the collected data. Specifically, it uses the Pandas library to impute missing values ​​and Scikit-learn to normalize the data. It also performs one-hot encoding on categorical data. The input is the collected decision data, and the output is the preprocessed data.

[0353] Step 3: Model training

[0354] The server uses the preprocessed data to train an AI model. Specifically, it uses TensorFlow to build a neural network model and trains the model using the preprocessed data as input. The input is the preprocessed data, and the output is a trained AI model.

[0355] Step 4: Collect emotional data

[0356] The device collects the user's emotional data. Specifically, it captures the user's facial expressions with the device's camera, records audio with the microphone, and obtains text messages. These data are sent to the server in real time via WebSocket. The input is the user's facial expressions, audio, and text messages, and the output is the collected emotional data.

[0357] Step 5: Emotion Recognition

[0358] The server analyzes the received emotion data. Specifically, it uses OpenCV to analyze facial expressions and IBM Watson to analyze voice and text messages. For example, if a user smiles, the server recognizes that information as "joy." The input is the collected emotion data, and the output is the recognized emotion information.

[0359] Step 6: Decision support

[0360] The server uses the trained AI model to assist the user in decision-making. Specifically, when the user inputs a prompt such as "Please give me the best proposal for the next new product launch," the server generates the best proposal based on past decision-making patterns. The inputs are the user's prompt and the trained AI model, and the output is the generated proposal.

[0361] Step 7: User Feedback

[0362] The user receives suggestions from the server and provides feedback on the suggestions. Specifically, the user indicates whether they are satisfied with the suggestions using facial expressions and voice. The device collects this feedback and sends it to the server. The server uses this feedback to further improve the AI ​​model. The input is the user's feedback, and the output is an improved AI model.

[0363] (Application example 1)

[0364] Next, a description will be given of Application Example 1 of Embodiment 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."

[0365] Conventional factory management systems have difficulty reflecting the decision-making patterns of managers in real time, and lack support that takes into account the emotions of factory staff, resulting in problems that lead to reduced production efficiency and staff satisfaction. For this reason, there is a need for a system that can learn the decision-making patterns of managers, optimize factory production lines, and recognize staff emotions to provide appropriate feedback.

[0366] 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.

[0367] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for providing feedback based on the user's emotions using an emotion engine that recognizes the user's emotions, and means for learning the decision-making patterns of managers and optimizing factory production lines.This makes it possible to reflect the manager's decision-making patterns in real time and provide support that takes the emotions of factory staff into consideration.

[0368] "Generative AI" is an algorithm that uses artificial intelligence techniques to generate new information and patterns from data.

[0369] A "virtual brain" is a model that uses artificial intelligence technology to mimic the decision-making process of the human brain.

[0370] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0371] "Personalization" refers to providing services and functions that are customized to suit the characteristics and needs of individual users.

[0372] An "emotion engine" is a technology for recognizing and analyzing emotions from a user's facial expressions, voice, text messages, etc.

[0373] "Managerial decision-making patterns" are decision-making tendencies and rules extracted based on data on decisions made by managers in the past.

[0374] "Production line optimization" refers to making adjustments and improvements to streamline a factory's production processes, reducing costs and improving productivity.

[0375] "Feedback" refers to evaluations and advice provided regarding the system or user's behavior.

[0376] As an embodiment of this invention, we will explain a system for smart factory management robots as an example. This system uses generative AI to build a virtual brain model for decision-making and optimize the factory production line. It also uses an emotion engine to recognize the emotions of factory staff and provide appropriate feedback.

[0377] The server uses generative AI to build a virtual brain model that makes decisions. This model learns from the manager's past decision-making data and extracts patterns, making it possible to imitate the manager's decision-making.

[0378] Next, the server speeds up the PDCA cycle to improve the accuracy of the model. This cycle consists of four steps: Plan, Do, Check, and Act, which are repeated to continuously improve the accuracy of the model.

[0379] Furthermore, the server learns and applies data obtained from the individual's brain to build a personalized virtual brain, enabling decision-making support tailored to the characteristics and needs of each individual manager.

[0380] The emotion engine recognizes emotions from users' facial expressions, voice, text messages, etc. For example, a camera can be used to capture the facial expressions of factory staff, and the emotion engine can analyze the expressions to extract emotions. This allows the system to provide feedback based on the staff's emotions.

[0381] Generative AI models that learn the decision-making patterns of managers can optimize factory production lines. For example, if a production line stops, the model can quickly resolve the issue by referring to past decision-making patterns. It can also obtain production data in real time and create optimal production plans.

[0382] For example, if a production line in a factory stops, the robot will refer to the manager's past decision-making patterns to quickly resolve the problem. Also, if a factory staff member is feeling stressed, the robot will use its emotion engine to recognize that emotion and provide appropriate support.

[0383] Examples of prompts to input to a generative AI model include:

[0384] "If the production line stops in a factory, how can you quickly solve the problem by referring to the decision-making patterns of past managers?"

[0385] In this way, smart factory management robots can improve factory productivity and efficiency while taking into account management decision-making patterns and staff emotions.

[0386] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0387] Step 1:

[0388] The server acquires images from cameras installed in the factory and captures the facial expressions of factory staff in real time. The input is the camera image, and the output is the captured image data.

[0389] Step 2:

[0390] The server inputs the captured image data into the emotion engine and recognizes emotions from facial expressions. The emotion engine analyzes the image data using a facial expression recognition model and extracts the user's emotion. The input is image data, and the output is recognized emotion data.

[0391] Step 3:

[0392] The server provides appropriate feedback to the user based on the recognized emotion data. For example, if the user feels stressed, the server provides advice on how to relax. The input is emotion data, and the output is a feedback message.

[0393] Step 4:

[0394] The server inputs the manager's past decision-making data into the generative AI model to learn the decision-making patterns. The generative AI model analyzes the past data and extracts the manager's decision-making patterns. The input is the decision-making data, and the output is the learned decision-making patterns.

[0395] Step 5:

[0396] The server inputs production data acquired in real time into the generative AI model and creates an optimal production plan. The generative AI model optimizes the production line based on the learned decision-making patterns. The input is production data, and the output is an optimized production plan.

[0397] Step 6:

[0398] The server sends the optimized production plan to the factory robots to execute it. The factory robots adjust the production line based on the received production plan to ensure efficient production. The input is the optimized production plan, and the output is the adjusted production line.

[0399] Step 7:

[0400] The server monitors the operation of the factory robots and re-optimizes them as necessary, thereby always maintaining optimal production conditions. The input is the robot's operation data, and the output is a re-optimized production plan.

[0401] In this way, the server can use generative AI models and emotion engines to improve factory productivity and staff satisfaction.

[0402] Example 2

[0403] Next, a description will be given of Example 2 of Form 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."

[0404] Conventional decision-making support systems do not take into account the emotional state of the user when making decisions, making it difficult to respond quickly to changes in the business environment. In addition, there was a lack of means to improve the accuracy of the model by quickly implementing the PDCA cycle.

[0405] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for constructing a virtual brain model that makes decisions using a generative AI; means for speeding up the PDCA cycle and improving the accuracy of the model; means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain; means for receiving user input and generating emotion data through an emotion engine; means for transmitting the generated emotion data to the server; means for the server to obtain the latest management data and feed it back to the AI ​​model; means for making decisions based on the user's emotion data and the latest management data using the generative AI model; and means for displaying the decision-making results to the user. This enables fast and accurate decision-making that takes the user's emotional state into consideration.

[0406] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and decisions from data.

[0407] A "virtual brain" is a system that uses artificial intelligence technology to mimic the functions of the human brain and make decisions.

[0408] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0409] "Personalization" refers to optimizing systems and services to suit the characteristics and needs of individual users.

[0410] An "emotion engine" is a system that uses natural language processing technology to analyze and generate emotional data from user input.

[0411] "Emotion data" is data that expresses the user's emotional state as numerical values ​​or categories.

[0412] A "server" is a computer system that processes data and provides services over a network.

[0413] "Management data" refers to data that includes information related to management, such as corporate performance and market trends.

[0414] "Decision-making results" are the optimal actions and strategies proposed by the generative AI model based on user input, emotional data, and management data.

[0415] MODE FOR CARRYING OUT THE INVENTION

[0416] This invention is a system that uses generative AI to build a virtual brain model that makes decisions, improves the accuracy of the model by quickly implementing the PDCA cycle, and provides decision-making that takes into account the emotional state of the user. Specific embodiments of this system are described below.

[0417] Hardware and software used

[0418] Server: A computer system used to process data and train AI models. For example, an AWS EC2 instance.

[0419] Terminal: A device used by a user to input information, such as a PC or smartphone.

[0420] Emotion engine: Software for analyzing and generating emotion data from user input, for example using an NLP library (such as spaCy).

[0421] Generative AI models: Algorithms for generating new information and decisions from data, for example using TensorFlow or PyTorch.

[0422] Explanation of program processing

[0423] 1. User input:

[0424] A user inputs a question or request into the system through a terminal interface, for example, "What are the sales forecasts for the next quarter?"

[0425] 2. Emotion data generation:

[0426] The device sends the user's input to the emotion engine, which analyzes the user's emotional state. The emotion engine uses NLP techniques to extract emotion data from the user's input. For example, if the user's input indicates a positive emotion, the emotion engine generates emotion data for "joy."

[0427] 3. Sending Emotional Data:

[0428] The terminal transmits the generated emotion data to the server. For example, the emotion data of "joy" and the user's input are transmitted to the server.

[0429] 4. Management data acquisition and feedback:

[0430] The server retrieves the latest management data from the database. For example, it retrieves the latest sales data and market trend data. The retrieved data is fed back to the AI ​​model. The server preprocesses the data using Python and retrains the model using machine learning libraries such as TensorFlow and PyTorch.

[0431] 5. Decision making:

[0432] The server uses a generative AI model to make optimal decisions in response to user requests. The generative AI model makes decisions by taking into account the user's emotional data and the latest management data. For example, if a user inputs, "What is the sales forecast for next quarter?", the generative AI model will provide a positive sales forecast based on the emotional data of "happiness" and the latest sales data.

[0433] 6. Sending and displaying decision results:

[0434] The server sends the decision-making results output by the generative AI model to the device. For example, it sends the results of a sales forecast to the device. The device then displays the decision-making results received from the server to the user. For example, it displays, "Sales forecast for the next quarter are expected to increase by 10% compared to the previous year."

[0435] Examples of concrete examples and prompts

[0436] Examples:

[0437] User types, "What is your forecast for next quarter?"

[0438] The emotion engine generates emotion data for "joy."

[0439] The server retrieves the latest sales data and feeds it back into the generative AI model.

[0440] The generative AI model generates a positive sales forecast and sends it to the device.

[0441] The device displays, "Sales forecast for next quarter are expected to increase 10% year over year."

[0442] Example prompt sentence:

[0443] "What are your sales forecasts for the next quarter?"

[0444] "Please tell us how you feel about this week's performance."

[0445] "Please suggest a marketing strategy for the new product."

[0446] In this way, users receive decisions that take their emotional state into account and can respond quickly to changes in the business environment.

[0447] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0448] Step 1:

[0449] The user types into the terminal.

[0450] A user inputs a question or request into the system through the terminal interface, for example, "What is the sales forecast for the next quarter?" The input text data is sent to the terminal.

[0451] Step 2:

[0452] The device generates emotion data through an emotion engine.

[0453] The device sends the user's input text data to the emotion engine, which then uses NLP technology to extract emotion data from the user's input. For example, if the user's input indicates a positive emotion, the emotion engine generates emotion data for "joy." The input is the user's text data, and the output is emotion data.

[0454] Step 3:

[0455] The device transmits the emotion data to the server.

[0456] The device sends the generated emotion data and the user's input in JSON format to the server. The input is the emotion data and the user's text data, and the output is the data to be sent to the server.

[0457] Step 4:

[0458] The server obtains the latest management data and provides feedback to the AI ​​model.

[0459] The server retrieves the latest management data from the database using SQL queries. For example, it retrieves the latest sales data or market trend data. The retrieved data is preprocessed using a Python script and fed back to the AI ​​model. The input is management data from the database, and the output is the preprocessed data.

[0460] Step 5:

[0461] The server makes decisions using generative AI models.

[0462] The server inputs emotional data and the latest business data into the generative AI model to make optimal decisions based on the user's request. For example, if a user inputs "Please tell me the sales forecast for next quarter," the generative AI model generates a positive sales forecast based on the emotional data of "happiness" and the latest sales data. The inputs are emotional data and business data, and the output is the decision-making result.

[0463] Step 6:

[0464] The server transmits the decision result to the terminal.

[0465] The server sends the decision-making results output by the generative AI model to the terminal in JSON format. The input is the decision-making results, and the output is the data sent to the terminal.

[0466] Step 7:

[0467] The terminal displays the results to the user.

[0468] The terminal displays the decision-making results received from the server to the user. For example, it displays "Sales forecast for the next quarter are expected to increase by 10% compared to the previous year." The input is the decision-making results from the server, and the output is the data displayed to the user.

[0469] (Application example 2)

[0470] Next, a description will be given of Application Example 2 of Form 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."

[0471] Conventional decision-making systems make decisions without considering the user's emotional state, which leads to problems such as reduced user satisfaction and engagement. Furthermore, it is difficult to recognize emotions in real time and propose optimal products and services based on those emotions, making it difficult to improve the user experience.

[0472] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[0473] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for providing emotion data obtained from an emotion engine to the generative AI and making decisions taking into account the user's emotional state, means for suggesting optimal products and services based on the user's emotional state, means for capturing the user's face using a camera and performing emotion recognition, means for transmitting emotion data to the end point of the generative AI model and obtaining recommended products and services, and means for displaying the recommended products and services in real time. This enables personalized decision-making that takes into account the user's emotional state, improving the user experience.

[0474] "Generative AI" is a system that uses artificial intelligence technology to generate new information and decisions from data.

[0475] A "virtual brain" is a model that uses artificial intelligence technology to mimic the functions of the human brain and make decisions.

[0476] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0477] "Personalization" means optimizing services and information to suit the characteristics and preferences of individual users.

[0478] An "emotion engine" is a system for detecting and analyzing a user's emotional state.

[0479] "Emotion data" is data that indicates the user's emotional state and is acquired by the emotion engine.

[0480] An "endpoint" is a URL or URI that allows access to a particular service or function on a network.

[0481] "Recommended products and services" are products and services suggested based on the user's emotional state and preferences.

[0482] "Real-time" means that data acquisition and processing occurs immediately.

[0483] A "camera" is a device for taking images or videos.

[0484] "Emotion recognition" is a technology that determines a user's emotional state from facial expressions, voice, etc.

[0485] A system for carrying out this invention has the following configuration: A server includes means for constructing a virtual brain model that makes decisions using a generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for providing emotion data obtained from an emotion engine to the generative AI and making decisions taking into account the user's emotional state, means for suggesting optimal products and services based on the user's emotional state, means for capturing the user's face using a camera and recognizing emotions, means for transmitting emotion data to an endpoint of the generative AI model and obtaining recommended products and services, and means for displaying the recommended products and services in real time.

[0486] Hardware and software used

[0487] Hardware

[0488] Smart glasses or smartphone: A device worn or carried by the user that contains a camera.

[0489] Camera: A device for capturing the user's face and performing emotion recognition.

[0490] software

[0491] OpenCV: An image processing library used for camera capture and face detection.

[0492] Keras: A deep learning library used to build and run emotion recognition models.

[0493] Requests: An HTTP request library used to send emotion data to the generative AI model endpoint.

[0494] Data processing and calculation

[0495] 1. Camera image capture: Use the camera on the smart glasses or smartphone to capture the user's face in real time. OpenCV is used to acquire the image and perform face detection.

[0496] 2. Emotion Recognition: The captured facial image is input into an emotion recognition model built with Keras to determine the user's emotional state. Emotion data is obtained from the user's facial expressions.

[0497] 3. Sending Emotion Data: The determined emotion data is sent to the generative AI model endpoint using the Requests library. The generative AI then uses this emotion data to make decisions that take into account the user's emotional state.

[0498] 4. Acquisition and display of recommended products: The recommended products and services returned by the generation AI are displayed on the user's device in real time, allowing the user to receive suggestions for the most suitable products and services on the spot.

[0499] Specific examples

[0500] When a user wears the smart glasses and walks around the store, the camera captures the user's face and performs emotion recognition. For example, if the user shows joy, the generative AI will display recommendations such as, "This dress from our new collection is perfect for you!"

[0501] Prompt Sentence Examples

[0502] Take user emotion data as input and generate product or service recommendations when the user is expressing happiness.

[0503] Emotion: Joy

[0504] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0505] Step 1:

[0506] The device captures the user's face in real time using the camera on the smart glasses or smartphone. The input is the camera image, and the output is the captured face image. Specifically, OpenCV is used to acquire the image and perform face detection.

[0507] Step 2:

[0508] The device inputs the captured facial image into an emotion recognition model built with Keras to determine the user's emotional state. The input is a facial image, and the output is emotion data. Specifically, the facial image is preprocessed and input into the emotion recognition model to determine the emotion.

[0509] Step 3:

[0510] The device sends the determined emotion data to the generative AI model's endpoint using the Requests library. The input is emotion data, and the output is recommended products and services returned by the generative AI. Specifically, it generates an HTTP request and sends the emotion data.

[0511] Step 4:

[0512] The server processes the emotional data using a generative AI model and generates recommended products and services that take the user's emotional state into account. The input is emotional data, and the output is recommended products and services. Specifically, the server generates prompt sentences based on the emotional data, which are then input into the generative AI model to obtain recommendations.

