system

Abstract

We provide the system. [Solution] In order to reuse existing large-scale language models, a means for selecting and loading older generation large-scale language models to be linked to new generation large-scale language models, A means for the new generation of large-scale language models to readjust the old generation of large-scale language models using datasets related to a specific field, A means of monitoring the readjustment process using an auto-tuning function and terminating the learning process at the optimal time, A system that includes this.

Description

Technical Field

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

Background Art

[0002] Patent Document 1 discloses a method for controlling a persona chatbot, which is performed by at least one processor, including the steps of receiving a user utterance, adding the user utterance to a prompt including an instruction sentence related to an explanation of a character of the chatbot, 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

Summary of the Invention

Problems to be Solved by the Invention

[0004] In the development competition of rapidly evolving large language models, old-generation models become unnecessary due to the emergence of new-generation models and are often discarded as they are. This causes waste of resources and poses a problem of being uneconomical and inefficient for enterprises. Also, although there is a possibility that they can be reused for specialized applications in specific fields, an efficient and effective method for realizing this has not been established, which is an issue.

Means for Solving the Problems

[0005] This invention solves these problems by providing a system for reusing older generation large-scale language models. Specifically, it links new generation large-scale language models with older generation models via a data communication channel and efficiently readjusts them using datasets relevant to specific domains. Furthermore, it generates reusable, specialized models by monitoring the learning process using an auto-tuning function and terminating learning at the optimal time. This enables companies to make the most of their existing resources and improve economic efficiency and productivity.

[0006] A "large-scale language model" is an artificial intelligence model trained on extremely large datasets, designed to understand and generate human language.

[0007] A "next-generation large-scale language model" is a large-scale language model that employs the latest technologies and algorithms to achieve higher performance and efficiency than previous generations.

[0008] An "older generation large-scale language model" is a large-scale language model that uses relatively older technologies compared to newly developed models, but still has value.

[0009] A "link" is the establishment of a connection for communicating data between two or more large language models.

[0010] A "dataset" is an organized collection of data used to train a model on knowledge in a specific field.

[0011] "Readjustment" is the process of adjusting the parameters of an existing model to optimize it for a specific purpose.

[0012] The "auto-tuning function" is a feature that automatically monitors the learning process and adjusts tuning parameters as needed.

[0013] A "specialized model" is a large-scale language model optimized to deliver performance suitable for a specific field or application. [Brief explanation of the drawing]

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

Embodiments for Carrying Out the Invention

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

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

[0017] In the following embodiments, a numbered processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units 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), and the like.

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

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

[0020] In the following embodiments, the signed communication interface (I / F) is an interface that includes a communication processor and an antenna, etc. The communication interface manages communication between multiple computers. Examples of communication standards applicable to the communication interface include wireless communication standards such as 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark).

[0021] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0022] [First Embodiment]

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

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

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

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

[0027] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0028] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0029] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

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

[0031] As shown in Figure 2, in the data processing device 12, a specific processing 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" related to the technology of this 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 according to the specific processing program 56 executed on the RAM 30.

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

[0033] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

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

[0035] This invention relates to a system that provides a method for making older, large-scale language models reusable in specific fields.

[0036] First, the server loads a large-scale, older-generation language model into memory. This is done by selecting it from storage and preparing the necessary parameters and weights.

[0037] Next, the server links the new generation large-scale language model with the old generation model. This enables data communication between the two models, creating a state where the new generation model can effectively utilize the insights and data of the old generation model.

[0038] Subsequently, the server prepares a dataset for a specific field and provides it to the next-generation model. For example, in the medical field, it would select the latest medical papers and diagnostic data as the dataset. This prepares the next-generation model to effectively learn the information required in that specific field.

[0039] During the retuning phase, the new generation of large-scale language models is retrained using specific domain datasets, referencing the output of the previous generation models. This improves the model's performance as a specialized model for that particular domain.

[0040] By utilizing the auto-tuning function, the server monitors the model's learning state in real time and dynamically adjusts training parameters as needed. This mechanism reduces wasted resources and enables effective reuse.

[0041] Finally, users utilize these retrained, specialized models to apply them to specific domain tasks and problem-solving. This allows them to maximize the knowledge gained from older models while enabling advanced applications made possible by the latest technologies.

[0042] For example, when a user builds a diagnostic support system for a medical setting, using the specialized model obtained through this process makes it possible to achieve more efficient and accurate diagnostic support.

[0043] The following describes the processing flow.

[0044] Step 1:

[0045] The server selects a large-scale language model from the previous generation and loads it from storage into memory. This completes the restoration of the previous generation model, making it ready for retraining.

[0046] Step 2:

[0047] The server links the new generation of large-scale language models with the old generation models. It establishes a data communication channel and configures it so that the output of the old generation model is taken in as input to the new generation model.

[0048] Step 3:

[0049] The server retrieves datasets related to a specific field from the database and provides them to the next-generation model. These datasets contain knowledge in a specific field and form the basis for learning.

[0050] Step 4:

[0051] A new generation of large-scale language models retunes itself using specific domain datasets, referencing the output of older models. This enhances the model's capabilities as a domain-specific model.

[0052] Step 5:

[0053] The server uses an auto-tuning function to monitor the learning process in real time. It adjusts the learning rate and parameters as needed to achieve optimal training results.

[0054] Step 6:

[0055] Users utilize retrained, specialized models to apply them to tasks in specific fields. For example, a model specialized in diagnostic support in the medical field can leverage its capabilities to support users' work.

[0056] (Example 1)

[0057] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0058] When reusing older generation information processing equipment in specific industrial fields, the challenge lies in effectively utilizing that knowledge with newer equipment while minimizing wasteful resource consumption and achieving rapid and advanced adaptation. In particular, there is a problem in efficiently exchanging information between older and newer generation information processing equipment.

[0059] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0060] In this invention, the server includes means for selecting and loading older generation information processing devices as information formats from a digital data storage device; means for connecting a new generation information processing device with an older generation information processing device to enable information exchange; means for the new generation information processing device to readjust the older generation information processing device using a collection of information related to a specific industrial field; and means for monitoring the readjustment procedure using a self-optimization function and adjusting the learning procedure in a variable manner. This makes it possible to adapt the knowledge of the older generation information processing device to modern requirements and enable efficient information processing.

[0061] A "digital data storage device" is a recording device used to store information for long periods of time, and it is the basis for selecting and loading older generation information processing devices in a computer system.

[0062] "Information format" refers to the structure of data and programs, and is the basic unit by which computer systems read and process information.

[0063] An "information processing device" is a general term for hardware and software that receives digital data as input and converts it into output through calculations, analysis, and other processes.

[0064] "Means of selection and loading" refers to the processes and techniques for selecting a specific information processing device from storage and loading it into memory.

[0065] "Integration" refers to the process of sharing and coordinating data and functions among multiple information processing devices.

[0066] "Information exchange" refers to the process of sending and receiving data between different information processing devices and sharing knowledge and functions.

[0067] An "information collection" refers to a set of data, texts, and other materials related to a specific industrial field, and is used to acquire new knowledge.

[0068] "Self-optimization function" refers to a mechanism by which an information processing device adjusts its own parameters and configuration to maximize efficiency.

[0069] A "readjustment procedure" is the process of adjusting the parameters and operation of an information processing device in response to new conditions or information.

[0070] A "variable method" refers to procedures or strategies that can be flexibly changed depending on the situation or conditions.

[0071] This invention provides a method for realizing information processing in a new specific business field using older generation information processing equipment. Specifically, it utilizes newer generation information processing equipment to reuse knowledge from previous generations and efficiently adapt to new requirements.

[0072] First, the server selects and loads older generation information processing equipment from digital data storage devices. This involves the specific step of placing the parameters and weights required for the older generation model into memory within the computer system, typically using a fast, accessible memory device such as an SSD.

[0073] Next, the server connects the new generation of information processing equipment with the older generation. Here, a data communication channel is established via an API or software developed as needed, enabling the exchange of information. This creates an environment where the new generation model can leverage the valuable knowledge held by the older generation.

[0074] Furthermore, the server prepares a collection of information relevant to a specific industry and provides it to the next generation of information processing devices. This collection of information, for example in the medical field, includes the latest research materials and clinical data. In this way, the next generation of models can learn based on the latest information in their specific field.

[0075] During the automatic tuning phase, the server uses its self-optimization function to monitor the readjustment process. This allows for variable adjustment of the learning parameters, maximizing processing efficiency.

[0076] Finally, users utilize the retrained, specialized models to solve problems and perform tasks in specific business areas. For example, when building a diagnostic support system in a medical setting, this model is extremely useful in providing more accurate diagnostic results.

[0077] As a concrete example, by prompting the AI ​​generator with "Please suggest ways to improve the diabetes diagnostic model based on the latest medical papers," the user can obtain information that meets their specific requirements.

[0078] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0079] Step 1:

[0080] The server selects and loads an older generation information processing unit from a digital data storage device. It receives an older generation model file stored in storage as input. This file is loaded into memory, and the model's parameters and weights are expanded to make the model processable. Specifically, memory management software is used to place the data into high-speed memory. The output is the model expanded in memory.

[0081] Step 2:

[0082] The server links the new generation information processing unit with the old generation model. It receives interface information from both the old and new generation models as input. It establishes a data communication channel, enabling data exchange between the two models via APIs or custom programs. Specifically, it uses communication protocol setup software to establish the link. The output is the link state, indicating that information exchange is now possible.

[0083] Step 3:

[0084] The server collects data related to a specific industry sector and provides it to a next-generation model. As input, it extracts specialized datasets and performs data cleaning. The data is then converted into a format that the model can learn from using preprocessing software. Specifically, it uses data format conversion utilities to reshape the data. The output is a trainable dataset.

[0085] Step 4:

[0086] The server utilizes a new generation model and readjusts it based on the output of the previous generation model. The inputs used are the training dataset and the output data from the previous generation model. The new generation model updates its training parameters based on these inputs, learning domain-specific knowledge. Specifically, it executes machine learning algorithms and adjusts the parameters. The output is the adaptively trained new generation model.

[0087] Step 5:

[0088] The server monitors the learning state using its self-optimization function and optimizes the training process. It takes current learning state data and prediction accuracy information as input and makes real-time adjustments using dynamic learning management software. Specifically, it adjusts the learning rate and batch size using an optimization algorithm. The output is a new generation model with optimal performance.

[0089] Step 6:

[0090] Users utilize specialized models to perform tasks in specific business areas. The input consists of prompts for the generating AI model, such as, "Please suggest ways to improve the diabetes diagnostic model based on the latest medical papers." The model calculates based on these prompts and provides the user with improvements and related information as output. The results are displayed through a user interface. The output consists of specific information and suggestions tailored to the user's requests.