[0513] Step 5:

[0514] The terminal displays the recommended products and services returned by the generation AI on the user's device in real time. The input is the recommended products and services, and the output is the information displayed to the user's visual sense. Specifically, it updates the UI to display the recommended information on the screen.

[0515] Example 3

[0516] Next, a description will be given of a third embodiment of the third embodiment. 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."

[0517] Conventional decision-making support systems have difficulty fully considering the decision-making style and emotional state of individual managers, and can only make proposals based on general decision-making models. Furthermore, they are unable to reflect the user's emotional state in real time, making it difficult to provide appropriate decision-making support that responds to stress and emotional fluctuations.

[0518] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0519] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from individual brains to construct a personalized virtual brain, means for collecting user emotional data in real time, and means for adjusting decision-making priorities based on the collected emotional data. This enables appropriate decision-making support that reflects the decision-making style of each manager and corresponds to the emotional state of the user.

[0520] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and models from data.

[0521] A "virtual brain" is a system that uses artificial intelligence technology to mimic the decision-making process of the human brain.

[0522] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0523] "Personalization" refers to optimizing systems and services to suit the characteristics and needs of individual users.

[0524] "Electroencephalogram data" is data obtained by measuring the electrical activity of the brain, and is information that indicates the state and activity of the brain.

[0525] "Behavioral data" is data that records the user's behavior and activities, and is information that indicates the user's behavioral patterns during specific time periods and situations.

[0526] "Emotion data" is data that indicates the user's emotional state, and is information obtained from facial expressions, voice, brain waves, and the like.

[0527] "Decision-making priority" is a criterion for determining which of multiple options or tasks should be executed first.

[0528] "Real-time" means that data is collected and processed immediately, and the results are reflected without delay.

[0529] This invention is a system that uses generative AI to build a virtual brain model for decision-making, and provides a personalized virtual brain by learning and applying data obtained from an individual's brain. It also includes a function to collect user emotional data in real time and adjust decision-making priorities based on the collected emotional data.

[0530] Data collection

[0531] The server collects the manager's brainwave and behavioral data. Specifically, it uses an EEG measurement device (e.g., Emotiv Epoc+) and behavior tracking software (e.g., RescueTime) to obtain the manager's brain activity patterns and daily behavioral data in real time.

[0532] Data Preprocessing

[0533] The server preprocesses the collected data. For example, it removes noise and filters EEG data, and categorizes and tags behavioral data. This is done using Python libraries (e.g., NumPy, Pandas). For example, it analyzes the frequency of application usage during a specific time period.

[0534] Training an AI model

[0535] The server uses the preprocessed data to train the AI ​​model. Specifically, it uses TensorFlow and PyTorch to train the model on data on how managers make decisions in specific business situations. For example, the model can reflect the decision-making patterns of managers based on data on the success and failure of past projects.

[0536] Building a Virtual Brain

[0537] The server builds a personalized virtual brain based on a trained AI model. This virtual brain reflects the individual decision-making style of each manager. Specifically, it replicates the manager's past decision-making patterns and makes the optimal decision in similar situations.

[0538] Collecting Emotional Data

[0539] The device collects the user's emotional data in real time. Specifically, it uses emotion recognition software (e.g., Affectiva) to analyze the user's emotional state from their facial expressions and voice. For example, it can detect whether the user is feeling stressed in real time.

[0540] Applying Emotion Data

[0541] The server adjusts decision-making priorities based on the collected emotional data. For example, if the user is feeling stressed, the virtual brain will prioritize options to reduce stress, such as reprioritizing tasks or suggesting taking a break.

[0542] Specific examples

[0543] Example 1: Decision making in a business situation

[0544] Suppose a user is considering starting a new project. The virtual brain uses past data to learn how managers have made decisions in similar situations and suggests optimal decisions. For example, it suggests strategies to minimize risk based on data from past successful projects.

[0545] Example 2: Applying Emotion Data

[0546] If the user is feeling stressed, the virtual brain suggests options to reduce stress (e.g., reprioritizing tasks, taking a break). For example, if the user is feeling stressed before an important meeting, the virtual brain suggests taking a short break to relax.

[0547] Prompt Sentence Examples

[0548] "Please build an AI model that uses the brainwave and behavioral data of business managers to learn how to make decisions in specific business situations. Also, please add a function that adjusts the priority of decision-making by taking into account the user's emotional data."

[0549] In this way, a personalized virtual brain that reflects the manager's decision-making style is constructed, and a system that provides decision-making according to the user's emotional state is completed. The flow of the specific processing in Example 3 will be described with reference to FIG.

[0550] Step 1:

[0551] The server collects the manager's EEG data and behavioral data. Specifically, it acquires EEG data using an EEG measurement device (e.g., Emotiv Epoc+) and collects behavioral data using behavioral tracking software (e.g., RescueTime). The input is data from the EEG measurement device and behavioral tracking software, and the output is the collected raw data.

[0552] Step 2:

[0553] The server preprocesses the collected data. It performs noise removal and filtering on EEG data, and categorizes and tags behavioral data. Specifically, it cleans the data using Python libraries (e.g., NumPy, Pandas). The input is the collected raw data, and the output is the preprocessed, clean data.

[0554] Step 3:

[0555] The server uses the preprocessed data to train the AI ​​model. Specifically, it uses TensorFlow and PyTorch to train data on the decisions managers made in specific business situations. The input is the preprocessed data, and the output is the trained AI model.

[0556] Step 4:

[0557] The server builds a personalized virtual brain based on a trained AI model. Specifically, it reproduces the decision-making patterns of managers in the past, and the virtual brain makes optimal decisions in similar situations. The input is the trained AI model, and the output is the personalized virtual brain.

[0558] Step 5:

[0559] The device collects the user's emotional data in real time. Specifically, it uses emotion recognition software (e.g., Affectiva) to analyze the user's emotional state from their facial expressions and voice. The input is the user's facial and voice data, and the output is the analyzed emotional data.

[0560] Step 6:

[0561] The server adjusts decision-making priorities based on the collected emotional data. Specifically, if the user is feeling stressed, the virtual brain prioritizes options that will reduce stress. The input is the analyzed emotional data, and the output is the adjusted decision-making priorities.

[0562] (Application example 3)

[0563] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0564] In modern factories, improving production line efficiency and troubleshooting are important issues. However, solving these issues requires an advanced system that reflects the manager's decision-making style. It is also necessary to adjust the workload taking into account the emotional state of the operator. Conventional systems have difficulty meeting these requirements, making it difficult to achieve efficient production management.

[0565] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0566] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from individual brains to construct a personalized virtual brain, means for optimizing production lines in factories, means for providing optimal solutions to problems that occur in factories, and means for monitoring the emotional state of operators in real time and adjusting their workloads.This enables efficient production management that reflects the decision-making style of managers and adjustment of workloads that take into account the emotional state of operators.

[0567] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and models from data.

[0568] A "virtual brain" is an artificial intelligence model built based on an individual's brainwave and behavioral data, which mimics a specific decision-making style.

[0569] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0570] "Personalization" means customizing something to meet specific needs and requirements based on individual characteristics and data.

[0571] "Production line optimization" means maximizing production efficiency by efficiently arranging production processes within a factory and optimizing work flow.

[0572] "Troubleshooting" is a technique for quickly identifying and resolving problems or failures that occur in systems or processes.

[0573] "Emotional state" refers to an individual's psychological state or mood, and includes factors such as stress and satisfaction.

[0574] "Workload" refers to the amount of effort or time required for a particular task or job.

[0575] To implement the present invention, the following system configuration and program are required.

[0576] System Configuration

[0577] 1. Hardware:

[0578] Factory robots: Robots used to carry out production line tasks.

[0579] Brainwave sensor: A device for collecting brainwave data from business managers.

[0580] Emotion Monitoring Device: A device for monitoring the emotional state of the operator in real time.

[0581] 2. Software:

[0582] Python: A programming language used to implement the program.

[0583] scikit-learn: A library used to build machine learning models.

[0584] NumPy: A library used for data preprocessing and numerical computation.

[0585] Program processing explanation

[0586] The server first collects the manager's brainwave and behavioral data, and then uses this data to build a virtual brain. The virtual brain is a model that uses generative AI to mimic the manager's decision-making style. This model is used to optimize production lines and troubleshoot problems within factories.

[0587] Specifically, the server performs the following process.

[0588] 1. Data Collection:

[0589] Brainwave data of managers is collected using an EEG sensor.

[0590] Emotional data from operators is collected using an emotion monitoring device.

[0591] 2. Data Preprocessing:

[0592] Normalize the collected data using NumPy.

[0593] 3. Train the model:

[0594] We build a virtual brain using scikit-learn's MLPRegressor.

[0595] The system learns the brainwave and behavioral data of managers.

[0596] 4. Prediction and optimization:

[0597] Input data for a new production line and the virtual brain predicts the optimal layout.

[0598] We provide optimal solutions to problems that occur within the factory.

[0599] 5. Considering emotional data:

[0600] Monitor operators' emotional state in real time and adjust their workload if stress levels increase.

[0601] Specific examples

[0602] For example, if a manager learns from data on past decisions made to resolve bottlenecks on a production line and inputs the current production line data, the virtual brain will suggest optimal layouts. It will also make decisions to reduce the workload of operators if they are feeling stressed.

[0603] Prompt Sentence Examples

[0604] "Using the brainwave and behavioral data of managers, please create a virtual brain that can optimize production lines in factories. When you input data on a new production line, please create a model that will suggest the optimal layout."

[0605] In this way, by applying a virtual brain that reflects the decision-making style of managers to factory robots, we can expect to improve production efficiency and speed up troubleshooting.

[0606] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0607] Step 1:

[0608] The server collects the manager's brainwave data using an EEG sensor. Specifically, the EEG sensor is worn on the manager's head and captures brainwave data in real time. This data is sent to the server and stored in a database. The input is raw data from the EEG sensor, and the output is stored brainwave data.

[0609] Step 2:

[0610] The server collects the operator's emotional data using an emotion monitoring device. Specifically, the emotion monitoring device is worn on the operator's body and captures physiological data such as heart rate and galvanic skin response. This data is sent to the server and stored in a database. The input is the raw data from the emotion monitoring device, and the output is the stored emotional data.

[0611] Step 3:

[0612] The server preprocesses the collected EEG and emotion data. Specifically, it uses NumPy to standardize the data, fill in missing values, and remove noise. The input is the stored EEG and emotion data, and the output is the preprocessed data.

[0613] Step 4:

[0614] The server uses the preprocessed data to construct a virtual brain. Specifically, it trains a generative AI model using scikit-learn's MLPRegressor. The input is preprocessed EEG data and behavioral data, and the output is a trained virtual brain model.

[0615] Step 5:

[0616] The server receives data on the new production line as input and uses the virtual brain model to predict the optimal layout. Specifically, data on each station on the production line is input, and the virtual brain model calculates the optimal layout. The input is the data on the new production line, and the output is the optimized production line layout.

[0617] Step 6:

[0618] The server provides optimal solutions to problems that occur within the factory. Specifically, data on the type of problem and its location is input, and the virtual brain model proposes optimal solutions based on past data. The input is data on the problem, and the output is the proposed solution.

[0619] Step 7:

[0620] The server monitors the emotional state of the operator in real time and adjusts the workload. Specifically, it analyzes data from the emotion monitoring device in real time and issues instructions to reduce the workload if stress levels are high. The input is real-time emotional data, and the output is the adjusted workload.

[0621] 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.

[0622] The data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of the data generation model 58 is Chat (registered trademark) (Internet search<URL: https: / / openai.com / blog / chatgpt> ) 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.

[0623] Another example of generative AI is Gemini (registered trademark) (Internet search engine). <url: https: gemini.google.com ?hl="ja">) are mentioned.

[0624] 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.

[0625] [Second embodiment]

[0626] FIG. 3 shows an example of the configuration of a data processing system 210 according to the second embodiment.

[0627] 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.

[0628] 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).

[0629] 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.

[0630] 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.

[0631] 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).

[0632] 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.

[0633] 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.

[0634] 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.

[0635] 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.

[0636] 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.

[0637] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[0638] "Example 1"

[0639] In one embodiment of the present invention, the means for constructing a virtual brain model using generative AI includes an AI algorithm for learning the decision-making patterns of managers. Specifically, data on past decisions made by managers is used as learning data, and the AI ​​learns those patterns.

[0640] "Example 2"

[0641] Furthermore, to speed up the PDCA cycle and improve the accuracy of the model, AI periodically updates its learning data and learns new decision-making patterns, enabling it to make decisions that respond to changes in the business environment.

[0642] "Example 3"

[0643] Furthermore, as a means of learning and applying data obtained from an individual's brain, a personalized virtual brain is constructed by acquiring the manager's brainwave data and behavioral data and using this as learning data for the AI. This makes it possible to construct a virtual brain that reflects the individual decision-making style of each manager. As a specific example, the AI ​​can learn data on how a manager made decisions in specific business situations, and the virtual brain can make decisions based on that data.

[0644] The processing flow of each embodiment will be described below.

[0645] "Example 1"

[0646] Step 1: Collect data on management's past decisions. This includes information on the results of decisions made by management in the past and the circumstances at the time.

[0647] Step 2: The collected data is input into the generative AI, which then learns the manager's decision-making patterns. This learning process builds a model for the AI ​​to imitate the manager's decision-making style.

[0648] Step 3: Use the constructed model to predict managerial decisions for new business situations.

[0649] "Example 2"

[0650] Step 1: The AI ​​periodically updates its learning data by adding newly acquired managerial decision-making data to the learning data.

[0651] Step 2: Using the updated learning data, the AI ​​learns new decision-making patterns. Step 3: Using the new decision-making patterns it has learned, it predicts decisions that respond to changes in the business environment.

[0652] "Example 3"

[0653] Step 1: Obtain the manager's brainwave data, behavioral data, etc. These data are used to reflect the manager's individual decision-making style.

[0654] Step 2: The acquired data is used as training data for the AI, which then learns the decision-making styles of individual managers.

[0655] Step 3: Build a personalized virtual brain using the learned decision-making styles of individual managers.

[0656] Example 1

[0657] Next, a description will be given of Example 1 of Form 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."

[0658] Conventional decision support systems lacked personalization and were unable to fully reflect the individual decision-making patterns of managers. Furthermore, data collection, preprocessing, model training, and evaluation were often performed manually, resulting in inefficiency and low accuracy. Furthermore, generating and inputting prompts was cumbersome, and the output of prediction results was not intuitive, making it difficult for managers to make quick decisions.

[0659] 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.

[0660] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for collecting past decision-making data and storing it in a database, means for preprocessing the collected data, means for training the AI ​​model using the preprocessed data, means for evaluating the trained model, means for generating prompt sentences and inputting them into the AI ​​model, and means for outputting prediction results.This allows for the construction of a personalized virtual brain that reflects the individual decision-making patterns of a manager, enabling efficient and highly accurate decision-making support.

[0661] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new data and models.

[0662] A "virtual brain" is a system that uses artificial intelligence technology to mimic the functions of the human brain and make decisions.

[0663] "PDCA" refers to the cycle of planning (Plan), doing (Do), checking (Check), and improving (Act), and is a method for making continuous improvements.

[0664] "Personalization" refers to optimizing systems and services based on the characteristics and behavioral patterns of individual users.

[0665] A "database" is a system for efficiently storing, managing, and retrieving data.

[0666] "Preprocessing" is the process of cleaning and transforming data before data analysis or machine learning.

[0667] "Training" is the process by which a machine learning model learns patterns from data.

[0668] "Evaluation" is the process of measuring the performance of a trained model using validation data.

[0669] A "prompt" is an instruction given to a generative AI model, containing instructions for performing a specific task.

[0670] "Prediction result" refers to the result that a generative AI model outputs based on a prompt, including predictions of future events or data.

[0671] This invention is a system for constructing a virtual brain model that uses generative AI to support managerial decision-making. Specific embodiments of this system are described below.

[0672] First, the server collects data on past decisions made by management, including sales data, market trends, economic indicators, competitor trends, etc. The server stores this data in a database management system (e.g., MySQL).

[0673] The server then preprocesses the collected data, using the Python Pandas library to impute missing values, normalize the data, and detect and remove outliers, thereby improving the quality of the data.

[0674] The server then uses the preprocessed data to train an AI model, using a machine learning framework (e.g., TensorFlow) to learn the decision-making patterns of managers. Training involves using past decision-making data as input so the model can recognize patterns.

[0675] After training is complete, the server evaluates the trained model using a validation dataset to measure the model's precision and recall, allowing the server to verify the model's performance and make adjustments if necessary.

[0676] The user inputs a prompt to the generative AI model. For example, the user inputs a prompt such as, "Please predict sales for the next quarter based on sales data from the past five years and market trends." The input prompt is sent to the server.

[0677] The server runs the AI ​​model based on the input prompt and generates a prediction result. The prediction result is displayed to the user. The user can use this prediction result as a reference to develop future management strategies.

[0678] For example, data on a manager's past decisions may include the following:

[0679] Sales Data

[0680] market trends

[0681] Economic indicators

[0682] Competitor trends

[0683] The server collects this data and inputs it into the AI ​​model. For example, the following prompt sentence is input into the generative AI model:

[0684] "Please forecast sales for the next quarter based on sales data from the past five years and market trends."

[0685] Based on this prompt, the AI ​​model analyzes past data and provides a sales forecast for the next quarter, which users can use to plan future business strategies.

[0686] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0687] Step 1: Data collection

[0688] The server collects data on past decisions made by management. Specifically, it obtains data such as sales data, market trends, economic indicators, and competitor trends using APIs and web scraping. This data is saved as a CSV file and imported into a database management system (e.g., MySQL). The input is raw data from various data sources, and the output is structured data stored in the database.

[0689] Step 2: Data Preprocessing

[0690] The server preprocesses the collected data. Specifically, it uses the Python Pandas library to impute missing values, normalize the data, and detect and remove outliers. For example, missing values ​​are imputed with the mean value, and data normalization uses min-max scaling. The input is the raw data stored in the database, and the output is the preprocessed, clean data.

[0691] Step 3: Train the model

[0692] The server trains an AI model using the preprocessed data. It uses a machine learning framework (e.g., TensorFlow) to build a neural network model. It processes the training data in batches and updates the model weights every epoch. The input is the preprocessed data, and the output is the trained AI model.

[0693] Step 4: Evaluate the model

[0694] The server evaluates the trained model. For evaluation, a validation dataset is used to measure the model's precision and recall. Evaluation metrics include mean absolute error (MAE) and coefficient of determination (R²). The input is the trained AI model and the validation dataset, and the output is the evaluation result.

[0695] Step 5: Generate and enter a prompt

[0696] The user inputs a prompt to the generative AI model. For example, the prompt might be, "Please predict sales for the next quarter based on sales data from the past five years and market trends." The input is the prompt entered by the user, and the output is the prompt sent to the server.

[0697] Step 6: Output prediction results

[0698] The server runs the AI ​​model based on the input prompt and generates a prediction result. The prediction result is displayed in graph or table format on a web interface. The input is the prompt and the trained AI model, and the output is the prediction result. Users can use this prediction result as a reference to develop future management strategies.

[0699] (Application example 1)

[0700] Next, a description will be given of Application Example 1 of Form 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."

[0701] With conventional production management systems, it was difficult to mimic management decisions in real time and perform optimal production management. Furthermore, there was a lack of means to effectively utilize management's past decision-making data, making it impossible to achieve efficient production management. Furthermore, systems for collecting and analyzing data in real time were complex, resulting in significant costs for implementation and operation.

[0702] 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.

[0703] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from individual brains to construct a personalized virtual brain, means for collecting and storing data from production lines in factories in real time, means for preprocessing the collected data and training the AI ​​model, and means for analyzing production line data in real time using the trained AI model to make optimal decisions. This enables efficient production management by imitating management decision-making in real time.

[0704] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new data and models.

[0705] A "virtual brain" is a model that uses artificial intelligence technology to mimic the decision-making process of the human brain.

[0706] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0707] "Personalization" refers to the individual optimization of systems and services based on the characteristics and behavior of individual users.

[0708] A "production line" refers to a series of machines or work stations arranged to efficiently produce a product.

[0709] "Real-time data collection" refers to obtaining data immediately from ongoing processes or events.

[0710] "Data preprocessing" refers to processing collected data to convert it into a format suitable for analysis and model training.

[0711] "Training an AI model" refers to the learning process of using collected data to improve the performance of an artificial intelligence model.

[0712] "Making optimal decisions" refers to determining the most effective actions or choices based on collected data and trained AI models.

[0713] A system for implementing this invention has the following configuration. First, the server uses a generative AI to build a virtual brain model that makes decisions. Specifically, the manager's past decision-making data is used as learning data, and the generative AI learns the patterns. This builds a virtual brain that mimics the manager's decision-making.

[0714] Next, the server has the means to speed up the PDCA cycle and improve the accuracy of the model, which allows the virtual brain model to be continuously improved and enable more accurate decision-making.

[0715] Furthermore, the server has the means to learn and apply data obtained from individual brains to build a personalized virtual brain, thereby providing optimized decision-making support for each manager.

[0716] The server also includes a means for collecting and storing real-time data from the production line in the factory. Specifically, sensors are used to obtain data on product quality, production speed, machine operation status, etc., and the data is stored in an SQL database.

[0717] The server also includes a means to preprocess the collected data and train the AI ​​model. Python is used to fill in missing values ​​and remove outliers, and TensorFlow is used to train the AI ​​model.

[0718] The server also includes a means to analyze production line data in real time using trained AI models to make optimal decisions, suggesting actions to adjust production speed or control quality.

[0719] For example, if product quality is declining, the AI ​​model will learn from past data what measures management has taken and suggest similar measures, such as temporarily slowing down production speed or performing maintenance on specific machines.

[0720] An example of a prompt is, "Please learn about the measures taken by management when product quality deteriorated in the past, and propose the optimal measures based on the current production line data."

[0721] In this way, the server can mimic the manager's decision-making in real time and achieve efficient production management.

[0722] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0723] Step 1:

[0724] The server collects data in real time from the production line in the factory. Specifically, sensors are used to obtain data on product quality, production speed, machine operation status, etc. The input is raw data from the sensors, and the output is data stored in an SQL database.

[0725] Step 2:

[0726] The server preprocesses the collected data. Specifically, it uses Python to fill in missing values ​​and remove outliers. The input is raw data stored in an SQL database, and the output is the preprocessed, clean data.

[0727] Step 3:

[0728] The server uses the preprocessed data to train the AI ​​model. Specifically, it uses TensorFlow to learn from managers' past decision-making data and build a generative AI model. The input is the preprocessed data, and the output is the trained AI model.

[0729] Step 4:

[0730] The server uses the trained AI model to analyze production line data in real time. Specifically, it inputs current production line data into the AI ​​model to make optimal decisions. The input is production line data collected in real time, and the output is a proposal for optimal decisions.

[0731] Step 5:

[0732] The server adjusts the production line based on optimal decision-making. Specifically, it proposes actions for adjusting production speed and quality control. The input is the decision-making proposal from the AI ​​model, and the output is the adjustment results for the production line.

[0733] Step 6:

[0734] The user checks the suggestions from the server and manually adjusts them as necessary. Specifically, based on the suggestions from the server, the user can take actions such as temporarily slowing down the production speed or performing maintenance on a specific machine. The input is the suggestion from the server, and the output is the result of the adjustment made by the user.

[0735] Step 7:

[0736] The server collects the adjustment results as data again and saves them as training data for the next run. This completes the PDCA cycle, improving the accuracy of the model. The input is the adjustment results made by the user, and the output is the training data for the next run.

[0737] Example 2

[0738] Next, a description will be given of Example 2 of Form 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."

[0739] Today's business environment is rapidly changing, requiring managers to make quick and accurate decisions. However, traditional decision support systems often require manual data updates and model retraining, making it difficult to quickly implement the PDCA cycle. They also lack personalized support that reflects the decision-making patterns of individual managers. This makes it difficult for managers to make quick decisions to adapt to a changing environment.

[0740] 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.

[0741] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for periodically updating the learning data and learning new decision-making patterns, means for retraining and evaluating the generative AI model, means for inputting prompt sentences to request specific predictions or proposals, and means for providing the generated predictions or proposals, thereby enabling managers to make quick and accurate decisions.

[0742] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and patterns from data.

[0743] A "virtual brain" is a model that uses artificial intelligence technology to mimic the decision-making process of the human brain.

[0744] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0745] "Personalization" refers to optimizing systems and services to suit the characteristics and needs of individual users.

[0746] "Training data" is a data set used to train an artificial intelligence model.

[0747] "Retraining" is the process of adding new data to an existing artificial intelligence model and training it again.

[0748] "Evaluation" is the process of measuring the performance of an artificial intelligence model and confirming its accuracy and effectiveness.

[0749] A "prompt sentence" is an instruction sentence that a user enters to request a specific prediction or suggestion from the system.

[0750] "Prediction" refers to an artificial intelligence model's estimation of future events or outcomes based on past data.

[0751] A "suggestion" is a specific recommendation of action or strategy that an AI model provides to a user.

[0752] This invention is a system that uses generative AI to build a virtual brain model for decision-making and improves the accuracy of the model by quickly implementing the PDCA cycle. Furthermore, it aims to build a personalized virtual brain by learning and applying data obtained from individual brains.

[0753] Hardware and software used

[0754] The server uses a computer equipped with a high-performance GPU (e.g., NVIDIA Tesla V100), and the software uses a machine learning framework (e.g., TensorFlow, PyTorch), which enables rapid processing of large amounts of data and efficient retraining of generative AI models.

[0755] Program processing

[0756] The server first collects data provided by the user, including past sales data, seasonal fluctuation data, competitor trends, etc. This data is uploaded in CSV or Excel file format.

[0757] The server then preprocesses the collected data, specifically by completing missing values, detecting and correcting outliers, and normalizing the data, thereby improving the quality of the data and the accuracy of the model.

[0758] The server retrains the generative AI model using the preprocessed data. For retraining, a dataset divided into training data and test data is used. The server initializes the generative AI model and retrains the model using the training data. After retraining, the performance of the model is evaluated using the test data.

[0759] Once the evaluation is complete, the model is used to assist the user in making decisions. The user inputs a prompt to request a specific prediction or suggestion. The server uses the generative AI model to generate the requested prediction or suggestion and provides it to the user.

[0760] Specific examples

[0761] Consider a company that wants to forecast market demand. Users upload sales data from the past three years, seasonal fluctuation data, and competitor trends to a server. The server uses this data to retrain a generative AI model and generate demand forecasts for the next quarter.

[0762] Example prompt sentence:

[0763] "Retrain your generative AI model using the past three years of sales data, seasonal fluctuation data, and competitor trends data to forecast demand for product A for the next quarter."

[0764] By entering this prompt, the server automatically collects data, retrains the model, and provides demand forecasts, allowing businesses to make fast and accurate decisions.

[0765] The flow of the identification process in the second embodiment will be described with reference to FIG.

[0766] Step 1: Data collection

[0767] Users upload past sales data, seasonal fluctuation data, and competitor trends data to the server. Data in CSV or Excel file format is used as input. The server receives these files and stores them in the database. Specifically, the server validates the file format and stores them appropriately in the database.

[0768] Step 2: Data Preprocessing

[0769] The server pre-processes the collected data. As input, it uses the raw data stored in the database. Specifically, the server:

[0770] Imputing missing values: Imputing missing values ​​with the mean or median.

[0771] Outlier detection and correction: Statistical methods are used to detect outliers and correct them appropriately.

[0772] Data normalization: Standardizing or normalizing data.

[0773] As an output, pre-processed data is obtained.

[0774] Step 3: Retrain the model

[0775] The server retrains the generative AI model using the preprocessed data. The preprocessed data is used as input. Specifically, the server performs the following operations:

[0776] Data Split: Split the preprocessed data into training data and test data.

[0777] Initialize model: Initialize the generative AI model.

[0778] Train the model: Retrain the model using the training data.

[0779] The output is a retrained generative AI model.

[0780] Step 4: Model evaluation

[0781] The server evaluates the performance of the retrained model. The test data and the retrained model are used as input. Specifically, the server performs the following operations:

[0782] Run prediction: Make a prediction for the model using the test data.

[0783] Calculate evaluation metrics: Calculate evaluation metrics such as precision, recall, and F1 score.

[0784] As an output, the evaluation result is obtained.

[0785] Step 5: Decision support

[0786] The user inputs a prompt to request a specific prediction or suggestion. The prompt and the retrained model are used as input. Specifically, the server performs the following actions:

[0787] Parsing prompts: Parse the prompts entered by the user.

[0788] Generate predictions and recommendations: Use generative AI models to generate the requested predictions and recommendations.

[0789] As an output, the generated predictions and suggestions are provided to the user.

[0790] Example prompt sentence:

[0791] "Retrain your generative AI model using the past three years of sales data, seasonal fluctuation data, and competitor trends data to forecast demand for product A for the next quarter."

[0792] (Application example 2)

[0793] Next, a description will be given of Application Example 2 of Form 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."

[0794] Inventory management and delivery route optimization are important issues for modern logistics centers. However, traditional manual management methods have limitations when it comes to quickly responding to changes in the business environment. In particular, delays in updating inventory and delivery data make efficient operations difficult, resulting in increased costs and decreased customer satisfaction. To solve these issues, an automated decision-making support system using AI is needed.

[0795] 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 constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for periodically updating inventory data and delivery data of the logistics center and proposing optimal inventory management and delivery routes, and means for periodically updating data and relearning the model to make decisions in response to changes in the business environment. This makes it possible to optimize inventory management and delivery routes at the logistics center, enabling rapid response to changes in the business environment.

[0796] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and patterns from data.

[0797] A "virtual brain" is a model that uses artificial intelligence technology to mimic the decision-making process of the human brain.

[0798] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0799] "Personalization" refers to the individual optimization of services and content based on the characteristics and behavior of individual users.

[0800] "Inventory data" refers to data that includes information such as the quantity of products at the logistics center, arrival date, and shipping date.

[0801] "Delivery data" refers to data that includes information such as the delivery route, delivery time, and destination of the product departing from the logistics center.

[0802] The "business environment" refers to the external and internal factors that affect a company's business operations, including market trends, competitive conditions, and regulations.

[0803] A "decision support system" is a system that uses data analysis and artificial intelligence technology to provide information and suggestions to managers and executives to help them make optimal decisions.

[0804] The system for implementing this invention aims to optimize inventory management and delivery routes in logistics centers. The system uses generative AI to build a virtual brain model for decision-making, speeding up the PDCA cycle and improving the accuracy of the model. It also learns and applies data obtained from individual brains to build a personalized virtual brain.

[0805] The server uses the following hardware and software:

[0806] Hardware: Server or cloud environment (e.g. AWS, Google Cloud)

[0807] Software: Python, pandas (data manipulation), scikit-learn (machine learning), schedule (regular execution)

[0808] The server first periodically updates inventory and delivery data, allowing it to propose optimal inventory management and delivery routes based on the latest data. Specifically, inventory data includes information such as product ID, stock quantity, arrival date, and shipping date, while delivery data includes information such as delivery ID, departure point, destination, and delivery time.

[0809] The server then uses generative AI to build a virtual brain model. This model is then incorporated into the "Smart Logistics Manager" smartphone application used by logistics center operators. The application provides optimized inventory management, delivery routes, real-time data updates and analysis, and decision-making support in response to changes in the business environment.

[0810] The server periodically updates the data and retrains the model to quickly respond to changes in the business environment, enabling optimization of inventory management and delivery routes at logistics centers and achieving efficient operations.

[0811] As a concrete example, the following prompt sentence can be used:

[0812] "Use inventory and delivery data from your distribution center to create an AI model that suggests optimal inventory management and delivery routes. The model will update the data regularly and make decisions that respond to changes in the business environment."

[0813] In this way, the server can provide the logistics center operator with information to help them make optimal decisions.

[0814] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[0815] Step 1:

[0816] The server reads inventory data and delivery data. Specifically, the server obtains inventory data (product ID, stock quantity, arrival date, shipping date, etc.) and delivery data (delivery ID, departure point, destination, delivery time, etc.) from a CSV file or database. The input is inventory data and delivery data, and the output is a data frame containing these data.

[0817] Step 2:

[0818] The server preprocesses the acquired data, specifically by imputing missing values, normalizing data, encoding categorical data, etc. The input is the data frame acquired in step 1, and the output is the preprocessed data frame.

[0819] Step 3:

[0820] The server trains a generative AI model using the preprocessed data. Specifically, it splits the data into training data and test data, and trains it using a random forest regression model. The input is the preprocessed data frame, and the output is the trained generative AI model.

[0821] Step 4:

[0822] The server uses a trained generative AI model to optimize inventory management and delivery routes. Specifically, it inputs new data into the model and predicts optimal inventory levels and delivery routes. The input is new inventory data and delivery data, and the output is a proposal for optimal inventory levels and delivery routes.

[0823] Step 5:

[0824] The server evaluates the accuracy of the generative AI model. Specifically, it evaluates the model's predictive accuracy using test data and adjusts the model's parameters as necessary. The input is the test data and the trained generative AI model, and the output is the model's accuracy evaluation result.

[0825] Step 6:

[0826] The server periodically updates the data and retrains the model. Specifically, it uses a schedule library to perform data updates and model retraining at midnight every day. The input is the latest inventory and delivery data, and the output is an updated generative AI model.

[0827] Step 7:

[0828] The user uses the smartphone application "Smart Logistics Manager" to check the optimal inventory management and delivery route proposals provided by the server. Specifically, the application retrieves data from the server and displays it to the user. The input is the proposal data from the server, and the output is the optimization proposal displayed to the user.

[0829] Example 3

[0830] Next, a description will be given of Example 3 of Form Example 3. 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."

[0831] Conventional decision support systems have difficulty reflecting the decision-making styles of individual managers, and can only provide support based on general algorithms. Therefore, personalized support that takes into account the individual decision-making criteria and styles of managers is required. In addition, efficient and accurate methods are required for preprocessing collected data and model training.

[0832] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[0833] In this invention, the server includes means for collecting individual electroencephalogram data, means for collecting behavioral data, means for preprocessing the collected electroencephalogram data and behavioral data, means for training a generative AI model using the preprocessed data, means for constructing a personalized virtual brain using the trained generative AI model, and means for displaying the decision-making results of the virtual brain, thereby enabling personalized support that reflects the individual decision-making style of the manager.

[0834] "Individual electroencephalogram data" refers to data of electrical signals acquired to measure an individual's brain activity.

[0835] "Behavioral data" refers to data that records an individual's behavior, operation history, activity log, etc.

[0836] "Preprocessing" is the process of performing processes such as noise removal and normalization on collected data to convert it into a format suitable for machine learning models.

[0837] A "generative AI model" is an artificial intelligence model that is generated using machine learning algorithms to learn from data and perform specific tasks.

[0838] A "personalized virtual brain" is a virtual brain model designed to reflect the decision-making style of an individual manager.

[0839] "Decision-making results" are the results of judgments and choices made by the virtual brain in a specific business situation.

[0840] "Display means" refers to devices or software that visually present the decision-making results of the virtual brain to the user.