[0091] (Application Example 1)

[0092] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0093] The challenge lies in providing efficient methods that effectively utilize existing information processing models to enable data analysis and real-time problem detection in specific domains. In particular, in manufacturing operations, there is a need to leverage past data patterns to improve quality control and optimize production processes.

[0094] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0095] In this invention, the server includes means for selecting and loading an older generation information processing model linked to a new generation information processing model in order to reuse an existing data processing model; means for the new generation information processing model to readjust the older generation information processing model using a data set related to a specific domain; means for monitoring the readjustment process using an automatic adjustment function and terminating the learning process at the optimal timing; and means for acquiring information from measuring devices and image acquisition devices, comparing it with past data patterns derived from the older generation model to detect problems, and suggesting improvement measures. This enables increased efficiency and improved quality in the manufacturing process.

[0096] An "existing data processing model" refers to an information analysis system developed in the past that is used to analyze specific data.

[0097] A "next-generation information processing model" is an information analysis system built using the latest technology, which works in conjunction with older generation models to perform more advanced data processing.

[0098] A "specific domain" refers to a collection of knowledge and data related to a particular industry or business, and in this invention, the subject is particularly manufacturing operations.

[0099] A "data set" refers to a collection of information or records related to a specific domain, and is used by next-generation information processing models for learning and analysis.

[0100] The "automatic adjustment function" is a mechanism that monitors the learning process of an information processing model in real time and determines the optimal point at which learning should end.

[0101] "Measurement devices and image acquisition devices" are hardware devices used to collect data about the physical environment and objects, and they provide the input data necessary for information processing models.

[0102] "Past data patterns" refer to the results of past information analysis and trends accumulated by older generation data processing models, which newer generation models utilize for problem detection and improvement.

[0103] "A means of detecting problems and suggesting solutions" refers to the process by which an information processing model grasps specific conditions or phenomena in real time and derives appropriate solutions or guidelines.

[0104] The system that implements this invention is primarily executed by the server's functions. The server first selects an older generation information processing model to be linked to a newer generation information processing model in order to use an existing data processing model, and loads it into memory. This process involves reading model data from a storage device and deploying the model using deep learning frameworks such as TENSORFLOW® or PyTorch.

[0105] Next, the server prepares a data set related to a specific area. This data includes historical manufacturing data and quality control standards, and is integrated with real-time data from measuring and image acquisition devices. This data set is then provided to the next-generation model using data stream processing technology that includes real-time information.

[0106] During the learning process, the server utilizes its automatic tuning function to readjust the next-generation information processing model to ensure optimal performance. This includes monitoring the model's learning state and terminating the learning process at the appropriate time. At this stage, logic is incorporated to refer to past data patterns and suggest the most suitable improvement measures for the current situation.

[0107] For example, if the model detects a frequently occurring error on an assembly line for a particular part, it will immediately analyze similar past cases and suggest effective improvement measures. Based on these suggestions, the user can make specific adjustments to improve production efficiency. This system leverages the knowledge derived by the model to optimize the entire manufacturing process.

[0108] An example of a prompt to input into a generative AI model is: "Based on last year's quality inspection data, please make suggestions to improve the product defect rate. Pay particular attention to issues that are often overlooked in optical inspections." In response to this prompt, the model will provide specific suggestions and countermeasures in real time.

[0109] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0110] Step 1:

[0111] The server selects a legacy data processing model from storage and loads it into memory. At this stage, the server reads a file containing the parameters and weights of the legacy model and deploys the model into memory using a framework such as TensorFlow or PyTorch. The input is the model file, and the output is the legacy model deployed in memory.

[0112] Step 2:

[0113] The server prepares a data set related to a specific domain. This set includes historical manufacturing data and quality control standards, and is integrated with data collected in real time from measuring and image acquisition devices. The input is raw data acquired from databases and sensors, and the output is a data set ready for analysis.

[0114] Step 3:

[0115] The server links the old-generation model with the new-generation information processing model. This means setting up a data communication channel and creating a state where data can be exchanged between the two models. The input is the old-generation model and the new-generation model, and the output is the communicative, linked state.

[0116] Step 4:

[0117] The server trains a new generation information processing model using a prepared data set. During training, the new generation model self-adjusts by referencing data patterns obtained from the old generation model. The input is a data set in a specific domain, and the output is the adjusted new generation model.

[0118] Step 5:

[0119] The server utilizes an auto-tuning function to continuously monitor and readjust the model to ensure optimal performance. Here, performance metrics are checked in real time, and parameters are adjusted as needed. The input is the model state during training, and the output is the optimized model state.

[0120] Step 6:

[0121] Users receive improvement suggestions from the server and use them to optimize their production processes. Based on the information provided, users create concrete action plans and apply them to the manufacturing line. The input is suggestions from the new generation model, and the output is a concrete action plan.

[0122] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0123] This invention relates to a system that links older generation large-scale language models with newer generation models, and further combines them with an emotion engine that recognizes user emotions.

[0124] First, the server selects a large-scale language model from the older generation and loads it from storage into memory to prepare it for reuse. This step prepares the older model as a preliminary step before retraining.

[0125] Next, the server links the new generation large-scale language model with the old generation model via a data communication channel. This makes the insights derived from the old generation model available to the new generation model.

[0126] Subsequently, the server retrieves datasets related to a specific field from the database and provides them to the next-generation model. For example, in the medical field, the latest research data and patient information on diseases would be used as datasets.

[0127] Next, the emotion engine collects user emotion data through the device. This emotion data is obtained from user interactions and interpreted using natural language processing. As a result, the next generation of models receive emotion information as additional adaptive feedback, improving the accuracy of their learning.

[0128] During the retuning process, the server provides the new generation model with specialized datasets and sentiment data, allowing the model to be readjusted to deepen its understanding of specific domains. The auto-tuning function monitors the entire process and efficiently adjusts parameters.

[0129] Finally, users utilize retrained, specialized models to respond to specific domain needs and tasks. At this stage, the addition of sentiment-based analysis provides more personalized responses and support to the user.

[0130] As a concrete example, in the field of education, we can consider a system that allows teachers to evaluate students' participation based on their reactions and use that information to improve individualized instruction. This system uses an emotion engine to analyze students' responses in real time and adjust the educational content using that information, thereby enabling the delivery of more effective education.

[0131] The following describes the processing flow.

[0132] Step 1:

[0133] The server selects a large-scale language model from the previous generation and loads it from storage into memory. This makes the data and parameters of the older model available.

[0134] Step 2:

[0135] The server establishes a data communication channel to link the new generation of large-scale language models with the older generation models. This link allows the new generation models to leverage the knowledge of the older generation models.

[0136] Step 3:

[0137] The server retrieves datasets related to a specific field from a database and supplies them to the next-generation model. For example, it prepares specific medical data to enable the model to learn field-specific knowledge.

[0138] Step 4:

[0139] The emotion engine built into the device collects emotional data through interaction with the user. This includes voice tone and facial recognition data, which are then analyzed through natural language processing.

[0140] Step 5:

[0141] The server provides the next-generation model with domain-specific datasets and sentiment data collected from users, and performs retuning. The sentiment data is used as learning feedback and contributes to the model's refinement.

[0142] Step 6:

[0143] The server monitors the entire learning process using an auto-tuning function. It evaluates the model's learning effectiveness in real time and optimizes parameters as needed. As a result, effective model readjustment is achieved.

[0144] Step 7:

[0145] Users can utilize retrained, specialized models to obtain responses to specific challenges and needs in particular fields. For example, it becomes possible to provide personalized medical advice or educational content based on the user's emotions.

[0146] (Example 2)

[0147] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0148] To meet the evolving needs of information processing models and diverse application fields, there is a demand for personalized responses that reflect user sentiment while leveraging the knowledge gained from older models. Furthermore, the readjustment of information processing models for efficient operation in specific application fields remains a challenge.

[0149] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0150] In this invention, the server includes means for selecting and loading an older generation information processing model into memory; means for linking a new information processing model to the older generation information processing model using a communication path; means for acquiring a set of knowledge related to a specific application field and providing it to the new information processing model; means for collecting user emotional information through a terminal and interpreting it using natural language processing; means for incorporating the emotional information into the new information processing model and readjusting it; and means for monitoring the readjustment process and efficiently adjusting it by utilizing an automatic adjustment function. This makes it possible to provide personalized responses in a specific application field.

[0151] An "older generation information processing model" is a collection of information processing algorithms developed in previous generations, and it possesses more fundamental knowledge and data processing processes compared to models based on new technologies.

[0152] A "new information processing model" is an information processing system that employs the latest technologies and integrates knowledge from previous generation models to possess the ability to perform advanced data analysis and inference suitable for the application field.

[0153] A "communication path" is a physical or logical connection means for transmitting data and insights between information processing models, enabling the sending and receiving of information.

[0154] A "knowledge set" is a collection of data and information relating to a specific application field, including the latest case studies and research results related to that application field.

[0155] "Automatic adjustment function" is a technology that monitors information processing processes in real time and autonomously makes adjustments necessary to maintain optimal performance.

[0156] "User emotional information" refers to data about emotional responses and states obtained through interaction with the user, and is interpreted using natural language processing.

[0157] "Personalized responses" refer to information provision and actions tailored to the individual user's needs and circumstances, enabling effective support in specific application areas.

[0158] This invention is a system that provides personalized responses by efficiently combining older and newer information processing models, and further utilizing user emotional information.

[0159] First, the server selects an older generation information processing model and loads it from storage into memory. This prepares the older generation model for linking with the newer information processing model. Next, the server links the newer information processing model with the older generation model using a data communication path. This link allows the newer model to leverage the insights from the older generation model. The hardware used includes a database server and a communication network. The software employs machine learning algorithms and natural language processing tools.

[0160] The server also retrieves knowledge sets relevant to specific application areas from the database and provides them to a new information processing model. This is a step for the new model to learn information specific to the application area. For example, in the medical field, the latest research data and patient health information are used as the knowledge set.

[0161] The device plays a role in collecting user emotional information. This involves analyzing the user's facial expressions and voice using the camera and microphone, and interpreting the emotions through natural language processing. This process provides data that helps understand the user's current situation and needs, contributing to improving the learning accuracy of new information processing models.

[0162] Users receive personalized responses through the system. For example, in the field of education, teachers can analyze student responses and optimize learning plans based on that data. Examples of specific prompts include, "Please tell me about recent advances in medical technology," or "Please suggest ways to improve students' learning progress."

[0163] In this way, combining emotional information with older and newer information processing models enables effective and flexible responses in specific application areas.

[0164] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0165] Step 1:

[0166] The server selects an older generation information processing model from storage and loads it into memory. It uses the identifier and associated file path of the older generation model as input and outputs the model data loaded into memory. Specifically, the server executes the process of reading the model file from storage and placing it into memory.

[0167] Step 2:

[0168] The server links the new information processing model to the old generation information processing model using a data communication path. It takes specific information about the old and new generation models as input and outputs a state where the communication path between the models has been established. Specifically, the server connects to both models via an API and sets up a state where data can be exchanged.

[0169] Step 3:

[0170] The server retrieves a set of knowledge relevant to a specific application domain from a database and provides it to a new information processing model. It uses application domain query data as input and provides the model with appropriate knowledge information as output. Specifically, the server executes database queries and passes the extracted data to the model's learning system.

[0171] Step 4:

[0172] The device collects user emotional information. It uses signal data of the user's facial expressions and voice, acquired through the camera and microphone, as input, and obtains analyzed emotional data as output. Specifically, this involves an emotion analysis engine built into the device processing the signal data, identifying emotional patterns, and transmitting them as numerical data.

[0173] Step 5:

[0174] The server incorporates emotional data into the new information processing model and readjusts it. It uses user emotional data and a specialized knowledge set as input and designs the adjusted model parameters as output. Specifically, the server generates a feedback loop using the emotional data to optimize the model so that it outputs a response that is more appropriate for the application field.

[0175] Step 6:

[0176] The user receives personalized responses from the system. The user sends prompts as input and receives optimized response data as output. Specifically, the process involves the user communicating questions and requests to the system via a terminal, and the model returning information tailored to those requests.

[0177] (Application Example 2)

[0178] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0179] In modern information processing technology, providing personalized information that responds to user emotions is a crucial challenge. However, conventional technologies have struggled to accurately analyze user emotions and recommend optimized information based on that data. Furthermore, there is a need for a solution that effectively links and reuses both old and new information processing technologies.

[0180] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0181] In this invention, the server includes means for selecting and loading older generation information processing technologies linked to new generation information processing technologies in order to reuse existing information processing technologies; means for the new generation information processing technologies to readjust the older generation information processing technologies using a set of information related to a specific field; and means for generating personalized information using sentiment analysis technology. This enables precise information recommendations based on the user's emotions.

[0182] "Information processing technology" refers to the technology of processing, analyzing, transforming, or optimizing digital data to serve a specific purpose.

[0183] An "information set" is a collection of data and information related to a specific field, which is then analyzed and processed.

[0184] "Emotional analysis technology" is a technology that analyzes users' emotions and sensitivities and optimizes data processing and service provision based on the results.

[0185] "Communication technology" refers to technologies for transmitting, receiving, and exchanging data and information, and enables the coordination of different technologies and systems.

[0186] "Personalized information" refers to information customized based on the individual user's characteristics and preferences, aiming to provide more appropriate and efficient information.

[0187] The system of the present invention links older and newer information processing technologies to provide personalized information based on user sentiment data. The server first selects and loads older information processing technologies to reuse existing ones. Next, the newer information processing technologies readjust the older technologies using a set of information relevant to a specific field.

[0188] This process uses emotion analysis technology to acquire user emotion data and optimize the service. Users can provide their emotional information to the system through devices such as smartphones and smart glasses, and the server processes this data to provide personalized information.

[0189] Specifically, the server monitors the readjustment process using an auto-tuning function, completing the learning process at the optimal time. In this process, older and newer information processing technologies are effectively coordinated using communication technology. As a result, precise information recommendations based on user emotions are realized.

[0190] For example, an entertainment information application can recommend movies that match the user's current mood when they choose a film. If the user is feeling sad, the system will recommend a more cheerful comedy film, providing a service that is considerate of the user's feelings.

[0191] An example of a prompt message might be: "Recommend a movie using the emotions the user is currently feeling. If the user is sad, choose a comedy movie to cheer them up."

[0192] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0193] Step 1:

[0194] The server selects older generation information processing technologies and loads them from storage into memory. The input here is the identification information of the older generation technology, and the output is an instance of the loaded technology. This loading operation prepares the older generation technology for reuse.

[0195] Step 2:

[0196] The server uses communication technology to link new-generation information processing technology with older-generation technology. The input to this step is information from both the old and new generations of technology, and the output is the established link. This link allows the new generation technology to utilize the knowledge of the old generation.

[0197] Step 3:

[0198] The server retrieves a set of information related to a specific field from a database. The input is a request related to that field, and the output is the retrieved information set. The server uses this information to prepare for the re-engineering of next-generation technologies.

[0199] Step 4:

[0200] The device collects user emotional information using emotion analysis technology and sends it to a server. The input is data of the user's facial expressions and voice, and the output is the analyzed emotional information. This provides data based on the user's emotions.

[0201] Step 5:

[0202] The server uses emotional information and data sets to retune next-generation technologies. The input is emotional information and data sets, and the output is the parameters of the updated technologies. This allows the system to provide users with more relevant information.

[0203] Step 6:

[0204] The server monitors the readjustment process using an auto-tuning function and completes it at the optimal time. The input is the readjustment progress data, and the output is the readjustment completion time. This enables an efficient learning process.

[0205] Step 7:

[0206] The server uses tuned information processing technology to provide users with personalized information. The input consists of the user's request and the tuned technology, while the output is information optimized for the user. As a result, users can receive information that resonates with their own emotions.

[0207] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating 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.

[0208] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0209] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0210] [Second Embodiment]

[0211] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0212] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0213] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0214] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0215] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0216] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0217] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0218] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0219] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0220] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0221] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0222] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0223] This invention relates to a system that provides a method for making older, large-scale language models reusable in specific fields.

[0224] First, the server loads a large-scale, older-generation language model into memory. This is done by selecting it from storage and preparing the necessary parameters and weights.

[0225] Next, the server links the new generation large-scale language model with the old generation model. This enables data communication between the two models, creating a state where the new generation model can effectively utilize the insights and data of the old generation model.

[0226] Subsequently, the server prepares a dataset for a specific field and provides it to the next-generation model. For example, in the medical field, it would select the latest medical papers and diagnostic data as the dataset. This prepares the next-generation model to effectively learn the information required in that specific field.

[0227] During the retuning phase, the new generation of large-scale language models is retrained using specific domain datasets, referencing the output of the previous generation models. This improves the model's performance as a specialized model for that particular domain.

[0228] By utilizing the auto-tuning function, the server monitors the model's learning state in real time and dynamically adjusts training parameters as needed. This mechanism reduces wasted resources and enables effective reuse.

[0229] Finally, users utilize these retrained, specialized models to apply them to specific domain tasks and problem-solving. This allows them to maximize the knowledge gained from older models while enabling advanced applications made possible by the latest technologies.

[0230] For example, when a user builds a diagnostic support system for a medical setting, using the specialized model obtained through this process makes it possible to achieve more efficient and accurate diagnostic support.

[0231] The following describes the processing flow.

[0232] Step 1:

[0233] The server selects a large-scale language model from the previous generation and loads it from storage into memory. This completes the restoration of the previous generation model, making it ready for retraining.

[0234] Step 2:

[0235] The server links the new generation of large-scale language models with the old generation models. It establishes a data communication channel and configures it so that the output of the old generation model is taken in as input to the new generation model.

[0236] Step 3:

[0237] The server retrieves datasets related to a specific field from the database and provides them to the next-generation model. These datasets contain knowledge in a specific field and form the basis for learning.

[0238] Step 4:

[0239] A new generation of large-scale language models retunes itself using specific domain datasets, referencing the output of older models. This enhances the model's capabilities as a domain-specific model.

[0240] Step 5:

[0241] The server uses an auto-tuning function to monitor the learning process in real time. It adjusts the learning rate and parameters as needed to achieve optimal training results.

[0242] Step 6:

[0243] Users utilize retrained, specialized models to apply them to tasks in specific fields. For example, a model specialized in diagnostic support in the medical field can leverage its capabilities to support users' work.

[0244] (Example 1)

[0245] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0246] When reusing older generation information processing equipment in specific industrial fields, the challenge lies in effectively utilizing that knowledge with newer equipment while minimizing wasteful resource consumption and achieving rapid and advanced adaptation. In particular, there is a problem in efficiently exchanging information between older and newer generation information processing equipment.

[0247] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0248] In this invention, the server includes means for selecting and loading older generation information processing devices as information formats from a digital data storage device; means for connecting a new generation information processing device with an older generation information processing device to enable information exchange; means for the new generation information processing device to readjust the older generation information processing device using a collection of information related to a specific industrial field; and means for monitoring the readjustment procedure using a self-optimization function and adjusting the learning procedure in a variable manner. This makes it possible to adapt the knowledge of the older generation information processing device to modern requirements and enable efficient information processing.

[0249] A "digital data storage device" is a recording device used to store information for long periods of time, and it is the basis for selecting and loading older generation information processing devices in a computer system.

[0250] "Information format" refers to the structure of data and programs, and is the basic unit by which computer systems read and process information.

[0251] An "information processing device" is a general term for hardware and software that receives digital data as input and converts it into output through calculations, analysis, and other processes.

[0252] "Means of selection and loading" refers to the processes and techniques for selecting a specific information processing device from storage and loading it into memory.

[0253] "Integration" refers to the process of sharing and coordinating data and functions among multiple information processing devices.

[0254] "Information exchange" refers to the process of sending and receiving data between different information processing devices and sharing knowledge and functions.

[0255] An "information collection" refers to a set of data, texts, and other materials related to a specific industrial field, and is used to acquire new knowledge.

[0256] "Self-optimization function" refers to a mechanism by which an information processing device adjusts its own parameters and configuration to maximize efficiency.

[0257] A "readjustment procedure" is the process of adjusting the parameters and operation of an information processing device in response to new conditions or information.

[0258] A "variable method" refers to procedures or strategies that can be flexibly changed depending on the situation or conditions.

[0259] This invention provides a method for realizing information processing in a new specific business field using older generation information processing equipment. Specifically, it utilizes newer generation information processing equipment to reuse knowledge from previous generations and efficiently adapt to new requirements.

[0260] First, the server selects and loads older generation information processing equipment from digital data storage devices. This involves the specific step of placing the parameters and weights required for the older generation model into memory within the computer system, typically using a fast, accessible memory device such as an SSD.

[0261] Next, the server connects the new generation of information processing equipment with the older generation. Here, a data communication channel is established via an API or software developed as needed, enabling the exchange of information. This creates an environment where the new generation model can leverage the valuable knowledge held by the older generation.

[0262] Furthermore, the server prepares a collection of information relevant to a specific industry and provides it to the next generation of information processing devices. This collection of information, for example in the medical field, includes the latest research materials and clinical data. In this way, the next generation of models can learn based on the latest information in their specific field.

[0263] During the automatic tuning phase, the server uses its self-optimization function to monitor the readjustment process. This allows for variable adjustment of the learning parameters, maximizing processing efficiency.