[0841] This invention is a system that collects individual electroencephalogram (EEG) data and behavioral data, trains a generative AI model based on the data, and constructs a personalized virtual brain. Specific embodiments of this system are described below.

[0842] First, the user wears an EEG device to collect the manager's brainwave data. The device acquires the brainwave data from the EEG device in real time and sends it to a server. For example, a manager wears the EEG device during a meeting, collecting brainwave data during that time. The user also installs dedicated software on their PC or smartphone to collect the manager's behavioral data. The device records the manager's operation history and activity log and sends it to the server. For example, it collects the operation history when the manager checks email or edits documents.

[0843] The server then stores the received EEG and behavioral data in a database. The server then preprocesses the data, specifically removing noise and normalizing the data to convert it into a format suitable for machine learning models. For example, the server filters the EEG data and checks the consistency of the timestamps of the behavioral data. The server uses a filtering algorithm to remove noise from the EEG data and standardize the timestamps of the behavioral data.

[0844] The server then uses the preprocessed data to train a generative AI model. Specifically, it uses machine learning libraries such as Python's TensorFlow and PyTorch. The server trains the model based on past data to learn the manager's decision-making style. For example, it trains the model using brainwave and behavioral data from when a manager decides whether to enter a new market. The server generates a training dataset and optimizes the model's parameters.

[0845] Next, the server uses the trained generative AI model to construct a virtual brain that reflects the manager's decision-making style. The server sets the virtual brain's parameters and simulates the manager's decision-making in specific business situations. For example, when the virtual brain decides whether to enter a new market, it reproduces the decision-making process based on past data. The server saves the results of the virtual brain's simulation and provides them to the user.

[0846] Finally, the user applies the constructed virtual brain to a real business situation. The terminal displays the virtual brain's decision-making results and provides them to the user. For example, the user can refer to the virtual brain's judgment when considering whether to enter a new market. The terminal displays the virtual brain's decision-making results in graphs and text format, making them easy for the user to understand.

[0847] Examples of prompts include:

[0848] "When a manager decides whether to enter a new market, the virtual brain learns what factors he or she considers based on past brainwave and behavioral data, and then makes a decision in a similar situation."

[0849] In this way, the server collects the manager's EEG data and behavioral data and uses the generative AI model to build a personalized virtual brain. This enables the virtual brain, which reflects the manager's decision-making style, to make decisions in specific business situations. The flow of the specific processing in Example 3 will be explained using Figure 15.

[0850] Step 1:

[0851] The user wears an EEG device to collect the executive's brainwave data. The terminal acquires the brainwave data from the EEG device in real time and transmits it to the server. The input is the brainwave data from the EEG device, and the output is the brainwave data transmitted to the server. Specifically, the executive wears the EEG device during a meeting, and brainwave data is collected during that time.

[0852] Step 2:

[0853] To collect behavioral data on managers, users install dedicated software on their PCs or smartphones. The devices record the manager's operation history and activity logs and send them to a server. The input is the operation history and activity logs from the PC or smartphone, and the output is the behavioral data sent to the server. Specifically, the software collects operation history when the manager checks email or edits documents.

[0854] Step 3:

[0855] The server stores the received EEG data and behavioral data in a database. The input is the EEG data and behavioral data sent to the server, and the output is the data stored in the database. Specifically, the server creates a new entry in the database and stores the received data.

[0856] Step 4:

[0857] The server preprocesses the data. Specifically, it removes noise and normalizes the data, converting it into a format suitable for machine learning models. The input is EEG data and behavioral data stored in a database, and the output is the preprocessed data. Specifically, the server uses a filtering algorithm to remove noise from the EEG data and standardize the timestamps of the behavioral data.

[0858] Step 5:

[0859] The server uses the preprocessed data to train the generative AI model. Specifically, it uses machine learning libraries such as Python's TensorFlow and PyTorch. The input is the preprocessed EEG data and behavioral data, and the output is a trained generative AI model. Specifically, the server generates a training dataset and optimizes the model parameters.

[0860] Step 6:

[0861] The server uses a trained generative AI model to build a virtual brain that reflects the manager's decision-making style. The input is the trained generative AI model, and the output is the virtual brain. Specifically, the server sets the parameters of the virtual brain and simulates the manager's decision-making in specific business situations.

[0862] Step 7:

[0863] The user applies the constructed virtual brain to actual business situations. The terminal displays the decision-making results of the virtual brain and provides them to the user. The input is the decision-making results of the virtual brain, and the output is the results displayed on the terminal. In concrete terms, the terminal displays the decision-making results of the virtual brain in graphs and text format, making them easy for the user to understand.

[0864] (Application example 3)

[0865] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[0866] In conventional factory operations, it is difficult to optimize production lines and troubleshoot problems that reflect the manager's decision-making style, which hinders efficient operations. In addition, there is no system that can mimic the manager's decision-making, which makes it difficult to respond quickly when the manager is absent.

[0867] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[0868] In this invention, the server includes a means for constructing a virtual brain model that makes decisions using generative AI, a means for speeding up the PDCA cycle and improving the accuracy of the model, a means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, and a means for using the virtual brain that reflects the manager's decision-making style to optimize and troubleshoot production lines in the factory, thereby enabling efficient factory operation that reflects the manager's decision-making style.

[0869] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and models from data.

[0870] A "virtual brain" is an artificial intelligence model built based on an individual's brainwave and behavioral data, which mimics a specific decision-making style.

[0871] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0872] "Personalization" refers to individually optimizing systems and services based on individual characteristics and behavior.

[0873] "Production line optimization" refers to making adjustments and improvements to streamline production processes within a factory and achieve maximum productivity.

[0874] "Troubleshooting" refers to procedures and methods for identifying and resolving problems or failures that occur in a system or process.

[0875] "Managerial decision-making style" refers to the tendency or pattern of how managers make judgments and choices in specific situations.

[0876] To implement the present invention, the following system configuration and program are required.

[0877] System Configuration

[0878] 1. Hardware:

[0879] Brainwave sensor: A device for acquiring brainwave data from business managers.

[0880] Behavioral tracking system: A device for capturing behavioral data of business managers.

[0881] Server: A computer system for processing data and constructing a virtual brain.

[0882] Factory robots: Robots for production line optimization and troubleshooting.

[0883] 2. Software:

[0884] Data collection software: Programs for collecting data from EEG sensors and behavioral tracking systems.

[0885] Generative AI model: An algorithm for constructing a virtual brain based on collected data. Specifically, it uses a multi-layer perceptron (MLPClassifier) ​​using the scikit-learn library.

[0886] Optimization software: A program that uses a virtual brain to optimize and troubleshoot production lines.

[0887] Program processing explanation

[0888] The server first collects the manager's brainwave and behavioral data from EEG sensors and a behavioral tracking system. This data is then sent to the server via data collection software. The server then uses a generative AI model to build a virtual brain based on the collected data. This virtual brain is designed to mimic the manager's decision-making style.

[0889] After the virtual brain is built, the server uses optimization software to optimize and troubleshoot the production line. Specifically, the virtual brain uses the current state of the production line in the factory as input and makes decisions to make optimal adjustments. When a problem occurs, the virtual brain uses the problem data as input and proposes the optimal solution.

[0890] Specific examples

[0891] For example, if a machine breaks down in a factory, the server will suggest the optimal repair procedure based on past data. The virtual brain, which has learned the decisions managers have made in the past in response to similar breakdowns, will suggest repair procedures in real time, enabling a fast and efficient response.

[0892] Prompt Sentence Examples

[0893] "A machine has broken down in the factory. Please suggest the optimal repair procedure based on past data."

[0894] In this way, by using a virtual brain that reflects the manager's decision-making style, production efficiency within the factory can be improved.

[0895] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[0896] Step 1:

[0897] The server collects the manager's EEG data and behavioral data from the EEG sensor and behavior tracking system. These data are sent to the server through the data collection software. The input is EEG data and behavioral data, and the output is raw data stored on the server.

[0898] Step 2:

[0899] The server preprocesses the collected EEG and behavioral data, specifically by removing noise and normalizing the data. The input is raw data, and the output is preprocessed data.

[0900] Step 3:

[0901] The server uses the preprocessed data to train a generative AI model. Specifically, it uses the scikit-learn library to build a multilayer perceptron (MLPClassifier) ​​and trains it on the data. The input is the preprocessed data, and the output is the trained generative AI model.

[0902] Step 4:

[0903] The server uses a trained generative AI model to build a virtual brain, which is designed to mimic the decision-making style of a manager. The input is the trained generative AI model, and the output is the virtual brain.

[0904] Step 5:

[0905] The server inputs the current state of the production line in the factory and uses the virtual brain to make decisions for optimal adjustments. Specifically, production line data obtained from sensors is input into the virtual brain, which then outputs optimal adjustment proposals. The input is production line data, and the output is adjustment proposals.

[0906] Step 6:

[0907] When a problem occurs, the server inputs the problem data and uses the virtual brain to propose the optimal solution. Specifically, data on breakdowns and problems is input into the virtual brain, which then outputs the optimal repair procedures and solutions. The input is problem data, and the output is the solution.

[0908] Step 7:

[0909] The user adjusts the production line and performs troubleshooting based on the adjustment proposals and solutions provided by the server. Specifically, the user operates or repairs the machine according to instructions from the server. The input is the instruction from the server, and the output is the result of the adjustment or repair that was performed.

[0910] 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.

[0911] "Example 1"

[0912] In one embodiment of the present invention, the emotion engine recognizes the user's emotion by extracting the emotion from the user's facial expression, voice, text message, etc. For example, if the user shows an expression that expresses joy, the emotion engine recognizes the information as "joy."

[0913] "Example 2"

[0914] Furthermore, the emotion data obtained from the emotion engine is provided to the generative AI. The generative AI learns this emotion data and makes decisions that take into account the user's emotional state. For example, if the user expresses joy, the generative AI will take that information into account and provide decisions that are likely to make the user feel joyful.

[0915] "Example 3"

[0916] The personalized virtual brain also provides decision-making that takes into account the user's emotional data. Specifically, it adjusts the priority of decisions depending on the user's emotional state. For example, if the user is feeling stressed, it will prioritize options that will reduce stress.

[0917] The processing flow of each embodiment will be described below.

[0918] "Example 1"

[0919] Step 1: The emotion engine extracts emotions from the user's facial expressions, voice, text messages, etc.

[0920] Step 2: The emotion engine converts the extracted emotions into emotion labels such as "joy," "sadness," or "anger."

[0921] Step 3: The emotion engine provides emotion labels to the generative AI.

[0922] "Example 2"

[0923] Step 1: The generative AI receives the emotion labels provided by the emotion engine.

[0924] Step 2: The generative AI uses the received emotion labels as training data to build a decision-making model that takes the user's emotional state into account.

[0925] Step 3: The generative AI uses the constructed decision-making model to provide a decision that takes into account the user's emotional state.

[0926] "Example 3"

[0927] Step 1: The personalized virtual brain receives the decisions provided by the generative AI.

[0928] Step 2: The personalized virtual brain adjusts the received decisions according to the user's emotional state.

[0929] Step 3: The personalized virtual brain provides tailored decision-making to the user.

[0930] Example 1

[0931] Next, a description will be given of Example 1 of Form 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."

[0932] Conventional decision support systems have the problem that it is difficult to imitate the decision-making patterns of managers and are unable to provide decision support that takes user emotions into account. Furthermore, there is a lack of means to effectively utilize user feedback to improve the model. For this reason, there is a need for a system that can more accurately imitate managerial decision-making and provide personalized decision support that takes user emotions into account.

[0933] 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.

[0934] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for collecting user emotion data, means for analyzing the collected emotion data and recognizing the user's emotion, means for supporting decision-making based on the recognized emotion, and means for collecting user feedback and improving the model. This enables personalized decision-making support that accurately mimics managerial decision-making and takes user emotion into consideration.

[0935] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and models from data.

[0936] A "virtual brain" is a system that uses artificial intelligence technology to mimic the functions of the human brain and perform decision-making and information processing.

[0937] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0938] "Personalization" refers to optimizing systems and services to suit the characteristics and needs of individual users.

[0939] "Emotion data" is information about emotions extracted from the user's facial expressions, voice, text messages, etc.

[0940] "Emotion recognition" is a technology that analyzes collected emotional data and identifies the user's emotional state.

[0941] "Decision support" refers to providing necessary information and suggestions to support users when making decisions.

[0942] "Feedback" refers to the evaluation and reaction information provided by the user regarding the system's suggestions and results.

[0943] This invention is a system that uses generative AI to construct a virtual brain model that makes decisions and provides personalized decision-making support that takes into account the user's emotions. Specific embodiments of this system are described below.

[0944] The server uses generative AI to build a virtual brain model that makes decisions. Specifically, the server connects to the company's database (e.g., MySQL) and collects data on past decisions made by managers. The collected data is preprocessed using the Pandas library to impute missing values, normalize the data, and encode categorical data. A neural network model is then built using TensorFlow, and the model is trained using the preprocessed data. This model is then used to predict the manager's decision-making patterns.

[0945] The device collects the user's emotional data by capturing the user's facial expressions with the device's camera, recording audio with the microphone, and capturing text messages. This data is then sent to the server in real time via WebSocket.

[0946] The server analyzes the received emotion data. Specifically, it uses OpenCV to analyze facial expressions and IBM Watson to analyze voice and text messages. For example, if a user smiles, the server recognizes that information as "joy."

[0947] The server uses the trained AI model to assist users in decision-making. When a user asks a question about the launch of a new product, the server generates optimal suggestions based on past decision-making patterns. These suggestions are generated by the generative AI model.

[0948] As a concrete example, we use historical data on when a manager decides to launch a new product. The server provides this data to the AI, which then learns the patterns. When a user asks a question about launching a new product, the AI ​​suggests the best decision based on past patterns.

[0949] Example prompt sentence:

[0950] "Please provide optimal recommendations for the next new product launch based on past new product launch decision-making data."

[0951] The user receives suggestions from the server and provides feedback on the suggestions. Specifically, the user indicates their satisfaction with the suggestions through facial expressions and voice. The device collects this feedback and sends it to the server. The server uses this feedback to further improve the AI ​​model.

[0952] Example prompt sentence:

[0953] "Analyze how users feel when they receive a new product suggestion and tailor your next suggestion accordingly."

[0954] In this way, personalized decision support that accurately mimics the decision-making of managers and takes into account the user's emotions becomes possible.

[0955] The flow of the identification process in the first embodiment will be described with reference to FIG.

[0956] Step 1: Data collection

[0957] The server connects to the company's database and collects data on past decisions made by management. Specifically, the server executes the query "SELECT FROM decision_data WHERE date > '2020-01-01'" on the MySQL database to retrieve decision data. This data includes information such as the type of decision, date and time, and outcome. The input is the database query, and the output is the retrieved decision data.

[0958] Step 2: Data Preprocessing

[0959] The server preprocesses the collected data. Specifically, it uses the Pandas library to impute missing values ​​and Scikit-learn to normalize the data. It also performs one-hot encoding on categorical data. The input is the collected decision data, and the output is the preprocessed data.

[0960] Step 3: Model training

[0961] The server uses the preprocessed data to train an AI model. Specifically, it uses TensorFlow to build a neural network model and trains the model using the preprocessed data as input. The input is the preprocessed data, and the output is a trained AI model.

[0962] Step 4: Collect emotional data

[0963] The device collects the user's emotional data. Specifically, it captures the user's facial expressions with the device's camera, records audio with the microphone, and obtains text messages. These data are sent to the server in real time via WebSocket. The input is the user's facial expressions, audio, and text messages, and the output is the collected emotional data.

[0964] Step 5: Emotion Recognition

[0965] The server analyzes the received emotion data. Specifically, it uses OpenCV to analyze facial expressions and IBM Watson to analyze voice and text messages. For example, if a user smiles, the server recognizes that information as "joy." The input is the collected emotion data, and the output is the recognized emotion information.

[0966] Step 6: Decision support

[0967] The server uses the trained AI model to assist the user in decision-making. Specifically, when the user inputs a prompt such as "Please give me the best proposal for the next new product launch," the server generates the best proposal based on past decision-making patterns. The inputs are the user's prompt and the trained AI model, and the output is the generated proposal.

[0968] Step 7: User Feedback

[0969] The user receives suggestions from the server and provides feedback on the suggestions. Specifically, the user indicates whether they are satisfied with the suggestions using facial expressions and voice. The device collects this feedback and sends it to the server. The server uses this feedback to further improve the AI ​​model. The input is the user's feedback, and the output is an improved AI model.

[0970] (Application example 1)

[0971] Next, a description will be given of Application Example 1 of Form 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."

[0972] Conventional factory management systems have difficulty reflecting the decision-making patterns of managers in real time, and lack support that takes into account the emotions of factory staff, resulting in problems that lead to reduced production efficiency and staff satisfaction. For this reason, there is a need for a system that can learn the decision-making patterns of managers, optimize factory production lines, and recognize staff emotions to provide appropriate feedback.

[0973] 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.

[0974] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for providing feedback based on the user's emotions using an emotion engine that recognizes the user's emotions, and means for learning the decision-making patterns of managers and optimizing factory production lines.This makes it possible to reflect the manager's decision-making patterns in real time and provide support that takes the emotions of factory staff into consideration.

[0975] "Generative AI" is an algorithm that uses artificial intelligence techniques to generate new information and patterns from data.

[0976] A "virtual brain" is a model that uses artificial intelligence technology to mimic the decision-making process of the human brain.

[0977] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[0978] "Personalization" refers to providing services and functions that are customized to suit the characteristics and needs of individual users.

[0979] An "emotion engine" is a technology for recognizing and analyzing emotions from a user's facial expressions, voice, text messages, etc.