[0264] Finally, users utilize the retrained, specialized models to solve problems and perform tasks in specific business areas. For example, when building a diagnostic support system in a medical setting, this model is extremely useful in providing more accurate diagnostic results.

[0265] As a concrete example, by prompting the AI ​​generator with "Please suggest ways to improve the diabetes diagnostic model based on the latest medical papers," the user can obtain information that meets their specific requirements.

[0266] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0267] Step 1:

[0268] The server selects and loads an older generation information processing unit from a digital data storage device. It receives an older generation model file stored in storage as input. This file is loaded into memory, and the model's parameters and weights are expanded to make the model processable. Specifically, memory management software is used to place the data into high-speed memory. The output is the model expanded in memory.

[0269] Step 2:

[0270] The server links the new generation information processing unit with the old generation model. It receives interface information from both the old and new generation models as input. It establishes a data communication channel, enabling data exchange between the two models via APIs or custom programs. Specifically, it uses communication protocol setup software to establish the link. The output is the link state, indicating that information exchange is now possible.

[0271] Step 3:

[0272] The server collects data related to a specific industry sector and provides it to a next-generation model. As input, it extracts specialized datasets and performs data cleaning. The data is then converted into a format that the model can learn from using preprocessing software. Specifically, it uses data format conversion utilities to reshape the data. The output is a trainable dataset.

[0273] Step 4:

[0274] The server utilizes a new generation model and readjusts it based on the output of the previous generation model. The inputs used are the training dataset and the output data from the previous generation model. The new generation model updates its training parameters based on these inputs, learning domain-specific knowledge. Specifically, it executes machine learning algorithms and adjusts the parameters. The output is the adaptively trained new generation model.

[0275] Step 5:

[0276] The server monitors the learning state using its self-optimization function and optimizes the training process. It takes current learning state data and prediction accuracy information as input and makes real-time adjustments using dynamic learning management software. Specifically, it adjusts the learning rate and batch size using an optimization algorithm. The output is a new generation model with optimal performance.

[0277] Step 6:

[0278] Users utilize specialized models to perform tasks in specific business areas. The input consists of prompts for the generating AI model, such as, "Please suggest ways to improve the diabetes diagnostic model based on the latest medical papers." The model calculates based on these prompts and provides the user with improvements and related information as output. The results are displayed through a user interface. The output consists of specific information and suggestions tailored to the user's requests.

[0279] (Application Example 1)

[0280] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0281] The challenge lies in providing efficient methods that effectively utilize existing information processing models to enable data analysis and real-time problem detection in specific domains. In particular, in manufacturing operations, there is a need to leverage past data patterns to improve quality control and optimize production processes.

[0282] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0283] In this invention, the server includes means for selecting and loading an older generation information processing model linked to a new generation information processing model in order to reuse an existing data processing model; means for the new generation information processing model to readjust the older generation information processing model using a data set related to a specific domain; means for monitoring the readjustment process using an automatic adjustment function and terminating the learning process at the optimal timing; and means for acquiring information from measuring devices and image acquisition devices, comparing it with past data patterns derived from the older generation model to detect problems, and suggesting improvement measures. This enables increased efficiency and improved quality in the manufacturing process.

[0284] The "existing data processing model" is an information analysis system developed in the past and used to analyze specific data.

[0285] The "new-generation information processing model" is an information analysis system constructed using the latest technologies and is used to perform more advanced data processing in cooperation with the old-generation model.

[0286] The "specific area" refers to the collection of knowledge and data related to a specific industry or business, and in this invention, it particularly focuses on manufacturing operations.

[0287] The "data set" refers to the collection of information and records related to a specific area and is what the new-generation information processing model uses for learning and analysis.

[0288] The "automatic adjustment function" is a mechanism for monitoring the learning process of the information processing model in real time and determining the optimal end point of learning.

[0289] The "measurement device and image acquisition device" are hardware devices for collecting data related to the physical environment and objects and provide the input data required by the information processing model.

[0290] The "past data pattern" refers to the past information analysis results and trends accumulated by the old-generation data processing model and is what the new-generation model uses for problem detection and improvement.

[0291] The "means for detecting problems and presenting improvement measures" is a process for the information processing model to grasp specific conditions and phenomena in real time and derive appropriate solutions and guidelines.

[0292] The system that implements this invention is primarily executed by the server's functions. The server first selects an older generation information processing model to be linked to a newer generation information processing model in order to use an existing data processing model, and loads it into memory. This process involves reading model data from a storage device and deploying the model using deep learning frameworks such as TensorFlow or PyTorch.

[0293] Next, the server prepares a data set related to a specific area. This data includes historical manufacturing data and quality control standards, and is integrated with real-time data from measuring and image acquisition devices. This data set is then provided to the next-generation model using data stream processing technology that includes real-time information.

[0294] During the learning process, the server utilizes its automatic tuning function to readjust the next-generation information processing model to ensure optimal performance. This includes monitoring the model's learning state and terminating the learning process at the appropriate time. At this stage, logic is incorporated to refer to past data patterns and suggest the most suitable improvement measures for the current situation.

[0295] For example, if the model detects a frequently occurring error on an assembly line for a particular part, it will immediately analyze similar past cases and suggest effective improvement measures. Based on these suggestions, the user can make specific adjustments to improve production efficiency. This system leverages the knowledge derived by the model to optimize the entire manufacturing process.

[0296] An example of a prompt to input into a generative AI model is: "Based on last year's quality inspection data, please make suggestions to improve the product defect rate. Pay particular attention to issues that are often overlooked in optical inspections." In response to this prompt, the model will provide specific suggestions and countermeasures in real time.

[0297] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0298] Step 1:

[0299] The server selects a legacy data processing model from storage and loads it into memory. At this stage, the server reads a file containing the parameters and weights of the legacy model and deploys the model into memory using a framework such as TensorFlow or PyTorch. The input is the model file, and the output is the legacy model deployed in memory.

[0300] Step 2:

[0301] The server prepares a data set related to a specific domain. This set includes historical manufacturing data and quality control standards, and is integrated with data collected in real time from measuring and image acquisition devices. The input is raw data acquired from databases and sensors, and the output is a data set ready for analysis.

[0302] Step 3:

[0303] The server links the old-generation model with the new-generation information processing model. This means setting up a data communication channel and creating a state where data can be exchanged between the two models. The input is the old-generation model and the new-generation model, and the output is the communicative, linked state.

[0304] Step 4:

[0305] The server trains a new generation information processing model using a prepared data set. During training, the new generation model self-adjusts by referencing data patterns obtained from the old generation model. The input is a data set in a specific domain, and the output is the adjusted new generation model.

[0306] Step 5:

[0307] The server continues to monitor and readjust using an automatic adjustment function so that the model exhibits optimal performance. Here, performance metrics are checked in real time, and parameters are adjusted as necessary. The input is the model state during learning, and the output is the optimized model state.

[0308] Step 6:

[0309] The user receives the improvement measures presented by the server and utilizes them for optimizing the production process. The user creates a specific action plan based on the presented information and applies it on the production line. The input is the proposal from the new generation model, and the output is the specific action plan.

[0310] Furthermore, an emotion engine for estimating the user's emotion may be combined. That is, the specific processing unit 290 may estimate the user's emotion using the emotion recognition model 59 and perform specific processing using the user's emotion.

[0311] This invention relates to a system that links an old-generation large language model to a new-generation model and further combines an emotion engine for recognizing the user's emotion.

[0312] First, the server selects an old-generation large language model and loads it from storage to memory to prepare for reuse. By this step, the old-generation model is prepared as a pre-stage for re-education.

[0313] Next, the server links a new-generation large language model to the old-generation model via a data communication channel. This makes the findings derived from the old-generation model available for the new-generation model.

[0314] After that, the server obtains a dataset related to a specific field from the database and provides it to the new-generation model. For example, in the medical field, the latest research data on diseases and patient information are used as the dataset.

[0315] Next, the emotion engine collects user emotion data through the device. This emotion data is obtained from user interactions and interpreted using natural language processing. As a result, the next generation of models receive emotion information as additional adaptive feedback, improving the accuracy of their learning.

[0316] During the retuning process, the server provides the new generation model with specialized datasets and sentiment data, allowing the model to be readjusted to deepen its understanding of specific domains. The auto-tuning function monitors the entire process and efficiently adjusts parameters.

[0317] Finally, users utilize retrained, specialized models to respond to specific domain needs and tasks. At this stage, the addition of sentiment-based analysis provides more personalized responses and support to the user.

[0318] As a concrete example, in the field of education, we can consider a system that allows teachers to evaluate students' participation based on their reactions and use that information to improve individualized instruction. This system uses an emotion engine to analyze students' responses in real time and adjust the educational content using that information, thereby enabling the delivery of more effective education.

[0319] The following describes the processing flow.

[0320] Step 1:

[0321] The server selects a legacy large-scale language model and loads it from storage into memory. This makes the data and parameters of the legacy model available.

[0322] Step 2:

[0323] The server establishes a data communication channel to link the new generation of large-scale language models with the older generation models. This link allows the new generation models to leverage the knowledge of the older generation models.

[0324] Step 3:

[0325] The server retrieves datasets related to a specific field from a database and supplies them to the next-generation model. For example, it prepares specific medical data to enable the model to learn field-specific knowledge.

[0326] Step 4:

[0327] The emotion engine built into the device collects emotional data through interaction with the user. This includes voice tone and facial recognition data, which are then analyzed through natural language processing.

[0328] Step 5:

[0329] The server provides the next-generation model with domain-specific datasets and sentiment data collected from users, and performs retuning. The sentiment data is used as learning feedback and contributes to the model's refinement.

[0330] Step 6:

[0331] The server monitors the entire learning process using an auto-tuning function. It evaluates the model's learning effectiveness in real time and optimizes parameters as needed. As a result, effective model readjustment is achieved.

[0332] Step 7:

[0333] Users can utilize retrained, specialized models to obtain responses to specific challenges and needs in particular fields. For example, it becomes possible to provide personalized medical advice or educational content based on the user's emotions.

[0334] (Example 2)

[0335] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0336] To meet the evolving needs of information processing models and diverse application fields, there is a demand for personalized responses that reflect user sentiment while leveraging the knowledge gained from older models. Furthermore, the readjustment of information processing models for efficient operation in specific application fields remains a challenge.

[0337] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0338] In this invention, the server includes means for selecting and loading an older generation information processing model into memory; means for linking a new information processing model to the older generation information processing model using a communication path; means for acquiring a set of knowledge related to a specific application field and providing it to the new information processing model; means for collecting user emotional information through a terminal and interpreting it using natural language processing; means for incorporating the emotional information into the new information processing model and readjusting it; and means for monitoring the readjustment process and efficiently adjusting it by utilizing an automatic adjustment function. This makes it possible to provide personalized responses in a specific application field.

[0339] An "older generation information processing model" is a collection of information processing algorithms developed in previous generations, and it possesses more fundamental knowledge and data processing processes compared to models based on new technologies.

[0340] A "new information processing model" is an information processing system that employs the latest technologies and integrates knowledge from previous generation models to possess the ability to perform advanced data analysis and inference suitable for the application field.

[0341] A "communication path" is a physical or logical connection means for transmitting data and insights between information processing models, enabling the sending and receiving of information.

[0342] A "knowledge set" is a collection of data and information relating to a specific application field, including the latest case studies and research results related to that application field.

[0343] "Automatic adjustment function" is a technology that monitors information processing processes in real time and autonomously makes adjustments necessary to maintain optimal performance.

[0344] "User emotional information" refers to data about emotional responses and states obtained through interaction with the user, and is interpreted using natural language processing.

[0345] "Personalized responses" refer to information provision and actions tailored to the individual user's needs and circumstances, enabling effective support in specific application areas.

[0346] This invention is a system that provides personalized responses by efficiently combining older and newer information processing models, and further utilizing user emotional information.

[0347] First, the server selects an older generation information processing model and loads it from storage into memory. This prepares the older generation model for linking with the newer information processing model. Next, the server links the newer information processing model with the older generation model using a data communication path. This link allows the newer model to leverage the insights from the older generation model. The hardware used includes a database server and a communication network. The software employs machine learning algorithms and natural language processing tools.

[0348] The server also retrieves knowledge sets relevant to specific application areas from the database and provides them to a new information processing model. This is a step for the new model to learn information specific to the application area. For example, in the medical field, the latest research data and patient health information are used as the knowledge set.

[0349] The device plays a role in collecting user emotional information. This involves analyzing the user's facial expressions and voice using the camera and microphone, and interpreting the emotions through natural language processing. This process provides data that helps understand the user's current situation and needs, contributing to improving the learning accuracy of new information processing models.

[0350] Users receive personalized responses through the system. For example, in the field of education, teachers can analyze student responses and optimize learning plans based on that data. Examples of specific prompts include, "Please tell me about recent advances in medical technology," or "Please suggest ways to improve students' learning progress."

[0351] In this way, combining emotional information with older and newer information processing models enables effective and flexible responses in specific application areas.

[0352] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0353] Step 1:

[0354] The server selects an older generation information processing model from storage and loads it into memory. It uses the identifier and associated file path of the older generation model as input and outputs the model data loaded into memory. Specifically, the server executes the process of reading the model file from storage and placing it into memory.

[0355] Step 2:

[0356] The server links the new information processing model to the old generation information processing model using a data communication path. It takes specific information about the old and new generation models as input and outputs a state where the communication path between the models has been established. Specifically, the server connects to both models via an API and sets up a state where data can be exchanged.

[0357] Step 3:

[0358] The server retrieves a set of knowledge relevant to a specific application domain from a database and provides it to a new information processing model. It uses application domain query data as input and provides the model with appropriate knowledge information as output. Specifically, the server executes database queries and passes the extracted data to the model's learning system.

[0359] Step 4:

[0360] The device collects user emotional information. It uses signal data of the user's facial expressions and voice, acquired through the camera and microphone, as input, and obtains analyzed emotional data as output. Specifically, this involves an emotion analysis engine built into the device processing the signal data, identifying emotional patterns, and transmitting them as numerical data.

[0361] Step 5:

[0362] The server incorporates emotional data into the new information processing model and readjusts it. It uses user emotional data and a specialized knowledge set as input and designs the adjusted model parameters as output. Specifically, the server generates a feedback loop using the emotional data to optimize the model so that it outputs a response that is more appropriate for the application field.

[0363] Step 6:

[0364] The user receives personalized responses from the system. The user sends prompts as input and receives optimized response data as output. Specifically, the process involves the user communicating questions and requests to the system via a terminal, and the model returning information tailored to those requests.

[0365] (Application Example 2)

[0366] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0367] In modern information processing technology, providing personalized information that responds to user emotions is a crucial challenge. However, conventional technologies have struggled to accurately analyze user emotions and recommend optimized information based on that data. Furthermore, there is a need for a solution that effectively links and reuses both old and new information processing technologies.

[0368] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0369] In this invention, the server includes means for selecting and loading older generation information processing technologies linked to new generation information processing technologies in order to reuse existing information processing technologies; means for the new generation information processing technologies to readjust the older generation information processing technologies using a set of information related to a specific field; and means for generating personalized information using sentiment analysis technology. This enables precise information recommendations based on the user's emotions.

[0370] "Information processing technology" refers to the technology of processing, analyzing, transforming, or optimizing digital data to serve a specific purpose.

[0371] An "information set" is a collection of data and information related to a specific field, which is then analyzed and processed.

[0372] "Emotional analysis technology" is a technology that analyzes users' emotions and sensitivities and optimizes data processing and service provision based on the results.

[0373] "Communication technology" refers to technologies for transmitting, receiving, and exchanging data and information, and enables the coordination of different technologies and systems.

[0374] "Personalized information" refers to information customized based on the individual user's characteristics and preferences, aiming to provide more appropriate and efficient information.

[0375] The system of the present invention links older and newer information processing technologies to provide personalized information based on user sentiment data. The server first selects and loads older information processing technologies to reuse existing ones. Next, the newer information processing technologies readjust the older technologies using a set of information relevant to a specific field.

[0376] This process uses emotion analysis technology to acquire user emotion data and optimize the service. Users can provide their emotional information to the system through devices such as smartphones and smart glasses, and the server processes this data to provide personalized information.

[0377] Specifically, the server monitors the readjustment process using an auto-tuning function, completing the learning process at the optimal time. In this process, older and newer information processing technologies are effectively coordinated using communication technology. As a result, precise information recommendations based on user emotions are realized.

[0378] For example, an entertainment information application can recommend movies that match the user's current mood when they choose a film. If the user is feeling sad, the system will recommend a more cheerful comedy film, providing a service that is considerate of the user's feelings.

[0379] An example of a prompt message might be: "Recommend a movie using the emotions the user is currently feeling. If the user is sad, choose a comedy movie to cheer them up."

[0380] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0381] Step 1:

[0382] The server selects older generation information processing technologies and loads them from storage into memory. The input here is the identification information of the older generation technology, and the output is an instance of the loaded technology. This loading operation prepares the older generation technology for reuse.

[0383] Step 2:

[0384] The server uses communication technology to link new-generation information processing technology with older-generation technology. The input to this step is information from both the old and new generations of technology, and the output is the established link. This link allows the new generation technology to utilize the knowledge of the old generation.

[0385] Step 3:

[0386] The server retrieves a set of information related to a specific field from a database. The input is a request related to that field, and the output is the retrieved information set. The server uses this information to prepare for the re-engineering of next-generation technologies.

[0387] Step 4:

[0388] The device collects user emotional information using emotion analysis technology and sends it to a server. The input is data of the user's facial expressions and voice, and the output is the analyzed emotional information. This provides data based on the user's emotions.

[0389] Step 5:

[0390] The server uses emotional information and data sets to retune next-generation technologies. The input is emotional information and data sets, and the output is the parameters of the updated technologies. This allows the system to provide users with more relevant information.

[0391] Step 6:

[0392] The server monitors the readjustment process using an auto-tuning function and completes it at the optimal time. The input is the readjustment progress data, and the output is the readjustment completion time. This enables an efficient learning process.

[0393] Step 7:

[0394] The server uses tuned information processing technology to provide users with personalized information. The input consists of the user's request and the tuned technology, while the output is information optimized for the user. As a result, users can receive information that resonates with their own emotions.

[0395] 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 user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0396] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0397] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0398] [Third Embodiment]

[0399] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0400] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0401] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0402] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0403] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0404] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0405] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0406] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0407] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0408] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0409] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0410] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0411] This invention relates to a system that provides a method for making older, large-scale language models reusable in specific fields.

[0412] First, the server loads a large-scale, older-generation language model into memory. This is done by selecting it from storage and preparing the necessary parameters and weights.

[0413] Next, the server links the new generation large-scale language model with the old generation model. This enables data communication between the two models, creating a state where the new generation model can effectively utilize the insights and data of the old generation model.

[0414] Subsequently, the server prepares a dataset for a specific field and provides it to the next-generation model. For example, in the medical field, it would select the latest medical papers and diagnostic data as the dataset. This prepares the next-generation model to effectively learn the information required in that specific field.

[0415] During the retuning phase, the new generation of large-scale language models is retrained using specific domain datasets, referencing the output of the previous generation models. This improves the model's performance as a specialized model for that particular domain.

[0416] By utilizing the auto-tuning function, the server monitors the model's learning state in real time and dynamically adjusts training parameters as needed. This mechanism reduces wasted resources and enables effective reuse.

[0417] Finally, users utilize these retrained, specialized models to apply them to specific domain tasks and problem-solving. This allows them to maximize the knowledge gained from older models while enabling advanced applications made possible by the latest technologies.

[0418] For example, when a user builds a diagnostic support system for a medical setting, using the specialized model obtained through this process makes it possible to achieve more efficient and accurate diagnostic support.

[0419] The following describes the processing flow.

[0420] Step 1:

[0421] The server selects a large-scale language model from the previous generation and loads it from storage into memory. This completes the restoration of the previous generation model, making it ready for retraining.

[0422] Step 2:

[0423] The server links the new generation of large-scale language models with the old generation models. It establishes a data communication channel and configures it so that the output of the old generation model is taken in as input to the new generation model.

[0424] Step 3:

[0425] The server retrieves datasets related to a specific field from the database and provides them to the next-generation model. These datasets contain knowledge in a specific field and form the basis for learning.

[0426] Step 4:

[0427] A new generation of large-scale language models retunes itself using specific domain datasets, referencing the output of older models. This enhances the model's capabilities as a domain-specific model.

[0428] Step 5:

[0429] The server uses an auto-tuning function to monitor the learning process in real time. It adjusts the learning rate and parameters as needed to achieve optimal training results.

[0430] Step 6:

[0431] Users utilize retrained, specialized models to apply them to tasks in specific fields. For example, a model specialized in diagnostic support in the medical field can leverage its capabilities to support users' work.

[0432] (Example 1)

[0433] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0434] When reusing older generation information processing equipment in specific industrial fields, the challenge lies in effectively utilizing that knowledge with newer equipment while minimizing wasteful resource consumption and achieving rapid and advanced adaptation. In particular, there is a problem in efficiently exchanging information between older and newer generation information processing equipment.

[0435] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0436] In this invention, the server includes means for selecting and loading older generation information processing devices as information formats from a digital data storage device; means for connecting a new generation information processing device with an older generation information processing device to enable information exchange; means for the new generation information processing device to readjust the older generation information processing device using a collection of information related to a specific industrial field; and means for monitoring the readjustment procedure using a self-optimization function and adjusting the learning procedure in a variable manner. This makes it possible to adapt the knowledge of the older generation information processing device to modern requirements and enable efficient information processing.