[0980] "Managerial decision-making patterns" are decision-making tendencies and rules extracted based on data on decisions made by managers in the past.

[0981] "Production line optimization" refers to making adjustments and improvements to streamline a factory's production processes, reducing costs and improving productivity.

[0982] "Feedback" refers to evaluations and advice provided regarding the system or user's behavior.

[0983] As an embodiment of this invention, we will explain a system for smart factory management robots as an example. This system uses generative AI to build a virtual brain model for decision-making and optimize the factory production line. It also uses an emotion engine to recognize the emotions of factory staff and provide appropriate feedback.

[0984] The server uses generative AI to build a virtual brain model that makes decisions. This model learns from the manager's past decision-making data and extracts patterns, making it possible to imitate the manager's decision-making.

[0985] Next, the server speeds up the PDCA cycle to improve the accuracy of the model. This cycle consists of four steps: Plan, Do, Check, and Act, which are repeated to continuously improve the accuracy of the model.

[0986] Furthermore, the server learns and applies data obtained from the individual's brain to build a personalized virtual brain, enabling decision-making support tailored to the characteristics and needs of each individual manager.

[0987] The emotion engine recognizes emotions from users' facial expressions, voice, text messages, etc. For example, a camera can be used to capture the facial expressions of factory staff, and the emotion engine can analyze the expressions to extract emotions. This allows the system to provide feedback based on the staff's emotions.

[0988] Generative AI models that learn the decision-making patterns of managers can optimize factory production lines. For example, if a production line stops, the model can quickly resolve the issue by referring to past decision-making patterns. It can also obtain production data in real time and create optimal production plans.

[0989] For example, if a production line in a factory stops, the robot will refer to the manager's past decision-making patterns to quickly resolve the problem. Also, if a factory staff member is feeling stressed, the robot will use its emotion engine to recognize that emotion and provide appropriate support.

[0990] Examples of prompts to input to a generative AI model include:

[0991] "If the production line stops in a factory, how can you quickly solve the problem by referring to the decision-making patterns of past managers?"

[0992] In this way, smart factory management robots can improve factory productivity and efficiency while taking into account management decision-making patterns and staff emotions.

[0993] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[0994] Step 1:

[0995] The server acquires images from cameras installed in the factory and captures the facial expressions of factory staff in real time. The input is the camera image, and the output is the captured image data.

[0996] Step 2:

[0997] The server inputs the captured image data into the emotion engine and recognizes emotions from facial expressions. The emotion engine analyzes the image data using a facial expression recognition model and extracts the user's emotion. The input is image data, and the output is recognized emotion data.

[0998] Step 3:

[0999] The server provides appropriate feedback to the user based on the recognized emotion data. For example, if the user feels stressed, the server provides advice on how to relax. The input is emotion data, and the output is a feedback message.

[1000] Step 4:

[1001] The server inputs the manager's past decision-making data into the generative AI model to learn the decision-making patterns. The generative AI model analyzes the past data and extracts the manager's decision-making patterns. The input is the decision-making data, and the output is the learned decision-making patterns.

[1002] Step 5:

[1003] The server inputs production data acquired in real time into the generative AI model and creates an optimal production plan. The generative AI model optimizes the production line based on the learned decision-making patterns. The input is production data, and the output is an optimized production plan.

[1004] Step 6:

[1005] The server sends the optimized production plan to the factory robots to execute it. The factory robots adjust the production line based on the received production plan to ensure efficient production. The input is the optimized production plan, and the output is the adjusted production line.

[1006] Step 7:

[1007] The server monitors the operation of the factory robots and re-optimizes them as necessary, thereby always maintaining optimal production conditions. The input is the robot's operation data, and the output is a re-optimized production plan.

[1008] In this way, the server can use generative AI models and emotion engines to improve factory productivity and staff satisfaction.

[1009] Example 2

[1010] Next, a description will be given of Example 2 of Form 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."

[1011] Conventional decision-making support systems do not take into account the emotional state of the user when making decisions, making it difficult to respond quickly to changes in the business environment. In addition, there was a lack of means to improve the accuracy of the model by quickly implementing the PDCA cycle.

[1012] The specific processing by the specific processing unit 290 of the data processing device 12 in Example 2 is realized by the following means. In this invention, the server includes: means for constructing a virtual brain model that makes decisions using a generative AI; means for speeding up the PDCA cycle and improving the accuracy of the model; means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain; means for receiving user input and generating emotion data through an emotion engine; means for transmitting the generated emotion data to the server; means for the server to obtain the latest management data and feed it back to the AI ​​model; means for making decisions based on the user's emotion data and the latest management data using the generative AI model; and means for displaying the decision-making results to the user. This enables fast and accurate decision-making that takes the user's emotional state into consideration.

[1013] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and decisions from data.

[1014] A "virtual brain" is a system that uses artificial intelligence technology to mimic the functions of the human brain and make decisions.

[1015] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[1016] "Personalization" refers to optimizing systems and services to suit the characteristics and needs of individual users.

[1017] An "emotion engine" is a system that uses natural language processing technology to analyze and generate emotional data from user input.

[1018] "Emotion data" is data that expresses the user's emotional state as numerical values ​​or categories.

[1019] A "server" is a computer system that processes data and provides services over a network.

[1020] "Management data" refers to data that includes information related to management, such as corporate performance and market trends.

[1021] "Decision-making results" are the optimal actions and strategies proposed by the generative AI model based on user input, emotional data, and management data.

[1022] MODE FOR CARRYING OUT THE INVENTION

[1023] This invention is a system that uses generative AI to build a virtual brain model that makes decisions, improves the accuracy of the model by quickly implementing the PDCA cycle, and provides decision-making that takes into account the emotional state of the user. Specific embodiments of this system are described below.

[1024] Hardware and software used

[1025] Server: A computer system used to process data and train AI models. For example, an AWS EC2 instance.

[1026] Terminal: A device used by a user to input information, such as a PC or smartphone.

[1027] Emotion engine: Software for analyzing and generating emotion data from user input, for example using an NLP library (such as spaCy).

[1028] Generative AI models: Algorithms for generating new information and decisions from data, for example using TensorFlow or PyTorch.

[1029] Explanation of program processing

[1030] 1. User input:

[1031] A user inputs a question or request into the system through a terminal interface, for example, "What are the sales forecasts for the next quarter?"

[1032] 2. Emotion data generation:

[1033] The device sends the user's input to the emotion engine, which analyzes the user's emotional state. The emotion engine uses NLP techniques to extract emotion data from the user's input. For example, if the user's input indicates a positive emotion, the emotion engine generates emotion data for "joy."

[1034] 3. Sending Emotional Data:

[1035] The terminal transmits the generated emotion data to the server. For example, the emotion data of "joy" and the user's input are transmitted to the server.

[1036] 4. Management data acquisition and feedback:

[1037] The server retrieves the latest management data from the database. For example, it retrieves the latest sales data and market trend data. The retrieved data is fed back to the AI ​​model. The server preprocesses the data using Python and retrains the model using machine learning libraries such as TensorFlow and PyTorch.

[1038] 5. Decision making:

[1039] The server uses a generative AI model to make optimal decisions in response to user requests. The generative AI model makes decisions by taking into account the user's emotional data and the latest management data. For example, if a user inputs, "What is the sales forecast for next quarter?", the generative AI model will provide a positive sales forecast based on the emotional data of "happiness" and the latest sales data.

[1040] 6. Sending and displaying decision results:

[1041] The server sends the decision-making results output by the generative AI model to the device. For example, it sends the results of a sales forecast to the device. The device then displays the decision-making results received from the server to the user. For example, it displays, "Sales forecast for the next quarter are expected to increase by 10% compared to the previous year."

[1042] Examples of concrete examples and prompts

[1043] Examples:

[1044] User types, "What is your forecast for next quarter?"

[1045] The emotion engine generates emotion data for "joy."

[1046] The server retrieves the latest sales data and feeds it back into the generative AI model.

[1047] The generative AI model generates a positive sales forecast and sends it to the device.

[1048] The device displays, "Sales forecast for next quarter are expected to increase 10% year over year."

[1049] Example prompt sentence:

[1050] "What are your sales forecasts for the next quarter?"

[1051] "Please tell us how you feel about this week's performance."

[1052] "Please suggest a marketing strategy for the new product."

[1053] In this way, users receive decisions that take their emotional state into account and can respond quickly to changes in the business environment.

[1054] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1055] Step 1:

[1056] The user types into the terminal.

[1057] A user inputs a question or request into the system through the terminal interface, for example, "What is the sales forecast for the next quarter?" The input text data is sent to the terminal.

[1058] Step 2:

[1059] The device generates emotion data through an emotion engine.

[1060] The device sends the user's input text data to the emotion engine, which then uses NLP technology to extract emotion data from the user's input. For example, if the user's input indicates a positive emotion, the emotion engine generates emotion data for "joy." The input is the user's text data, and the output is emotion data.

[1061] Step 3:

[1062] The device transmits the emotion data to the server.

[1063] The device sends the generated emotion data and the user's input in JSON format to the server. The input is the emotion data and the user's text data, and the output is the data to be sent to the server.

[1064] Step 4:

[1065] The server obtains the latest management data and provides feedback to the AI ​​model.

[1066] The server retrieves the latest management data from the database using SQL queries. For example, it retrieves the latest sales data or market trend data. The retrieved data is preprocessed using a Python script and fed back to the AI ​​model. The input is management data from the database, and the output is the preprocessed data.

[1067] Step 5:

[1068] The server makes decisions using generative AI models.

[1069] The server inputs emotional data and the latest business data into the generative AI model to make optimal decisions based on the user's request. For example, if a user inputs "Please tell me the sales forecast for next quarter," the generative AI model generates a positive sales forecast based on the emotional data of "happiness" and the latest sales data. The inputs are emotional data and business data, and the output is the decision-making result.

[1070] Step 6:

[1071] The server transmits the decision result to the terminal.

[1072] The server sends the decision-making results output by the generative AI model to the terminal in JSON format. The input is the decision-making results, and the output is the data sent to the terminal.

[1073] Step 7:

[1074] The terminal displays the results to the user.

[1075] The terminal displays the decision-making results received from the server to the user. For example, it displays "Sales forecast for the next quarter are expected to increase by 10% compared to the previous year." The input is the decision-making results from the server, and the output is the data displayed to the user.

[1076] (Application example 2)

[1077] Next, a description will be given of Application Example 2 of Form 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."

[1078] Conventional decision-making systems make decisions without considering the user's emotional state, which leads to problems such as reduced user satisfaction and engagement. Furthermore, it is difficult to recognize emotions in real time and propose optimal products and services based on those emotions, making it difficult to improve the user experience.

[1079] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 2 is realized by the following means.

[1080] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for providing emotion data obtained from an emotion engine to the generative AI and making decisions taking into account the user's emotional state, means for suggesting optimal products and services based on the user's emotional state, means for capturing the user's face using a camera and performing emotion recognition, means for transmitting emotion data to the end point of the generative AI model and obtaining recommended products and services, and means for displaying the recommended products and services in real time. This enables personalized decision-making that takes into account the user's emotional state, improving the user experience.

[1081] "Generative AI" is a system that uses artificial intelligence technology to generate new information and decisions from data.

[1082] A "virtual brain" is a model that uses artificial intelligence technology to mimic the functions of the human brain and make decisions.

[1083] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[1084] "Personalization" means optimizing services and information to suit the characteristics and preferences of individual users.

[1085] An "emotion engine" is a system for detecting and analyzing a user's emotional state.

[1086] "Emotion data" is data that indicates the user's emotional state and is acquired by the emotion engine.

[1087] An "endpoint" is a URL or URI that allows access to a particular service or function on a network.

[1088] "Recommended products and services" are products and services suggested based on the user's emotional state and preferences.

[1089] "Real-time" means that data acquisition and processing occurs immediately.

[1090] A "camera" is a device for taking images or videos.

[1091] "Emotion recognition" is a technology that determines a user's emotional state from facial expressions, voice, etc.

[1092] A system for carrying out this invention has the following configuration: A server includes means for constructing a virtual brain model that makes decisions using a generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for providing emotion data obtained from an emotion engine to the generative AI and making decisions taking into account the user's emotional state, means for suggesting optimal products and services based on the user's emotional state, means for capturing the user's face using a camera and recognizing emotions, means for transmitting emotion data to an endpoint of the generative AI model and obtaining recommended products and services, and means for displaying the recommended products and services in real time.

[1093] Hardware and software used

[1094] Hardware

[1095] Smart glasses or smartphone: A device worn or carried by the user that contains a camera.

[1096] Camera: A device for capturing the user's face and performing emotion recognition.

[1097] software

[1098] OpenCV: An image processing library used for camera capture and face detection.

[1099] Keras: A deep learning library used to build and run emotion recognition models.

[1100] Requests: An HTTP request library used to send emotion data to the generative AI model endpoint.

[1101] Data processing and calculation

[1102] 1. Camera image capture: Use the camera on the smart glasses or smartphone to capture the user's face in real time. OpenCV is used to acquire the image and perform face detection.

[1103] 2. Emotion Recognition: The captured facial image is input into an emotion recognition model built with Keras to determine the user's emotional state. Emotion data is obtained from the user's facial expressions.

[1104] 3. Sending Emotion Data: The determined emotion data is sent to the generative AI model endpoint using the Requests library. The generative AI then uses this emotion data to make decisions that take into account the user's emotional state.

[1105] 4. Acquisition and display of recommended products: The recommended products and services returned by the generation AI are displayed on the user's device in real time, allowing the user to receive suggestions for the most suitable products and services on the spot.

[1106] Specific examples

[1107] When a user wears the smart glasses and walks around the store, the camera captures the user's face and performs emotion recognition. For example, if the user shows joy, the generative AI will display recommendations such as, "This dress from our new collection is perfect for you!"

[1108] Prompt Sentence Examples

[1109] Take user emotion data as input and generate product or service recommendations when the user is expressing happiness.

[1110] Emotion: Joy

[1111] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1112] Step 1:

[1113] The device captures the user's face in real time using the camera on the smart glasses or smartphone. The input is the camera image, and the output is the captured face image. Specifically, OpenCV is used to acquire the image and perform face detection.

[1114] Step 2:

[1115] The device inputs the captured facial image into an emotion recognition model built with Keras to determine the user's emotional state. The input is a facial image and the output is emotion data. Specifically, the facial image is preprocessed and input into the emotion recognition model to determine the emotion.

[1116] Step 3:

[1117] The device sends the determined emotion data to the generative AI model's endpoint using the Requests library. The input is emotion data, and the output is recommended products and services returned by the generative AI. Specifically, it generates an HTTP request and sends the emotion data.

[1118] Step 4:

[1119] The server processes the emotional data using a generative AI model and generates recommended products and services that take the user's emotional state into account. The input is emotional data, and the output is recommended products and services. Specifically, the server generates prompt sentences based on the emotional data, which are then input into the generative AI model to obtain recommendations.

[1120] Step 5:

[1121] The terminal displays the recommended products and services returned by the generation AI on the user's device in real time. The input is the recommended products and services, and the output is the information displayed to the user's visual sense. Specifically, it updates the UI to display the recommended information on the screen.

[1122] Example 3

[1123] Next, a description will be given of Example 3 of Form Example 3. 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."

[1124] Conventional decision-making support systems have difficulty fully considering the decision-making style and emotional state of individual managers, and can only make proposals based on general decision-making models. Furthermore, they are unable to reflect the user's emotional state in real time, making it difficult to provide appropriate decision-making support that responds to stress and emotional fluctuations.

[1125] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1126] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from individual brains to construct a personalized virtual brain, means for collecting user emotional data in real time, and means for adjusting decision-making priorities based on the collected emotional data. This enables appropriate decision-making support that reflects the decision-making style of each manager and corresponds to the emotional state of the user.

[1127] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and models from data.

[1128] A "virtual brain" is a system that uses artificial intelligence technology to mimic the decision-making process of the human brain.

[1129] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[1130] "Personalization" refers to optimizing systems and services to suit the characteristics and needs of individual users.

[1131] "Electroencephalogram data" is data obtained by measuring the electrical activity of the brain, and is information that indicates the state and activity of the brain.

[1132] "Behavioral data" is data that records the user's behavior and activities, and is information that indicates the user's behavioral patterns during specific time periods and situations.

[1133] "Emotion data" is data that indicates the user's emotional state, and is information obtained from facial expressions, voice, brain waves, and the like.

[1134] "Decision-making priority" is a criterion for determining which of multiple options or tasks should be executed first.

[1135] "Real-time" means that data is collected and processed immediately, and the results are reflected without delay.

[1136] This invention is a system that uses generative AI to build a virtual brain model for decision-making, and provides a personalized virtual brain by learning and applying data obtained from an individual's brain. It also includes a function to collect user emotional data in real time and adjust decision-making priorities based on the collected emotional data.

[1137] Data collection

[1138] The server collects the manager's brainwave and behavioral data. Specifically, it uses an EEG measurement device (e.g., Emotiv Epoc+) and behavior tracking software (e.g., RescueTime) to obtain the manager's brain activity patterns and daily behavioral data in real time.

[1139] Data Preprocessing

[1140] The server preprocesses the collected data. For example, it removes noise and filters EEG data, and categorizes and tags behavioral data. This is done using Python libraries (e.g., NumPy, Pandas). For example, it analyzes the frequency of application usage during a specific time period.

[1141] Training an AI model

[1142] The server uses the preprocessed data to train the AI ​​model. Specifically, it uses TensorFlow and PyTorch to train the model on data on how managers make decisions in specific business situations. For example, the model can reflect the decision-making patterns of managers based on data on the success and failure of past projects.