[0437] A "digital data storage device" is a recording device used to store information for long periods of time, and it serves as the basis for selecting and loading older generation information processing devices in a computer system.

[0438] "Information format" refers to the structure of data and programs, and is the basic unit by which computer systems read and process information.

[0439] An "information processing device" is a general term for hardware and software that receives digital data as input and converts it into output through calculations, analysis, and other processes.

[0440] "Means of selection and loading" refers to the processes and techniques for selecting a specific information processing device from storage and loading it into memory.

[0441] "Integration" refers to the process of sharing and coordinating data and functions among multiple information processing devices.

[0442] "Information exchange" refers to the process of sending and receiving data between different information processing devices and sharing knowledge and functions.

[0443] An "information collection" refers to a set of data, texts, and other materials related to a specific industrial field, and is used to acquire new knowledge.

[0444] "Self-optimization function" refers to a mechanism by which an information processing device adjusts its own parameters and configuration to maximize efficiency.

[0445] A "readjustment procedure" is the process of adjusting the parameters and operation of an information processing device in response to new conditions or information.

[0446] A "variable method" refers to procedures or strategies that can be flexibly changed depending on the situation or conditions.

[0447] This invention provides a method for realizing information processing in a new specific business field using older generation information processing equipment. Specifically, it utilizes newer generation information processing equipment to reuse knowledge from previous generations and efficiently adapt to new requirements.

[0448] First, the server selects and loads older generation information processing equipment from digital data storage devices. This involves the specific step of placing the parameters and weights required for the older generation model into memory within the computer system, typically using a fast, accessible memory device such as an SSD.

[0449] Next, the server connects the new generation of information processing equipment with the older generation. Here, a data communication channel is established via an API or software developed as needed, enabling the exchange of information. This creates an environment where the new generation model can leverage the valuable knowledge held by the older generation.

[0450] Furthermore, the server prepares a collection of information relevant to a specific industry and provides it to the next generation of information processing devices. This collection of information, for example in the medical field, includes the latest research materials and clinical data. In this way, the next generation of models can learn based on the latest information in their specific field.

[0451] During the automatic tuning phase, the server uses its self-optimization function to monitor the readjustment process. This allows for variable adjustment of the learning parameters, maximizing processing efficiency.

[0452] Finally, users utilize the retrained, specialized models to solve problems and perform tasks in specific business areas. For example, when building a diagnostic support system in a medical setting, this model is extremely useful in providing more accurate diagnostic results.

[0453] As a concrete example, by prompting the AI ​​generator with "Please suggest ways to improve the diabetes diagnostic model based on the latest medical papers," the user can obtain information that meets their specific requirements.

[0454] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0455] Step 1:

[0456] The server selects and loads an older generation information processing unit from a digital data storage device. It receives an older generation model file stored in storage as input. This file is loaded into memory, and the model's parameters and weights are expanded to make the model processable. Specifically, memory management software is used to place the data into high-speed memory. The output is the model expanded in memory.

[0457] Step 2:

[0458] The server links the new generation information processing unit with the old generation model. It receives interface information from both the old and new generation models as input. It establishes a data communication channel, enabling data exchange between the two models via APIs or custom programs. Specifically, it uses communication protocol setup software to establish the link. The output is the link state, indicating that information exchange is now possible.

[0459] Step 3:

[0460] The server collects data related to a specific industry sector and provides it to a next-generation model. As input, it extracts specialized datasets and performs data cleaning. The data is then converted into a format that the model can learn from using preprocessing software. Specifically, it uses data format conversion utilities to reshape the data. The output is a trainable dataset.

[0461] Step 4:

[0462] The server utilizes a new generation model and readjusts it based on the output of the previous generation model. The inputs used are the training dataset and the output data from the previous generation model. The new generation model updates its training parameters based on these inputs, learning domain-specific knowledge. Specifically, it executes machine learning algorithms and adjusts the parameters. The output is the adaptively trained new generation model.

[0463] Step 5:

[0464] The server monitors the learning state using its self-optimization function and optimizes the training process. It takes current learning state data and prediction accuracy information as input and makes real-time adjustments using dynamic learning management software. Specifically, it adjusts the learning rate and batch size using an optimization algorithm. The output is a new generation model with optimal performance.

[0465] Step 6:

[0466] Users utilize specialized models to perform tasks in specific business areas. The input consists of prompts for the generating AI model, such as, "Please suggest ways to improve the diabetes diagnostic model based on the latest medical papers." The model calculates based on these prompts and provides the user with improvements and related information as output. The results are displayed through a user interface. The output consists of specific information and suggestions tailored to the user's requests.

[0467] (Application Example 1)

[0468] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0469] The challenge lies in providing efficient methods that effectively utilize existing information processing models to enable data analysis and real-time problem detection in specific domains. In particular, in manufacturing operations, there is a need to leverage past data patterns to improve quality control and optimize production processes.

[0470] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0471] In this invention, the server includes means for selecting and loading an older generation information processing model linked to a new generation information processing model in order to reuse an existing data processing model; means for the new generation information processing model to readjust the older generation information processing model using a data set related to a specific domain; means for monitoring the readjustment process using an automatic adjustment function and terminating the learning process at the optimal timing; and means for acquiring information from measuring devices and image acquisition devices, comparing it with past data patterns derived from the older generation model to detect problems, and suggesting improvement measures. This enables increased efficiency and improved quality in the manufacturing process.

[0472] An "existing data processing model" refers to an information analysis system developed in the past that is used to analyze specific data.

[0473] A "next-generation information processing model" is an information analysis system built using the latest technology, which works in conjunction with older generation models to perform more advanced data processing.

[0474] A "specific domain" refers to a collection of knowledge and data related to a particular industry or business, and in this invention, the subject is particularly manufacturing operations.

[0475] A "data set" refers to a collection of information or records related to a specific domain, and is used by next-generation information processing models for learning and analysis.

[0476] The "automatic adjustment function" is a mechanism that monitors the learning process of an information processing model in real time and determines the optimal point at which learning should end.

[0477] "Measurement devices and image acquisition devices" are hardware devices used to collect data about the physical environment and objects, and they provide the input data necessary for information processing models.

[0478] "Past data patterns" refer to the results of past information analysis and trends accumulated by older generation data processing models, which newer generation models utilize for problem detection and improvement.

[0479] "A means of detecting problems and suggesting solutions" refers to the process by which an information processing model grasps specific conditions or phenomena in real time and derives appropriate solutions or guidelines.

[0480] The system that implements this invention is primarily executed by the server's functions. The server first selects an older generation information processing model to be linked to a newer generation information processing model in order to use an existing data processing model, and loads it into memory. This process involves reading model data from a storage device and deploying the model using deep learning frameworks such as TensorFlow or PyTorch.

[0481] Next, the server prepares a data set related to a specific area. This data includes historical manufacturing data and quality control standards, and is integrated with real-time data from measuring and image acquisition devices. This data set is then provided to the next-generation model using data stream processing technology that includes real-time information.

[0482] During the learning process, the server utilizes its automatic tuning function to readjust the next-generation information processing model to ensure optimal performance. This includes monitoring the model's learning state and terminating the learning process at the appropriate time. At this stage, logic is incorporated to refer to past data patterns and suggest the most suitable improvement measures for the current situation.

[0483] For example, if the model detects a frequently occurring error on an assembly line for a particular part, it will immediately analyze similar past cases and suggest effective improvement measures. Based on these suggestions, the user can make specific adjustments to improve production efficiency. This system leverages the knowledge derived by the model to optimize the entire manufacturing process.

[0484] An example of a prompt to input into a generative AI model is: "Based on last year's quality inspection data, please make suggestions to improve the product defect rate. Pay particular attention to issues that are often overlooked in optical inspections." In response to this prompt, the model will provide specific suggestions and countermeasures in real time.

[0485] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0486] Step 1:

[0487] The server selects a legacy data processing model from storage and loads it into memory. At this stage, the server reads a file containing the parameters and weights of the legacy model and deploys the model into memory using a framework such as TensorFlow or PyTorch. The input is the model file, and the output is the legacy model deployed in memory.

[0488] Step 2:

[0489] The server prepares a data set related to a specific domain. This set includes historical manufacturing data and quality control standards, and is integrated with data collected in real time from measuring and image acquisition devices. The input is raw data acquired from databases and sensors, and the output is a data set ready for analysis.

[0490] Step 3:

[0491] The server links the old-generation model with the new-generation information processing model. This means setting up a data communication channel and creating a state where data can be exchanged between the two models. The input is the old-generation model and the new-generation model, and the output is the communicative, linked state.

[0492] Step 4:

[0493] The server trains a new generation information processing model using a prepared data set. During training, the new generation model self-adjusts by referencing data patterns obtained from the old generation model. The input is a data set in a specific domain, and the output is the adjusted new generation model.

[0494] Step 5:

[0495] The server utilizes an auto-tuning function to continuously monitor and readjust the model to ensure optimal performance. Here, performance metrics are checked in real time, and parameters are adjusted as needed. The input is the model state during training, and the output is the optimized model state.

[0496] Step 6:

[0497] Users receive improvement suggestions from the server and use them to optimize their production processes. Based on the information provided, users create concrete action plans and apply them to the manufacturing line. The input is suggestions from the new generation model, and the output is a concrete action plan.

[0498] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0499] This invention relates to a system that links older generation large-scale language models with newer generation models, and further combines them with an emotion engine that recognizes user emotions.

[0500] First, the server selects a large-scale language model from the older generation and loads it from storage into memory to prepare it for reuse. This step prepares the older model as a preliminary step before retraining.

[0501] Next, the server links the new generation large-scale language model with the old generation model via a data communication channel. This makes the insights derived from the old generation model available to the new generation model.

[0502] Subsequently, the server retrieves datasets related to a specific field from the database and provides them to the next-generation model. For example, in the medical field, the latest research data and patient information on diseases would be used as datasets.

[0503] Next, the emotion engine collects user emotion data through the device. This emotion data is obtained from user interactions and interpreted using natural language processing. As a result, the next generation of models receive emotion information as additional adaptive feedback, improving the accuracy of their learning.

[0504] During the retuning process, the server provides the new generation model with specialized datasets and sentiment data, allowing the model to be readjusted to deepen its understanding of specific domains. The auto-tuning function monitors the entire process and efficiently adjusts parameters.

[0505] Finally, users utilize retrained, specialized models to respond to specific domain needs and tasks. At this stage, the addition of sentiment-based analysis provides more personalized responses and support to the user.

[0506] As a concrete example, in the field of education, we can consider a system that allows teachers to evaluate students' participation based on their reactions and use that information to improve individualized instruction. This system uses an emotion engine to analyze students' responses in real time and adjust the educational content using that information, thereby enabling the delivery of more effective education.

[0507] The following describes the processing flow.

[0508] Step 1:

[0509] The server selects a large-scale language model from the previous generation and loads it from storage into memory. This makes the data and parameters of the older model available.

[0510] Step 2:

[0511] The server establishes a data communication channel to link the new generation of large-scale language models with the older generation models. This link allows the new generation models to leverage the knowledge of the older generation models.

[0512] Step 3:

[0513] The server retrieves datasets related to a specific field from a database and supplies them to the next-generation model. For example, it prepares specific medical data to enable the model to learn field-specific knowledge.

[0514] Step 4:

[0515] The emotion engine built into the device collects emotional data through interaction with the user. This includes voice tone and facial recognition data, which are then analyzed through natural language processing.

[0516] Step 5:

[0517] The server provides the next-generation model with domain-specific datasets and sentiment data collected from users, and performs retuning. The sentiment data is used as learning feedback and contributes to the model's refinement.

[0518] Step 6:

[0519] The server monitors the entire learning process using an auto-tuning function. It evaluates the model's learning effectiveness in real time and optimizes parameters as needed. As a result, effective model readjustment is achieved.

[0520] Step 7:

[0521] Users can utilize retrained, specialized models to obtain responses to specific challenges and needs in particular fields. For example, it becomes possible to provide personalized medical advice or educational content based on the user's emotions.

[0522] (Example 2)

[0523] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0524] To meet the evolving needs of information processing models and diverse application fields, there is a demand for personalized responses that reflect user sentiment while leveraging the knowledge gained from older models. Furthermore, the readjustment of information processing models for efficient operation in specific application fields remains a challenge.

[0525] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0526] In this invention, the server includes means for selecting and loading an older generation information processing model into memory; means for linking a new information processing model to the older generation information processing model using a communication path; means for acquiring a set of knowledge related to a specific application field and providing it to the new information processing model; means for collecting user emotional information through a terminal and interpreting it using natural language processing; means for incorporating the emotional information into the new information processing model and readjusting it; and means for monitoring the readjustment process and efficiently adjusting it by utilizing an automatic adjustment function. This makes it possible to provide personalized responses in a specific application field.

[0527] An "older generation information processing model" is a collection of information processing algorithms developed in previous generations, and it possesses more fundamental knowledge and data processing processes compared to models based on new technologies.

[0528] A "new information processing model" is an information processing system that employs the latest technologies and integrates knowledge from previous generation models to possess the ability to perform advanced data analysis and inference suitable for the application field.

[0529] A "communication path" is a physical or logical connection means for transmitting data and insights between information processing models, enabling the sending and receiving of information.

[0530] A "knowledge set" is a collection of data and information relating to a specific application field, including the latest case studies and research results related to that application field.

[0531] "Automatic adjustment function" is a technology that monitors information processing processes in real time and autonomously makes adjustments necessary to maintain optimal performance.

[0532] "User emotional information" refers to data about emotional responses and states obtained through interaction with the user, and is interpreted using natural language processing.

[0533] "Personalized responses" refer to information provision and actions tailored to the individual user's needs and circumstances, enabling effective support in specific application areas.

[0534] This invention is a system that provides personalized responses by efficiently combining older and newer information processing models, and further utilizing user emotional information.

[0535] First, the server selects an older generation information processing model and loads it from storage into memory. This prepares the older generation model for linking with the newer information processing model. Next, the server links the newer information processing model with the older generation model using a data communication path. This link allows the newer model to leverage the insights from the older generation model. The hardware used includes a database server and a communication network. The software employs machine learning algorithms and natural language processing tools.

[0536] The server also retrieves knowledge sets relevant to specific application areas from the database and provides them to a new information processing model. This is a step for the new model to learn information specific to the application area. For example, in the medical field, the latest research data and patient health information are used as the knowledge set.

[0537] The device plays a role in collecting user emotional information. This involves analyzing the user's facial expressions and voice using the camera and microphone, and interpreting the emotions through natural language processing. This process provides data that helps understand the user's current situation and needs, contributing to improving the learning accuracy of new information processing models.

[0538] Users receive personalized responses through the system. For example, in the field of education, teachers can analyze student responses and optimize learning plans based on that data. Examples of specific prompts include, "Please tell me about recent advances in medical technology," or "Please suggest ways to improve students' learning progress."

[0539] In this way, combining emotional information with older and newer information processing models enables effective and flexible responses in specific application areas.

[0540] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0541] Step 1:

[0542] The server selects an older generation information processing model from storage and loads it into memory. It uses the identifier and associated file path of the older generation model as input and outputs the model data loaded into memory. Specifically, the server executes the process of reading the model file from storage and placing it into memory.

[0543] Step 2:

[0544] The server links the new information processing model to the old generation information processing model using a data communication path. It takes specific information about the old and new generation models as input and outputs a state where the communication path between the models has been established. Specifically, the server connects to both models via an API and sets up a state where data can be exchanged.

[0545] Step 3:

[0546] The server retrieves a set of knowledge relevant to a specific application domain from a database and provides it to a new information processing model. It uses application domain query data as input and provides the model with appropriate knowledge information as output. Specifically, the server executes database queries and passes the extracted data to the model's learning system.

[0547] Step 4:

[0548] The device collects user emotional information. It uses signal data of the user's facial expressions and voice, acquired through the camera and microphone, as input, and obtains analyzed emotional data as output. Specifically, this involves an emotion analysis engine built into the device processing the signal data, identifying emotional patterns, and transmitting them as numerical data.

[0549] Step 5:

[0550] The server incorporates emotional data into the new information processing model and readjusts it. It uses user emotional data and a specialized knowledge set as input and designs the adjusted model parameters as output. Specifically, the server generates a feedback loop using the emotional data to optimize the model so that it outputs a response that is more appropriate for the application field.

[0551] Step 6:

[0552] The user receives personalized responses from the system. The user sends prompts as input and receives optimized response data as output. Specifically, the process involves the user communicating questions and requests to the system via a terminal, and the model returning information tailored to those requests.

[0553] (Application Example 2)

[0554] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0555] In modern information processing technology, providing personalized information that responds to user emotions is a crucial challenge. However, conventional technologies have struggled to accurately analyze user emotions and recommend optimized information based on that data. Furthermore, there is a need for a solution that effectively links and reuses both old and new information processing technologies.

[0556] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0557] In this invention, the server includes means for selecting and loading older generation information processing technologies linked to new generation information processing technologies in order to reuse existing information processing technologies; means for the new generation information processing technologies to readjust the older generation information processing technologies using a set of information related to a specific field; and means for generating personalized information using sentiment analysis technology. This enables precise information recommendations based on the user's emotions.

[0558] "Information processing technology" refers to the technology of processing, analyzing, transforming, or optimizing digital data to serve a specific purpose.

[0559] An "information set" is a collection of data and information related to a specific field, which is then analyzed and processed.

[0560] "Emotional analysis technology" is a technology that analyzes users' emotions and sensitivities and optimizes data processing and service provision based on the results.

[0561] "Communication technology" refers to technologies for transmitting, receiving, and exchanging data and information, and enables the coordination of different technologies and systems.

[0562] "Personalized information" refers to information customized based on the individual user's characteristics and preferences, aiming to provide more appropriate and efficient information.

[0563] The system of the present invention links older and newer information processing technologies to provide personalized information based on user sentiment data. The server first selects and loads older information processing technologies to reuse existing ones. Next, the newer information processing technologies readjust the older technologies using a set of information relevant to a specific field.

[0564] This process uses emotion analysis technology to acquire user emotion data and optimize the service. Users can provide their emotional information to the system through devices such as smartphones and smart glasses, and the server processes this data to provide personalized information.

[0565] Specifically, the server monitors the readjustment process using an auto-tuning function, completing the learning process at the optimal time. In this process, older and newer information processing technologies are effectively coordinated using communication technology. As a result, precise information recommendations based on user emotions are realized.

[0566] For example, an entertainment information application can recommend movies that match the user's current mood when they choose a film. If the user is feeling sad, the system will recommend a more cheerful comedy film, providing a service that is considerate of the user's feelings.

[0567] An example of a prompt message might be: "Recommend a movie using the emotions the user is currently feeling. If the user is sad, choose a comedy movie to cheer them up."

[0568] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0569] Step 1:

[0570] The server selects older generation information processing technologies and loads them from storage into memory. The input here is the identification information of the older generation technology, and the output is an instance of the loaded technology. This loading operation prepares the older generation technology for reuse.

[0571] Step 2:

[0572] The server uses communication technology to link new-generation information processing technology with older-generation technology. The input to this step is information from both the old and new generations of technology, and the output is the established link. This link allows the new generation technology to utilize the knowledge of the old generation.

[0573] Step 3:

[0574] The server retrieves a set of information related to a specific field from a database. The input is a request related to that field, and the output is the retrieved information set. The server uses this information to prepare for the re-engineering of next-generation technologies.

[0575] Step 4:

[0576] The device collects user emotional information using emotion analysis technology and sends it to a server. The input is data of the user's facial expressions and voice, and the output is the analyzed emotional information. This provides data based on the user's emotions.

[0577] Step 5:

[0578] The server uses emotional information and data sets to retune next-generation technologies. The input is emotional information and data sets, and the output is the parameters of the updated technologies. This allows the system to provide users with more relevant information.

[0579] Step 6:

[0580] The server monitors the readjustment process using an auto-tuning function and completes it at the optimal time. The input is the readjustment progress data, and the output is the readjustment completion time. This enables an efficient learning process.

[0581] Step 7:

[0582] The server uses tuned information processing technology to provide users with personalized information. The input consists of the user's request and the tuned technology, while the output is information optimized for the user. As a result, users can receive information that resonates with their own emotions.

[0583] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0584] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0585] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0586] [Fourth Embodiment]

[0587] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0588] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0589] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0590] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0591] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0592] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0593] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0594] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0595] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0596] The specific processing program 56 is an example of a "program" relating to the technology of this 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.

[0597] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0598] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. 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 processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0599] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0600] This invention relates to a system that provides a method for making older, large-scale language models reusable in specific fields.

[0601] First, the server loads a large-scale, older-generation language model into memory. This is done by selecting it from storage and preparing the necessary parameters and weights.

[0602] Next, the server links the new generation large-scale language model with the old generation model. This enables data communication between the two models, creating a state where the new generation model can effectively utilize the insights and data of the old generation model.

[0603] Subsequently, the server prepares a dataset for a specific field and provides it to the next-generation model. For example, in the medical field, it would select the latest medical papers and diagnostic data as the dataset. This prepares the next-generation model to effectively learn the information required in that specific field.