[1143] Building a Virtual Brain

[1144] The server builds a personalized virtual brain based on a trained AI model. This virtual brain reflects the individual decision-making style of each manager. Specifically, it replicates the manager's past decision-making patterns and makes the optimal decision in similar situations.

[1145] Collecting Emotional Data

[1146] The device collects the user's emotional data in real time. Specifically, it uses emotion recognition software (e.g., Affectiva) to analyze the user's emotional state from their facial expressions and voice. For example, it can detect whether the user is feeling stressed in real time.

[1147] Applying Emotion Data

[1148] The server adjusts decision-making priorities based on the collected emotional data. For example, if the user is feeling stressed, the virtual brain will prioritize options to reduce stress, such as reprioritizing tasks or suggesting taking a break.

[1149] Specific examples

[1150] Example 1: Decision making in a business situation

[1151] Suppose a user is considering starting a new project. The virtual brain uses past data to learn how managers have made decisions in similar situations and suggests optimal decisions. For example, it suggests strategies to minimize risk based on data from past successful projects.

[1152] Example 2: Applying Emotion Data

[1153] If the user is feeling stressed, the virtual brain suggests options to reduce stress (e.g., reprioritizing tasks, taking a break). For example, if the user is feeling stressed before an important meeting, the virtual brain suggests taking a short break to relax.

[1154] Prompt Sentence Examples

[1155] "Please build an AI model that uses the brainwave and behavioral data of business managers to learn how to make decisions in specific business situations. Also, please add a function that adjusts the priority of decision-making by taking into account the user's emotional data."

[1156] In this way, a personalized virtual brain that reflects the manager's decision-making style is constructed, and a system that provides decision-making according to the user's emotional state is completed. The flow of the specific processing in Example 3 will be described with reference to FIG.

[1157] Step 1:

[1158] The server collects the manager's EEG data and behavioral data. Specifically, it acquires EEG data using an EEG measurement device (e.g., Emotiv Epoc+) and collects behavioral data using behavioral tracking software (e.g., RescueTime). The input is data from the EEG measurement device and behavioral tracking software, and the output is the collected raw data.

[1159] Step 2:

[1160] The server preprocesses the collected data. It performs noise removal and filtering on EEG data, and categorizes and tags behavioral data. Specifically, it cleans the data using Python libraries (e.g., NumPy, Pandas). The input is the collected raw data, and the output is the preprocessed, clean data.

[1161] Step 3:

[1162] The server uses the preprocessed data to train the AI ​​model. Specifically, it uses TensorFlow and PyTorch to train data on the decisions managers made in specific business situations. The input is the preprocessed data, and the output is the trained AI model.

[1163] Step 4:

[1164] The server builds a personalized virtual brain based on a trained AI model. Specifically, it reproduces the decision-making patterns of managers in the past, and the virtual brain makes optimal decisions in similar situations. The input is the trained AI model, and the output is the personalized virtual brain.

[1165] Step 5:

[1166] The device collects the user's emotional data in real time. Specifically, it uses emotion recognition software (e.g., Affectiva) to analyze the user's emotional state from their facial expressions and voice. The input is the user's facial and voice data, and the output is the analyzed emotional data.

[1167] Step 6:

[1168] The server adjusts decision-making priorities based on the collected emotional data. Specifically, if the user is feeling stressed, the virtual brain will prioritize options that will reduce stress. The input is the analyzed emotional data, and the output is the adjusted decision-making priorities.

[1169] (Application example 3)

[1170] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1171] In modern factories, improving production line efficiency and troubleshooting are important issues. However, solving these issues requires an advanced system that reflects the manager's decision-making style. It is also necessary to adjust the workload taking into account the emotional state of the operator. Conventional systems have difficulty meeting these requirements, making it difficult to achieve efficient production management.

[1172] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1173] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from individual brains to construct a personalized virtual brain, means for optimizing production lines in factories, means for providing optimal solutions to problems that occur in factories, and means for monitoring the emotional state of operators in real time and adjusting their workloads.This enables efficient production management that reflects the decision-making style of managers and adjustment of workloads that take into account the emotional state of operators.

[1174] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and models from data.

[1175] A "virtual brain" is an artificial intelligence model built based on an individual's brainwave and behavioral data, which mimics a specific decision-making style.

[1176] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[1177] "Personalization" means customizing something to meet specific needs and requirements based on individual characteristics and data.

[1178] "Production line optimization" means maximizing production efficiency by efficiently arranging production processes within a factory and optimizing work flow.

[1179] "Troubleshooting" is a technique for quickly identifying and resolving problems or failures that occur in systems or processes.

[1180] "Emotional state" refers to an individual's psychological state or mood, and includes factors such as stress and satisfaction.

[1181] "Workload" refers to the amount of effort or time required for a particular task or job.

[1182] To implement the present invention, the following system configuration and program are required.

[1183] System Configuration

[1184] 1. Hardware:

[1185] Factory robots: Robots used to carry out production line tasks.

[1186] Brainwave sensor: A device for collecting brainwave data from business managers.

[1187] Emotion Monitoring Device: A device for monitoring the emotional state of the operator in real time.

[1188] 2. Software:

[1189] Python: A programming language used to implement the program.

[1190] scikit-learn: A library used to build machine learning models.

[1191] NumPy: A library used for data preprocessing and numerical computation.

[1192] Program processing explanation

[1193] The server first collects the manager's brainwave and behavioral data, and then uses this data to build a virtual brain. The virtual brain is a model that uses generative AI to mimic the manager's decision-making style. This model is used to optimize production lines and troubleshoot problems within factories.

[1194] Specifically, the server performs the following process.

[1195] 1. Data Collection:

[1196] Brainwave data of managers is collected using an EEG sensor.

[1197] Emotional data from operators is collected using an emotion monitoring device.

[1198] 2. Data Preprocessing:

[1199] Normalize the collected data using NumPy.

[1200] 3. Train the model:

[1201] We build a virtual brain using scikit-learn's MLPRegressor.

[1202] The system learns the brainwave and behavioral data of managers.

[1203] 4. Prediction and optimization:

[1204] Input data for a new production line and the virtual brain predicts the optimal layout.

[1205] We provide optimal solutions to problems that occur within the factory.

[1206] 5. Considering emotional data:

[1207] Monitor operators' emotional state in real time and adjust their workload if stress levels increase.

[1208] Specific examples

[1209] For example, if a manager learns from data on past decisions made to resolve bottlenecks on a production line and inputs the current production line data, the virtual brain will suggest optimal layouts. It will also make decisions to reduce the workload of operators if they are feeling stressed.

[1210] Prompt Sentence Examples

[1211] "Using the brainwave and behavioral data of managers, please create a virtual brain that can optimize production lines in factories. When you input data on a new production line, please create a model that will suggest the optimal layout."

[1212] In this way, by applying a virtual brain that reflects the decision-making style of managers to factory robots, we can expect to improve production efficiency and speed up troubleshooting.

[1213] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1214] Step 1:

[1215] The server collects the manager's brainwave data using an EEG sensor. Specifically, the EEG sensor is worn on the manager's head and captures brainwave data in real time. This data is sent to the server and stored in a database. The input is raw data from the EEG sensor, and the output is stored brainwave data.

[1216] Step 2:

[1217] The server collects the operator's emotional data using an emotion monitoring device. Specifically, the emotion monitoring device is worn on the operator's body and captures physiological data such as heart rate and galvanic skin response. This data is sent to the server and stored in a database. The input is the raw data from the emotion monitoring device, and the output is the stored emotional data.

[1218] Step 3:

[1219] The server preprocesses the collected EEG and emotion data. Specifically, it uses NumPy to standardize the data, fill in missing values, and remove noise. The input is the stored EEG and emotion data, and the output is the preprocessed data.

[1220] Step 4:

[1221] The server uses the preprocessed data to construct a virtual brain. Specifically, it trains a generative AI model using scikit-learn's MLPRegressor. The input is preprocessed EEG data and behavioral data, and the output is a trained virtual brain model.

[1222] Step 5:

[1223] The server receives data on the new production line as input and uses the virtual brain model to predict the optimal layout. Specifically, data on each station on the production line is input, and the virtual brain model calculates the optimal layout. The input is the data on the new production line, and the output is the optimized production line layout.

[1224] Step 6:

[1225] The server provides optimal solutions to problems that occur within the factory. Specifically, data on the type of problem and its location is input, and the virtual brain model proposes optimal solutions based on past data. The input is data on the problem, and the output is the proposed solution.

[1226] Step 7:

[1227] The server monitors the emotional state of the operator in real time and adjusts the workload. Specifically, it analyzes data from the emotion monitoring device in real time and issues instructions to reduce the workload if stress levels are high. The input is real-time emotional data, and the output is the adjusted workload.

[1228] 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.

[1229] 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> ) 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.

[1230] Another example of generative AI is Gemini (internet search engine). <url: https: gemini.google.com ?hl="ja">) are listed.

[1231] 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.

[1232] [Third embodiment]

[1233] FIG. 5 shows an example of the configuration of a data processing system 310 according to the third embodiment.

[1234] 5, the data processing system 310 includes the data processing device 12 and a headset type terminal 314. An example of the data processing device 12 is a server.

[1235] 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).

[1236] 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.

[1237] 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.

[1238] 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).

[1239] 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.

[1240] 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.

[1241] 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.

[1242] 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.

[1243] 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.

[1244] Next, the specific processing by the specific processing unit 290 of the data processing device 12 will be described.

[1245] "Example 1"

[1246] In one embodiment of the present invention, the means for constructing a virtual brain model using generative AI includes an AI algorithm for learning the decision-making patterns of managers. Specifically, data on past decisions made by managers is used as learning data, and the AI ​​learns those patterns.

[1247] "Example 2"

[1248] Furthermore, to speed up the PDCA cycle and improve the accuracy of the model, AI periodically updates its learning data and learns new decision-making patterns, enabling it to make decisions that respond to changes in the business environment.

[1249] "Example 3"

[1250] Furthermore, as a means of learning and applying data obtained from an individual's brain, a personalized virtual brain is constructed by acquiring the manager's brainwave data and behavioral data and using this as learning data for the AI. This makes it possible to construct a virtual brain that reflects the individual decision-making style of each manager. As a specific example, the AI ​​can learn data on how a manager made decisions in specific business situations, and the virtual brain can make decisions based on that data.

[1251] The processing flow of each embodiment will be described below.

[1252] "Example 1"

[1253] Step 1: Collect data on management's past decisions. This includes information on the results of decisions made by management in the past and the circumstances at the time.

[1254] Step 2: The collected data is input into the generative AI, which then learns the manager's decision-making patterns. This learning process builds a model for the AI ​​to imitate the manager's decision-making style.

[1255] Step 3: Use the constructed model to predict managerial decisions for new business situations.

[1256] "Example 2"

[1257] Step 1: The AI ​​periodically updates its learning data by adding newly acquired managerial decision-making data to the learning data.

[1258] Step 2: Using the updated learning data, the AI ​​learns new decision-making patterns. Step 3: Using the new decision-making patterns it has learned, it predicts decisions that respond to changes in the business environment.

[1259] "Example 3"

[1260] Step 1: Obtain the manager's brainwave data, behavioral data, etc. These data are used to reflect the manager's individual decision-making style.

[1261] Step 2: The acquired data is used as training data for the AI, which then learns the decision-making styles of individual managers.

[1262] Step 3: Build a personalized virtual brain using the learned decision-making styles of individual managers.

[1263] Example 1

[1264] Next, a description will be given of Example 1 of Form 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."

[1265] Conventional decision support systems lacked personalization and were unable to fully reflect the individual decision-making patterns of managers. Furthermore, data collection, preprocessing, model training, and evaluation were often performed manually, resulting in inefficiency and low accuracy. Furthermore, generating and inputting prompts was cumbersome, and the output of prediction results was not intuitive, making it difficult for managers to make quick decisions.

[1266] 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.

[1267] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for collecting past decision-making data and storing it in a database, means for preprocessing the collected data, means for training the AI ​​model using the preprocessed data, means for evaluating the trained model, means for generating prompt sentences and inputting them into the AI ​​model, and means for outputting prediction results.This allows for the construction of a personalized virtual brain that reflects the individual decision-making patterns of a manager, enabling efficient and highly accurate decision-making support.

[1268] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new data and models.

[1269] A "virtual brain" is a system that uses artificial intelligence technology to mimic the functions of the human brain and make decisions.

[1270] "PDCA" refers to the cycle of planning (Plan), doing (Do), checking (Check), and improving (Act), and is a method for making continuous improvements.

[1271] "Personalization" refers to optimizing systems and services based on the characteristics and behavioral patterns of individual users.

[1272] A "database" is a system for efficiently storing, managing, and retrieving data.

[1273] "Preprocessing" is the process of cleaning and transforming data before data analysis or machine learning.

[1274] "Training" is the process by which a machine learning model learns patterns from data.

[1275] "Evaluation" is the process of measuring the performance of a trained model using validation data.

[1276] A "prompt" is an instruction given to a generative AI model, containing instructions for performing a specific task.

[1277] "Prediction result" refers to the result that a generative AI model outputs based on a prompt, including predictions of future events or data.

[1278] This invention is a system for constructing a virtual brain model that uses generative AI to support managerial decision-making. Specific embodiments of this system are described below.

[1279] First, the server collects data on past decisions made by management, including sales data, market trends, economic indicators, competitor trends, etc. The server stores this data in a database management system (e.g., MySQL).

[1280] The server then preprocesses the collected data, using the Python Pandas library to impute missing values, normalize the data, and detect and remove outliers, thereby improving the quality of the data.

[1281] The server then uses the preprocessed data to train an AI model, using a machine learning framework (e.g., TensorFlow) to learn the decision-making patterns of managers. Training involves using past decision-making data as input so the model can recognize patterns.

[1282] After training is complete, the server evaluates the trained model using a validation dataset to measure the model's precision and recall, allowing the server to verify the model's performance and make adjustments if necessary.

[1283] The user inputs a prompt to the generative AI model. For example, the user inputs a prompt such as, "Please predict sales for the next quarter based on sales data from the past five years and market trends." The input prompt is sent to the server.

[1284] The server runs the AI ​​model based on the input prompt and generates a prediction result. The prediction result is displayed to the user, who can use it as a reference to develop future management strategies.

[1285] For example, data on a manager's past decisions may include the following:

[1286] Sales Data

[1287] market trends

[1288] Economic indicators

[1289] Competitor trends

[1290] The server collects this data and inputs it into the AI ​​model. For example, the following prompt sentence is input into the generative AI model:

[1291] "Please provide a sales forecast for the next quarter based on sales data from the past five years and market trends."

[1292] Based on this prompt, the AI ​​model analyzes past data and provides a sales forecast for the next quarter, which users can use to plan future business strategies.

[1293] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1294] Step 1: Data collection

[1295] The server collects data on past decisions made by management. Specifically, it obtains data such as sales data, market trends, economic indicators, and competitor trends using APIs and web scraping. This data is saved as a CSV file and imported into a database management system (e.g., MySQL). The input is raw data from various data sources, and the output is structured data stored in the database.

[1296] Step 2: Data Preprocessing

[1297] The server preprocesses the collected data. Specifically, it uses the Python Pandas library to impute missing values, normalize the data, and detect and remove outliers. For example, missing values ​​are imputed with the mean value, and data normalization uses min-max scaling. The input is the raw data stored in the database, and the output is the preprocessed, clean data.

[1298] Step 3: Train the model

[1299] The server trains an AI model using the preprocessed data. It uses a machine learning framework (e.g., TensorFlow) to build a neural network model. It processes the training data in batches and updates the model weights every epoch. The input is the preprocessed data, and the output is the trained AI model.

[1300] Step 4: Evaluate the model

[1301] The server evaluates the trained model. For evaluation, a validation dataset is used to measure the model's precision and recall. Evaluation metrics include mean absolute error (MAE) and coefficient of determination (R²). The input is the trained AI model and the validation dataset, and the output is the evaluation result.

[1302] Step 5: Generate and enter a prompt

[1303] The user inputs a prompt to the generative AI model. For example, the prompt might be, "Please predict sales for the next quarter based on sales data from the past five years and market trends." The input is the prompt entered by the user, and the output is the prompt sent to the server.

[1304] Step 6: Output prediction results

[1305] The server runs the AI ​​model based on the input prompt and generates a prediction result. The prediction result is displayed in graph or table format on a web interface. The input is the prompt and the trained AI model, and the output is the prediction result. Users can use this prediction result as a reference to develop future management strategies.

[1306] (Application example 1)

[1307] Next, a description will be given of Application Example 1 of Form 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."

[1308] With conventional production management systems, it was difficult to mimic management decisions in real time and perform optimal production management. Furthermore, there was a lack of means to effectively utilize management's past decision-making data, making it impossible to achieve efficient production management. Furthermore, systems for collecting and analyzing data in real time were complex, resulting in significant costs for implementation and operation.

[1309] 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.

[1310] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from individual brains to construct a personalized virtual brain, means for collecting and storing data from production lines in factories in real time, means for preprocessing the collected data and training the AI ​​model, and means for analyzing production line data in real time using the trained AI model to make optimal decisions. This enables efficient production management by imitating management decision-making in real time.

[1311] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new data and models.

[1312] A "virtual brain" is a model that uses artificial intelligence technology to mimic the decision-making process of the human brain.

[1313] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[1314] "Personalization" refers to the individual optimization of systems and services based on the characteristics and behavior of individual users.

[1315] A "production line" refers to a series of machines or work stations arranged to efficiently produce a product.

[1316] "Real-time data collection" refers to obtaining data immediately from ongoing processes or events.

[1317] "Data preprocessing" refers to processing collected data to convert it into a format suitable for analysis and model training.

[1318] "Training an AI model" refers to the learning process of using collected data to improve the performance of an artificial intelligence model.

[1319] "Making optimal decisions" refers to determining the most effective actions or choices based on collected data and trained AI models.

[1320] A system for implementing this invention has the following configuration. First, the server uses a generative AI to build a virtual brain model that makes decisions. Specifically, the manager's past decision-making data is used as learning data, and the generative AI learns the patterns. This builds a virtual brain that mimics the manager's decision-making.

[1321] Next, the server has the means to speed up the PDCA cycle and improve the accuracy of the model, which allows the virtual brain model to be continuously improved and enable more accurate decision-making.

[1322] Furthermore, the server has the means to learn and apply data obtained from individual brains to build a personalized virtual brain, thereby providing optimized decision-making support for each manager.

[1323] The server also includes a means for collecting and storing real-time data from the production line in the factory. Specifically, sensors are used to obtain data on product quality, production speed, machine operation status, etc., and the data is stored in an SQL database.

[1324] The server also includes a means to preprocess the collected data and train the AI ​​model. Python is used to fill in missing values ​​and remove outliers, and TensorFlow is used to train the AI ​​model.

[1325] The server also includes a means to analyze production line data in real time using trained AI models to make optimal decisions, suggesting actions to adjust production speed or control quality.

[1326] For example, if product quality is declining, the AI ​​model will learn from past data what measures management has taken and suggest similar measures, such as temporarily slowing down production speed or performing maintenance on specific machines.

[1327] An example of a prompt is, "Please learn about the measures taken by management when product quality deteriorated in the past, and propose the optimal measures based on the current production line data."

[1328] In this way, the server can mimic the manager's decision-making in real time and achieve efficient production management.

[1329] The flow of the specific processing in the application example 1 will be described with reference to FIG.

[1330] Step 1:

[1331] The server collects data in real time from the production line in the factory. Specifically, it uses sensors to obtain data on product quality, production speed, machine operation status, etc. The input is raw data from the sensors, and the output is data stored in an SQL database.

[1332] Step 2:

[1333] The server preprocesses the collected data. Specifically, it uses Python to fill in missing values ​​and remove outliers. The input is raw data stored in an SQL database, and the output is the preprocessed, clean data.

[1334] Step 3:

[1335] The server uses the preprocessed data to train the AI ​​model. Specifically, it uses TensorFlow to learn from managers' past decision-making data and build a generative AI model. The input is the preprocessed data, and the output is the trained AI model.

[1336] Step 4:

[1337] The server uses the trained AI model to analyze production line data in real time. Specifically, it inputs current production line data into the AI ​​model to make optimal decisions. The input is production line data collected in real time, and the output is a proposal for optimal decisions.

[1338] Step 5:

[1339] The server adjusts the production line based on optimal decision-making. Specifically, it proposes actions for adjusting production speed and quality control. The input is the decision-making proposal from the AI ​​model, and the output is the adjustment results for the production line.

[1340] Step 6:

[1341] The user checks the suggestions from the server and manually adjusts them as necessary. Specifically, based on the suggestions from the server, the user can take actions such as temporarily slowing down the production speed or performing maintenance on a specific machine. The input is the suggestion from the server, and the output is the result of the adjustment made by the user.

[1342] Step 7:

[1343] The server collects the adjustment results as data again and saves them as training data for the next run. This completes the PDCA cycle, improving the accuracy of the model. The input is the adjustment results made by the user, and the output is the training data for the next run.

[1344] Example 2

[1345] Next, a description will be given of Example 2 of Form 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."

[1346] Today's business environment is rapidly changing, requiring managers to make quick and accurate decisions. However, traditional decision support systems often require manual data updates and model retraining, making it difficult to quickly implement the PDCA cycle. They also lack personalized support that reflects the decision-making patterns of individual managers. This makes it difficult for managers to make quick decisions to adapt to a changing environment.

[1347] 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.

[1348] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for periodically updating the learning data and learning new decision-making patterns, means for retraining and evaluating the generative AI model, means for inputting prompt sentences to request specific predictions or proposals, and means for providing the generated predictions or proposals, thereby enabling managers to make quick and accurate decisions.

[1349] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and patterns from data.

[1350] A "virtual brain" is a model that uses artificial intelligence technology to mimic the decision-making process of the human brain.

[1351] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), executing (Do), evaluating (Check), and improving (Act).

[1352] "Personalization" refers to optimizing systems and services to suit the characteristics and needs of individual users.

[1353] "Training data" is a data set used to train an artificial intelligence model.

[1354] "Retraining" is the process of adding new data to an existing artificial intelligence model and training it again.

[1355] "Evaluation" is the process of measuring the performance of an artificial intelligence model and confirming its accuracy and effectiveness.

[1356] A "prompt sentence" is an instruction sentence that a user enters to request a specific prediction or suggestion from the system.

[1357] "Prediction" refers to an artificial intelligence model's estimation of future events or outcomes based on past data.

[1358] A "suggestion" is a specific recommendation of action or strategy that an AI model provides to a user.

[1359] This invention is a system that uses generative AI to build a virtual brain model for decision-making and improves the accuracy of the model by quickly implementing the PDCA cycle. Furthermore, it aims to build a personalized virtual brain by learning and applying data obtained from individual brains.

[1360] Hardware and software used

[1361] The server uses a computer equipped with a high-performance GPU (e.g., NVIDIA Tesla V100), and the software uses a machine learning framework (e.g., TensorFlow, PyTorch), which enables rapid processing of large amounts of data and efficient retraining of generative AI models.

[1362] Program processing

[1363] The server first collects data provided by the user, including past sales data, seasonal fluctuation data, competitor trends, etc. This data is uploaded in CSV or Excel file format.

[1364] The server then preprocesses the collected data, specifically by completing missing values, detecting and correcting outliers, and normalizing the data, thereby improving the quality of the data and the accuracy of the model.

[1365] The server retrains the generative AI model using the preprocessed data. For retraining, a dataset divided into training data and test data is used. The server initializes the generative AI model and retrains the model using the training data. After retraining, the performance of the model is evaluated using the test data.

[1366] Once the evaluation is complete, the model is used to assist the user in making decisions. The user inputs a prompt to request a specific prediction or suggestion. The server uses the generative AI model to generate the requested prediction or suggestion and provides it to the user.

[1367] Specific examples

[1368] Consider a company that wants to forecast market demand. Users upload sales data from the past three years, seasonal fluctuation data, and competitor trends to a server. The server uses this data to retrain a generative AI model and generate demand forecasts for the next quarter.

[1369] Example prompt sentence:

[1370] "Retrain your generative AI model using the past three years of sales data, seasonal fluctuation data, and competitor trends data to forecast demand for product A for the next quarter."

[1371] By entering this prompt, the server automatically collects data, retrains the model, and provides demand forecasts, allowing businesses to make fast and accurate decisions.

[1372] The flow of the identification process in the second embodiment will be described with reference to FIG.

[1373] Step 1: Data collection

[1374] Users upload past sales data, seasonal fluctuation data, and competitor trends data to the server. Data in CSV or Excel file format is used as input. The server receives these files and stores them in the database. Specifically, the server validates the file format and stores them appropriately in the database.

[1375] Step 2: Data Preprocessing

[1376] The server pre-processes the collected data. As input, it uses the raw data stored in the database. Specifically, the server:

[1377] Imputing missing values: Imputing missing values ​​with the mean or median.

[1378] Outlier detection and correction: Statistical methods are used to detect outliers and correct them appropriately.

[1379] Data normalization: Standardizing or normalizing data.

[1380] As an output, pre-processed data is obtained.

[1381] Step 3: Retrain the model

[1382] The server retrains the generative AI model using the preprocessed data. The preprocessed data is used as input. Specifically, the server performs the following operations:

[1383] Data Split: Split the preprocessed data into training data and test data.

[1384] Initialize model: Initialize the generative AI model.

[1385] Train the model: Retrain the model using the training data.

[1386] The output is a retrained generative AI model.

[1387] Step 4: Model evaluation

[1388] The server evaluates the performance of the retrained model. The test data and the retrained model are used as input. Specifically, the server performs the following operations:

[1389] Run prediction: Make a prediction for the model using the test data.

[1390] Calculate evaluation metrics: Calculate evaluation metrics such as precision, recall, and F1 score.

[1391] As an output, the evaluation result is obtained.

[1392] Step 5: Decision support

[1393] The user inputs a prompt to request a specific prediction or suggestion. The prompt and the retrained model are used as input. Specifically, the server performs the following actions:

[1394] Parsing prompts: Parse the prompts entered by the user.

[1395] Generate predictions and recommendations: Use generative AI models to generate the requested predictions and recommendations.

[1396] As an output, the generated predictions and suggestions are provided to the user.

[1397] Example prompt sentence:

[1398] "Retrain your generative AI model using the past three years of sales data, seasonal fluctuation data, and competitor trends data to forecast demand for product A for the next quarter."

[1399] (Application example 2)

[1400] Next, a description will be given of Application Example 2 of Form 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."

[1401] Inventory management and delivery route optimization are important issues for modern logistics centers. However, traditional manual management methods have limitations when it comes to quickly responding to changes in the business environment. In particular, delays in updating inventory and delivery data make efficient operations difficult, resulting in increased costs and decreased customer satisfaction. To solve these issues, an automated decision-making support system using AI is needed.

[1402] 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 constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for periodically updating inventory data and delivery data of the logistics center and proposing optimal inventory management and delivery routes, and means for periodically updating data and relearning the model to make decisions in response to changes in the business environment. This makes it possible to optimize inventory management and delivery routes at the logistics center, enabling rapid response to changes in the business environment.

[1403] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and patterns from data.

[1404] A "virtual brain" is a model that uses artificial intelligence technology to mimic the decision-making process of the human brain.

[1405] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[1406] "Personalization" refers to the individual optimization of services and content based on the characteristics and behavior of individual users.

[1407] "Inventory data" refers to data that includes information such as the quantity of products at the logistics center, arrival date, and shipping date.

[1408] "Delivery data" refers to data that includes information such as the delivery route, delivery time, and destination of the product departing from the logistics center.

[1409] The "business environment" refers to the external and internal factors that affect a company's business operations, including market trends, competitive conditions, and regulations.

[1410] A "decision support system" is a system that uses data analysis and artificial intelligence technology to provide information and suggestions to managers and administrators to help them make optimal decisions.

[1411] The system for implementing this invention aims to optimize inventory management and delivery routes in logistics centers. The system uses generative AI to build a virtual brain model for decision-making, speeding up the PDCA cycle and improving the accuracy of the model. It also learns and applies data obtained from individual brains to build a personalized virtual brain.

[1412] The server uses the following hardware and software:

[1413] Hardware: Server or cloud environment (e.g. AWS, Google Cloud)

[1414] Software: Python, pandas (data manipulation), scikit-learn (machine learning), schedule (regular execution)

[1415] The server first periodically updates inventory and delivery data, allowing it to propose optimal inventory management and delivery routes based on the latest data. Specifically, inventory data includes information such as product ID, stock quantity, arrival date, and shipping date, while delivery data includes information such as delivery ID, departure point, destination, and delivery time.

[1416] The server then uses generative AI to build a virtual brain model. This model is then incorporated into the "Smart Logistics Manager" smartphone application used by logistics center operators. The application provides optimized inventory management, delivery routes, real-time data updates and analysis, and decision-making support in response to changes in the business environment.

[1417] The server periodically updates the data and retrains the model to quickly respond to changes in the business environment, enabling inventory management and delivery routes at logistics centers to be optimized and leading to more efficient operations.

[1418] As a concrete example, the following prompt sentence can be used:

[1419] "Use inventory and delivery data from your distribution center to create an AI model that suggests optimal inventory management and delivery routes. The model will update the data regularly and make decisions that respond to changes in the business environment."

[1420] In this way, the server can provide the logistics center operator with information to help them make optimal decisions.

[1421] The flow of the specific processing in the application example 2 will be described with reference to FIG.

[1422] Step 1:

[1423] The server reads inventory data and delivery data. Specifically, the server obtains inventory data (product ID, stock quantity, arrival date, shipping date, etc.) and delivery data (delivery ID, departure point, destination, delivery time, etc.) from a CSV file or database. The input is inventory data and delivery data, and the output is a data frame containing these data.

[1424] Step 2:

[1425] The server preprocesses the acquired data, specifically by imputing missing values, normalizing data, encoding categorical data, etc. The input is the data frame acquired in step 1, and the output is the preprocessed data frame.

[1426] Step 3:

[1427] The server trains a generative AI model using the preprocessed data. Specifically, it splits the data into training data and test data, and trains it using a random forest regression model. The input is the preprocessed data frame, and the output is the trained generative AI model.

[1428] Step 4:

[1429] The server uses a trained generative AI model to optimize inventory management and delivery routes. Specifically, it inputs new data into the model and predicts optimal inventory levels and delivery routes. The input is new inventory data and delivery data, and the output is a proposal for optimal inventory levels and delivery routes.

[1430] Step 5:

[1431] The server evaluates the accuracy of the generative AI model. Specifically, it evaluates the model's predictive accuracy using test data and adjusts the model's parameters as necessary. The input is the test data and the trained generative AI model, and the output is the model's accuracy evaluation result.

[1432] Step 6:

[1433] The server periodically updates the data and retrains the model. Specifically, it uses a schedule library to perform data updates and model retraining at midnight every day. The input is the latest inventory and delivery data, and the output is an updated generative AI model.

[1434] Step 7:

[1435] The user uses the smartphone application "Smart Logistics Manager" to check the optimal inventory management and delivery route proposals provided by the server. Specifically, the application retrieves data from the server and displays it to the user. The input is the proposal data from the server, and the output is the optimization proposal displayed to the user.

[1436] Example 3

[1437] Next, a third embodiment of the third embodiment will be described. 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."

[1438] Conventional decision support systems have difficulty reflecting the decision-making styles of individual managers, and can only provide support based on general algorithms. Therefore, personalized support that takes into account the individual decision-making criteria and styles of managers is required. In addition, efficient and accurate methods are required for preprocessing collected data and model training.

[1439] The specific processing by the specific processing unit 290 of the data processing device 12 in the third embodiment is realized by the following means.

[1440] In this invention, the server includes means for collecting individual electroencephalogram data, means for collecting behavioral data, means for preprocessing the collected electroencephalogram data and behavioral data, means for training a generative AI model using the preprocessed data, means for constructing a personalized virtual brain using the trained generative AI model, and means for displaying the decision-making results of the virtual brain, thereby enabling personalized support that reflects the individual decision-making style of the manager.

[1441] "Individual electroencephalogram data" refers to data of electrical signals acquired to measure an individual's brain activity.

[1442] "Behavioral data" refers to data that records an individual's behavior, operation history, activity log, etc.

[1443] "Preprocessing" is the process of performing processes such as noise removal and normalization on collected data to convert it into a format suitable for machine learning models.

[1444] A "generative AI model" is an artificial intelligence model that is generated using machine learning algorithms to learn from data and perform specific tasks.

[1445] A "personalized virtual brain" is a virtual brain model designed to reflect the decision-making style of an individual manager.

[1446] "Decision-making results" are the results of judgments and choices made by the virtual brain in a specific business situation.

[1447] "Display means" refers to devices or software that visually present the decision-making results of the virtual brain to the user.

[1448] This invention is a system that collects individual electroencephalogram (EEG) data and behavioral data, trains a generative AI model based on the data, and constructs a personalized virtual brain. Specific embodiments of this system are described below.

[1449] First, the user wears an EEG device to collect the manager's brainwave data. The device acquires the brainwave data from the EEG device in real time and sends it to a server. For example, a manager wears the EEG device during a meeting, collecting brainwave data during that time. The user also installs dedicated software on their PC or smartphone to collect the manager's behavioral data. The device records the manager's operation history and activity log and sends it to the server. For example, it collects the operation history when the manager checks email or edits documents.

[1450] The server then stores the received EEG and behavioral data in a database. The server then preprocesses the data, specifically removing noise and normalizing the data to convert it into a format suitable for machine learning models. For example, the server filters the EEG data and checks the consistency of the timestamps of the behavioral data. The server uses a filtering algorithm to remove noise from the EEG data and standardize the timestamps of the behavioral data.

[1451] The server then uses the preprocessed data to train a generative AI model. Specifically, it uses machine learning libraries such as Python's TensorFlow and PyTorch. The server trains the model based on past data to learn the manager's decision-making style. For example, it trains the model using brainwave and behavioral data from when a manager decides whether to enter a new market. The server generates a training dataset and optimizes the model's parameters.

[1452] Next, the server uses the trained generative AI model to construct a virtual brain that reflects the manager's decision-making style. The server sets the virtual brain's parameters and simulates the manager's decision-making in specific business situations. For example, when the virtual brain decides whether to enter a new market, it reproduces the decision-making process based on past data. The server saves the results of the virtual brain's simulation and provides them to the user.

[1453] Finally, the user applies the constructed virtual brain to a real business situation. The terminal displays the virtual brain's decision-making results and provides them to the user. For example, the user can refer to the virtual brain's judgment when considering whether to enter a new market. The terminal displays the virtual brain's decision-making results in graphs and text format, making them easy for the user to understand.

[1454] Examples of prompts include:

[1455] "When a manager decides whether to enter a new market, the virtual brain learns what factors he or she considers based on past brainwave and behavioral data, and then makes a decision in a similar situation."

[1456] In this way, the server collects the manager's EEG data and behavioral data and uses the generative AI model to build a personalized virtual brain. This enables the virtual brain, which reflects the manager's decision-making style, to make decisions in specific business situations. The flow of the specific processing in Example 3 will be explained using Figure 15.

[1457] Step 1:

[1458] The user wears an EEG device to collect the executive's brainwave data. The terminal acquires the brainwave data from the EEG device in real time and transmits it to the server. The input is the brainwave data from the EEG device, and the output is the brainwave data transmitted to the server. Specifically, the executive wears the EEG device during a meeting, and brainwave data is collected during that time.

[1459] Step 2:

[1460] To collect behavioral data on managers, users install dedicated software on their PCs or smartphones. The devices record the manager's operation history and activity logs and send them to a server. The input is the operation history and activity logs from the PC or smartphone, and the output is the behavioral data sent to the server. Specifically, the software collects operation history when the manager checks email or edits documents.

[1461] Step 3:

[1462] The server stores the received EEG data and behavioral data in a database. The input is the EEG data and behavioral data sent to the server, and the output is the data stored in the database. Specifically, the server creates a new entry in the database and stores the received data.

[1463] Step 4:

[1464] The server preprocesses the data. Specifically, it removes noise and normalizes the data, converting it into a format suitable for machine learning models. The input is EEG data and behavioral data stored in a database, and the output is the preprocessed data. Specifically, the server uses a filtering algorithm to remove noise from the EEG data and standardize the timestamps of the behavioral data.

[1465] Step 5:

[1466] The server uses the preprocessed data to train the generative AI model. Specifically, it uses machine learning libraries such as Python's TensorFlow and PyTorch. The input is the preprocessed EEG data and behavioral data, and the output is a trained generative AI model. Specifically, the server generates a training dataset and optimizes the model parameters.

[1467] Step 6:

[1468] The server uses a trained generative AI model to build a virtual brain that reflects the manager's decision-making style. The input is the trained generative AI model, and the output is the virtual brain. Specifically, the server sets the parameters of the virtual brain and simulates the manager's decision-making in specific business situations.

[1469] Step 7:

[1470] The user applies the constructed virtual brain to actual business situations. The terminal displays the decision-making results of the virtual brain and provides them to the user. The input is the decision-making results of the virtual brain, and the output is the results displayed on the terminal. In concrete terms, the terminal displays the decision-making results of the virtual brain in graphs and text format, making them easy for the user to understand.

[1471] (Application example 3)

[1472] Next, a description will be given of Application Example 3 of Form Example 3. 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."

[1473] In conventional factory operations, it is difficult to optimize production lines and troubleshoot problems that reflect the manager's decision-making style, which hinders efficient operations. Additionally, because there is no system that can mimic the manager's decision-making, it is difficult to respond quickly when the manager is absent.

[1474] The specific processing by the specific processing unit 290 of the data processing device 12 in the application example 3 is realized by the following means.

[1475] In this invention, the server includes a means for constructing a virtual brain model that makes decisions using generative AI, a means for speeding up the PDCA cycle and improving the accuracy of the model, a means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, and a means for using the virtual brain that reflects the manager's decision-making style to optimize and troubleshoot production lines in a factory, thereby enabling efficient factory operation that reflects the manager's decision-making style.

[1476] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and models from data.

[1477] A "virtual brain" is an artificial intelligence model built based on an individual's brainwave and behavioral data, which mimics a specific decision-making style.

[1478] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[1479] "Personalization" refers to individually optimizing systems and services based on individual characteristics and behavior.

[1480] "Production line optimization" refers to making adjustments and improvements to streamline production processes within a factory and achieve maximum productivity.

[1481] "Troubleshooting" refers to procedures and methods for identifying and resolving problems or failures that occur in a system or process.

[1482] "Managerial decision-making style" refers to the tendency or pattern of how managers make judgments and choices in specific situations.

[1483] To implement the present invention, the following system configuration and program are required.

[1484] System Configuration

[1485] 1. Hardware:

[1486] Brainwave sensor: A device for acquiring brainwave data from business managers.

[1487] Behavioral tracking system: A device for capturing behavioral data of business managers.

[1488] Server: A computer system for processing data and constructing a virtual brain.

[1489] Factory robots: Robots for production line optimization and troubleshooting.

[1490] 2. Software:

[1491] Data collection software: Programs for collecting data from EEG sensors and behavioral tracking systems.

[1492] Generative AI model: An algorithm for constructing a virtual brain based on collected data. Specifically, it uses a multi-layer perceptron (MLPClassifier) ​​using the scikit-learn library.

[1493] Optimization software: A program that uses a virtual brain to optimize and troubleshoot production lines.

[1494] Program processing explanation

[1495] The server first collects the manager's brainwave and behavioral data from EEG sensors and a behavioral tracking system. This data is then sent to the server via data collection software. The server then uses a generative AI model to build a virtual brain based on the collected data. This virtual brain is designed to mimic the manager's decision-making style.

[1496] After the virtual brain is built, the server uses optimization software to optimize and troubleshoot the production line. Specifically, the virtual brain uses the current state of the production line in the factory as input and makes decisions to make optimal adjustments. When a problem occurs, the virtual brain uses the problem data as input and proposes the optimal solution.

[1497] Specific examples

[1498] For example, if a machine breaks down in a factory, the server will suggest the optimal repair procedure based on past data. The virtual brain, which has learned the decisions managers have made in the past in response to similar breakdowns, will suggest repair procedures in real time, enabling a fast and efficient response.

[1499] Prompt Sentence Examples

[1500] "A machine has broken down in the factory. Please suggest the optimal repair procedure based on past data."

[1501] In this way, by using a virtual brain that reflects the manager's decision-making style, production efficiency within the factory can be improved.

[1502] The flow of the specific processing in Application Example 3 will be described with reference to FIG.

[1503] Step 1:

[1504] The server collects the manager's EEG data and behavioral data from the EEG sensor and behavior tracking system. These data are sent to the server through the data collection software. The input is EEG data and behavioral data, and the output is raw data stored on the server.

[1505] Step 2:

[1506] The server preprocesses the collected EEG and behavioral data, specifically by removing noise and normalizing the data. The input is raw data, and the output is preprocessed data.

[1507] Step 3:

[1508] The server uses the preprocessed data to train a generative AI model. Specifically, it uses the scikit-learn library to build a multilayer perceptron (MLPClassifier) ​​and trains it on the data. The input is the preprocessed data, and the output is the trained generative AI model.

[1509] Step 4:

[1510] The server uses a trained generative AI model to build a virtual brain, which is designed to mimic the decision-making style of a manager. The input is the trained generative AI model, and the output is the virtual brain.

[1511] Step 5:

[1512] The server inputs the current state of the production line in the factory and uses the virtual brain to make decisions for optimal adjustments. Specifically, production line data obtained from sensors is input into the virtual brain, which then outputs optimal adjustment proposals. The input is production line data, and the output is adjustment proposals.

[1513] Step 6:

[1514] When a problem occurs, the server inputs the problem data and uses the virtual brain to propose the optimal solution. Specifically, data on breakdowns and problems is input into the virtual brain, which then outputs the optimal repair procedures and solutions. The input is problem data, and the output is the solution.

[1515] Step 7:

[1516] The user adjusts the production line and performs troubleshooting based on the adjustment proposals and solutions provided by the server. Specifically, the user operates or repairs the machine according to instructions from the server. The input is the instruction from the server, and the output is the result of the adjustment or repair that was performed.

[1517] 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.

[1518] "Example 1"

[1519] In one embodiment of the present invention, the emotion engine recognizes the user's emotion by extracting the emotion from the user's facial expression, voice, text message, etc. For example, if the user shows an expression that expresses joy, the emotion engine recognizes the information as "joy."

[1520] "Example 2"

[1521] Furthermore, the emotion data obtained from the emotion engine is provided to the generative AI. The generative AI learns this emotion data and makes decisions that take into account the user's emotional state. For example, if the user expresses joy, the generative AI will take that information into account and provide decisions that are likely to make the user feel joyful.

[1522] "Example 3"

[1523] The personalized virtual brain also provides decision-making that takes into account the user's emotional data. Specifically, it adjusts the priority of decisions depending on the user's emotional state. For example, if the user is feeling stressed, it will prioritize options that will reduce stress.

[1524] The processing flow of each embodiment will be described below.

[1525] "Example 1"

[1526] Step 1: The emotion engine extracts emotions from the user's facial expressions, voice, text messages, etc.

[1527] Step 2: The emotion engine converts the extracted emotions into emotion labels such as "joy," "sadness," and "anger."

[1528] Step 3: The emotion engine provides emotion labels to the generative AI.

[1529] "Example 2"

[1530] Step 1: The generative AI receives the emotion labels provided by the emotion engine.

[1531] Step 2: The generative AI uses the received emotion labels as training data to build a decision-making model that takes the user's emotional state into account.

[1532] Step 3: The generative AI uses the constructed decision-making model to provide a decision that takes into account the user's emotional state.

[1533] "Example 3"

[1534] Step 1: The personalized virtual brain receives the decisions provided by the generative AI.

[1535] Step 2: The personalized virtual brain adjusts the received decisions according to the user's emotional state.

[1536] Step 3: The personalized virtual brain provides tailored decision-making to the user.

[1537] Example 1

[1538] Next, a description will be given of Example 1 of Form 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."

[1539] Conventional decision support systems have the problem that it is difficult to imitate the decision-making patterns of managers and are unable to provide decision support that takes user emotions into account. Furthermore, there is a lack of means to effectively utilize user feedback to improve the model. For this reason, there is a need for a system that can more accurately imitate managerial decision-making and provide personalized decision support that takes user emotions into account.

[1540] 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.

[1541] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for collecting user emotion data, means for analyzing the collected emotion data and recognizing the user's emotion, means for supporting decision-making based on the recognized emotion, and means for collecting user feedback and improving the model. This enables personalized decision-making support that accurately mimics managerial decision-making and takes user emotion into consideration.

[1542] "Generative AI" is an algorithm that uses artificial intelligence technology to generate new information and models from data.

[1543] A "virtual brain" is a system that uses artificial intelligence technology to mimic the functions of the human brain and perform decision-making and information processing.

[1544] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[1545] "Personalization" refers to optimizing systems and services to suit the characteristics and needs of individual users.

[1546] "Emotion data" is information about emotions extracted from the user's facial expressions, voice, text messages, and the like.

[1547] "Emotion recognition" is a technology that analyzes collected emotional data and identifies the user's emotional state.

[1548] "Decision support" refers to providing necessary information and suggestions to support users when making decisions.

[1549] "Feedback" refers to the evaluation and reaction information provided by the user regarding the system's suggestions and results.

[1550] This invention is a system that uses generative AI to construct a virtual brain model that makes decisions and provides personalized decision-making support that takes into account the user's emotions. Specific embodiments of this system are described below.

[1551] The server uses generative AI to build a virtual brain model that makes decisions. Specifically, the server connects to the company's database (e.g., MySQL) and collects data on past decisions made by managers. The collected data is preprocessed using the Pandas library to impute missing values, normalize the data, and encode categorical data. A neural network model is then built using TensorFlow, and the model is trained using the preprocessed data. This model is then used to predict the manager's decision-making patterns.

[1552] The device collects the user's emotional data by capturing the user's facial expressions with the device's camera, recording audio with the microphone, and capturing text messages. This data is then sent to the server in real time via WebSocket.

[1553] The server analyzes the received emotional data. Specifically, it uses OpenCV to analyze facial expressions and IBM Watson to analyze voice and text messages. For example, if a user smiles, the server recognizes that information as "joy."

[1554] The server uses the trained AI model to assist users in decision-making. When a user asks a question about the launch of a new product, the server generates optimal suggestions based on past decision-making patterns. These suggestions are generated by the generative AI model.

[1555] As a concrete example, we use historical data on when a manager decides to launch a new product. The server provides this data to the AI, which then learns the patterns. When a user asks a question about launching a new product, the AI ​​suggests the best decision based on past patterns.

[1556] Example prompt sentence:

[1557] "Please provide optimal recommendations for the next new product launch based on past new product launch decision-making data."

[1558] The user receives suggestions from the server and provides feedback on the suggestions. Specifically, the user indicates their satisfaction with the suggestions through facial expressions and voice. The device collects this feedback and sends it to the server. The server uses this feedback to further improve the AI ​​model.

[1559] Example prompt sentence:

[1560] "Analyze how users feel when they receive a new product suggestion and tailor your next suggestion accordingly."

[1561] In this way, personalized decision support that accurately mimics the decision-making of managers and takes into account the user's emotions becomes possible.

[1562] The flow of the identification process in the first embodiment will be described with reference to FIG.

[1563] Step 1: Data collection

[1564] The server connects to the company's database and collects data on past decisions made by management. Specifically, the server executes the query "SELECT FROM decision_data WHERE date > '2020-01-01'" on the MySQL database to retrieve decision data. This data includes information such as the type of decision, date and time, and outcome. The input is the database query, and the output is the retrieved decision data.

[1565] Step 2: Data Preprocessing

[1566] The server preprocesses the collected data. Specifically, it uses the Pandas library to impute missing values ​​and Scikit-learn to normalize the data. It also performs one-hot encoding on categorical data. The input is the collected decision data, and the output is the preprocessed data.

[1567] Step 3: Model training

[1568] The server uses the preprocessed data to train an AI model. Specifically, it uses TensorFlow to build a neural network model and trains the model using the preprocessed data as input. The input is the preprocessed data, and the output is a trained AI model.

[1569] Step 4: Collect emotional data

[1570] The device collects the user's emotional data. Specifically, it captures the user's facial expressions with the device's camera, records audio with the microphone, and obtains text messages. These data are sent to the server in real time via WebSocket. The input is the user's facial expressions, audio, and text messages, and the output is the collected emotional data.

[1571] Step 5: Emotion Recognition

[1572] The server analyzes the received emotion data. Specifically, it uses OpenCV to analyze facial expressions and IBM Watson to analyze voice and text messages. For example, if a user smiles, the server recognizes that information as "joy." The input is the collected emotion data, and the output is the recognized emotion information.

[1573] Step 6: Decision support

[1574] The server uses the trained AI model to assist the user in decision-making. Specifically, when the user inputs a prompt such as "Please give me the best proposal for the next new product launch," the server generates the best proposal based on past decision-making patterns. The inputs are the user's prompt and the trained AI model, and the output is the generated proposal.

[1575] Step 7: User Feedback

[1576] The user receives suggestions from the server and provides feedback on the suggestions. Specifically, the user indicates whether they are satisfied with the suggestions using facial expressions and voice. The device collects this feedback and sends it to the server. The server uses this feedback to further improve the AI ​​model. The input is the user's feedback, and the output is an improved AI model.

[1577] (Application example 1)

[1578] Next, a description will be given of Application Example 1 of Form 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."

[1579] Conventional factory management systems have difficulty reflecting the decision-making patterns of managers in real time, and lack support that takes into account the emotions of factory staff, resulting in problems that lead to reduced production efficiency and staff satisfaction. For this reason, there is a need for a system that can learn the decision-making patterns of managers, optimize factory production lines, and recognize staff emotions to provide appropriate feedback.

[1580] 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.

[1581] In this invention, the server includes means for constructing a virtual brain model that makes decisions using generative AI, means for speeding up the PDCA cycle and improving the accuracy of the model, means for learning and applying data obtained from an individual's brain to construct a personalized virtual brain, means for providing feedback based on the user's emotions using an emotion engine that recognizes the user's emotions, and means for learning the decision-making patterns of managers and optimizing factory production lines.This makes it possible to reflect the manager's decision-making patterns in real time and provide support that takes the emotions of factory staff into consideration.

[1582] "Generative AI" is an algorithm that uses artificial intelligence techniques to generate new information and patterns from data.

[1583] A "virtual brain" is a model that uses artificial intelligence technology to mimic the decision-making process of the human brain.

[1584] The PDCA cycle is a method for continuously improving business operations and processes by repeating the four steps of planning (Plan), doing (Do), evaluating (Check), and improving (Act).

[1585] "Personalization" refers to providing services and functions that are customized to suit the characteristics and needs of individual users.

[1586] An "emotion engine" is a technology for recognizing and analyzing emotions from a user's facial expressions, voice, text messages, etc.

[1587] "Managerial decision-making patterns" are decision-making tendencies and rules extracted based on data on decisions made by managers in the past.

[1588] "Production line optimization" refers to making adjustments and improvements to streamline a factory's production processes, reducing costs and improving productivity.

[1589] "Feedback" refers to evaluations and advice provided regarding the system or user's behavior.

[1590] As an embodiment of this invention, we will explain a system for smart factory management robots as an example. This system uses generative AI to build a virtual brain model for decision-making and optimize the factory production line. It also uses an emotion engine to recognize the emotions of factory staff and provide appropriate feedback.

[1591] The server uses generative AI to build a virtual brain model that makes decisions. This model learns from the manager's past decision-making data and extracts patterns, making it possible to imitate the manager's decision-making.

[1592] Next, the server speeds up the PDCA cycle to improve the accuracy of the model. This cycle consists of four steps: Plan, Do, Check, and Act, which are repeated to continuously improve the accuracy of the model.

[1593] Furthermore, the server learns and applies data obtained from the individual's brain to build a personalized virtual brain, enabling decision-making support tailored to the characteristics and needs of each ...

Claims

1. A means for constructing a virtual brain model that mimics the decision-making style of a manager using a trained generative AI; means for iteratively planning, executing, evaluating, and improving the model; and means for training said model using past decision-making data of said manager; A means for recognizing the emotions of the manager; A support means for supporting decision making by the model based on the emotion; a means for improving said model using said management's evaluations and reactions to said model's suggestions and decisions; A system including:

2. 2. The system according to claim 1, wherein, when a machine breakdown occurs in the factory, said support means proposes a repair procedure for the broken machine based on data relating to past failure cases.

3. 2. The system according to claim 1, wherein said support means optimizes the production line of the factory by learning the decision-making patterns of said manager using said model.

Images