[0604] During the retuning phase, the new generation of large-scale language models is retrained using specific domain datasets, referencing the output of the previous generation models. This improves the model's performance as a specialized model for that particular domain.

[0605] By utilizing the auto-tuning function, the server monitors the model's learning state in real time and dynamically adjusts training parameters as needed. This mechanism reduces wasted resources and enables effective reuse.

[0606] Finally, users utilize these retrained, specialized models to apply them to specific domain tasks and problem-solving. This allows them to maximize the knowledge gained from older models while enabling advanced applications made possible by the latest technologies.

[0607] For example, when a user builds a diagnostic support system for a medical setting, using the specialized model obtained through this process makes it possible to achieve more efficient and accurate diagnostic support.

[0608] The following describes the processing flow.

[0609] Step 1:

[0610] The server selects a large-scale language model from the previous generation and loads it from storage into memory. This completes the restoration of the previous generation model, making it ready for retraining.

[0611] Step 2:

[0612] The server links the new generation of large-scale language models with the old generation models. It establishes a data communication channel and configures it so that the output of the old generation model is taken in as input to the new generation model.

[0613] Step 3:

[0614] The server retrieves datasets related to a specific field from the database and provides them to the next-generation model. These datasets contain knowledge in a specific field and form the basis for learning.

[0615] Step 4:

[0616] A new generation of large-scale language models retunes itself using specific domain datasets, referencing the output of older models. This enhances the model's capabilities as a domain-specific model.

[0617] Step 5:

[0618] The server uses an auto-tuning function to monitor the learning process in real time. It adjusts the learning rate and parameters as needed to achieve optimal training results.

[0619] Step 6:

[0620] Users utilize retrained, specialized models to apply them to tasks in specific fields. For example, a model specialized in diagnostic support in the medical field can leverage its capabilities to support users' work.

[0621] (Example 1)

[0622] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0623] When reusing older generation information processing equipment in specific industrial fields, the challenge lies in effectively utilizing that knowledge with newer equipment while minimizing wasteful resource consumption and achieving rapid and advanced adaptation. In particular, there is a problem in efficiently exchanging information between older and newer generation information processing equipment.

[0624] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0625] In this invention, the server includes means for selecting and loading older generation information processing devices as information formats from a digital data storage device; means for connecting a new generation information processing device with an older generation information processing device to enable information exchange; means for the new generation information processing device to readjust the older generation information processing device using a collection of information related to a specific industrial field; and means for monitoring the readjustment procedure using a self-optimization function and adjusting the learning procedure in a variable manner. This makes it possible to adapt the knowledge of the older generation information processing device to modern requirements and enable efficient information processing.

[0626] A "digital data storage device" is a recording device used to store information for long periods of time, and it is the basis for selecting and loading older generation information processing devices in a computer system.

[0627] "Information format" refers to the structure of data and programs, and is the basic unit by which computer systems read and process information.

[0628] An "information processing device" is a general term for hardware and software that receives digital data as input and converts it into output through calculations, analysis, and other processes.

[0629] "Means of selection and loading" refers to the processes and techniques for selecting a specific information processing device from storage and loading it into memory.

[0630] "Integration" refers to the process of sharing and coordinating data and functions among multiple information processing devices.

[0631] "Information exchange" refers to the process of sending and receiving data between different information processing devices and sharing knowledge and functions.

[0632] An "information collection" refers to a set of data, texts, and other materials related to a specific industrial field, and is used to acquire new knowledge.

[0633] "Self-optimization function" refers to a mechanism by which an information processing device adjusts its own parameters and configuration to maximize efficiency.

[0634] A "readjustment procedure" is the process of adjusting the parameters and operation of an information processing device in response to new conditions or information.

[0635] A "variable method" refers to procedures or strategies that can be flexibly changed depending on the situation or conditions.

[0636] This invention provides a method for realizing information processing in a new specific business field using older generation information processing equipment. Specifically, it utilizes newer generation information processing equipment to reuse knowledge from previous generations and efficiently adapt to new requirements.

[0637] First, the server selects and loads older generation information processing equipment from digital data storage devices. This involves the specific step of placing the parameters and weights required for the older generation model into memory within the computer system, typically using a fast, accessible memory device such as an SSD.

[0638] Next, the server connects the new generation of information processing equipment with the older generation. Here, a data communication channel is established via an API or software developed as needed, enabling the exchange of information. This creates an environment where the new generation model can leverage the valuable knowledge held by the older generation.

[0639] Furthermore, the server prepares a collection of information relevant to a specific industry and provides it to the next generation of information processing devices. This collection of information, for example in the medical field, includes the latest research materials and clinical data. In this way, the next generation of models can learn based on the latest information in their specific field.

[0640] During the automatic tuning phase, the server uses its self-optimization function to monitor the readjustment process. This allows for variable adjustment of the learning parameters, maximizing processing efficiency.

[0641] Finally, users utilize the retrained, specialized models to solve problems and perform tasks in specific business areas. For example, when building a diagnostic support system in a medical setting, this model is extremely useful in providing more accurate diagnostic results.

[0642] As a concrete example, by prompting the AI ​​generator with "Please suggest ways to improve the diabetes diagnostic model based on the latest medical papers," the user can obtain information that meets their specific requirements.

[0643] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0644] Step 1:

[0645] The server selects and loads an older generation information processing unit from a digital data storage device. It receives an older generation model file stored in storage as input. This file is loaded into memory, and the model's parameters and weights are expanded to make the model processable. Specifically, memory management software is used to place the data into high-speed memory. The output is the model expanded in memory.

[0646] Step 2:

[0647] The server links the new generation information processing unit with the old generation model. It receives interface information from both the old and new generation models as input. It establishes a data communication channel, enabling data exchange between the two models via APIs or custom programs. Specifically, it uses communication protocol setup software to establish the link. The output is the link state, indicating that information exchange is now possible.

[0648] Step 3:

[0649] The server collects data related to a specific industry sector and provides it to a next-generation model. As input, it extracts specialized datasets and performs data cleaning. The data is then converted into a format that the model can learn from using preprocessing software. Specifically, it uses data format conversion utilities to reshape the data. The output is a trainable dataset.

[0650] Step 4:

[0651] The server utilizes a new generation model and readjusts it based on the output of the previous generation model. The inputs used are the training dataset and the output data from the previous generation model. The new generation model updates its training parameters based on these inputs, learning domain-specific knowledge. Specifically, it executes machine learning algorithms and adjusts the parameters. The output is the adaptively trained new generation model.

[0652] Step 5:

[0653] The server monitors the learning state using its self-optimization function and optimizes the training process. It takes current learning state data and prediction accuracy information as input and makes real-time adjustments using dynamic learning management software. Specifically, it adjusts the learning rate and batch size using an optimization algorithm. The output is a new generation model with optimal performance.

[0654] Step 6:

[0655] Users utilize specialized models to perform tasks in specific business areas. The input consists of prompts for the generating AI model, such as, "Please suggest ways to improve the diabetes diagnostic model based on the latest medical papers." The model calculates based on these prompts and provides the user with improvements and related information as output. The results are displayed through a user interface. The output consists of specific information and suggestions tailored to the user's requests.

[0656] (Application Example 1)

[0657] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0658] The challenge lies in providing efficient methods that effectively utilize existing information processing models to enable data analysis and real-time problem detection in specific domains. In particular, in manufacturing operations, there is a need to leverage past data patterns to improve quality control and optimize production processes.

[0659] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0660] In this invention, the server includes means for selecting and loading an older generation information processing model linked to a new generation information processing model in order to reuse an existing data processing model; means for the new generation information processing model to readjust the older generation information processing model using a data set related to a specific domain; means for monitoring the readjustment process using an automatic adjustment function and terminating the learning process at the optimal timing; and means for acquiring information from measuring devices and image acquisition devices, comparing it with past data patterns derived from the older generation model to detect problems, and suggesting improvement measures. This enables increased efficiency and improved quality in the manufacturing process.

[0661] An "existing data processing model" refers to an information analysis system developed in the past that is used to analyze specific data.

[0662] A "next-generation information processing model" is an information analysis system built using the latest technology, which works in conjunction with older generation models to perform more advanced data processing.

[0663] A "specific domain" refers to a collection of knowledge and data related to a particular industry or business, and in this invention, the subject is particularly manufacturing operations.

[0664] A "data set" refers to a collection of information or records related to a specific domain, and is used by next-generation information processing models for learning and analysis.

[0665] The "automatic adjustment function" is a mechanism that monitors the learning process of an information processing model in real time and determines the optimal point at which learning should end.

[0666] "Measurement devices and image acquisition devices" are hardware devices used to collect data about the physical environment and objects, and they provide the input data necessary for information processing models.

[0667] "Past data patterns" refer to the results of past information analysis and trends accumulated by older generation data processing models, which newer generation models utilize for problem detection and improvement.

[0668] "A means of detecting problems and suggesting solutions" refers to the process by which an information processing model grasps specific conditions or phenomena in real time and derives appropriate solutions or guidelines.

[0669] The system that implements this invention is primarily executed by the server's functions. The server first selects an older generation information processing model to be linked to a newer generation information processing model in order to use an existing data processing model, and loads it into memory. This process involves reading model data from a storage device and deploying the model using deep learning frameworks such as TensorFlow or PyTorch.

[0670] Next, the server prepares a data set related to a specific area. This data includes historical manufacturing data and quality control standards, and is integrated with real-time data from measuring and image acquisition devices. This data set is then provided to the next-generation model using data stream processing technology that includes real-time information.

[0671] During the learning process, the server utilizes its automatic tuning function to readjust the next-generation information processing model to ensure optimal performance. This includes monitoring the model's learning state and terminating the learning process at the appropriate time. At this stage, logic is incorporated to refer to past data patterns and suggest the most suitable improvement measures for the current situation.

[0672] For example, if the model detects a frequently occurring error on an assembly line for a particular part, it will immediately analyze similar past cases and suggest effective improvement measures. Based on these suggestions, the user can make specific adjustments to improve production efficiency. This system leverages the knowledge derived by the model to optimize the entire manufacturing process.

[0673] An example of a prompt to input into a generative AI model is: "Based on last year's quality inspection data, please make suggestions to improve the product defect rate. Pay particular attention to issues that are often overlooked in optical inspections." In response to this prompt, the model will provide specific suggestions and countermeasures in real time.

[0674] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0675] Step 1:

[0676] The server selects a legacy data processing model from storage and loads it into memory. At this stage, the server reads a file containing the parameters and weights of the legacy model and deploys the model into memory using a framework such as TensorFlow or PyTorch. The input is the model file, and the output is the legacy model deployed in memory.

[0677] Step 2:

[0678] The server prepares a data set related to a specific domain. This set includes historical manufacturing data and quality control standards, and is integrated with data collected in real time from measuring and image acquisition devices. The input is raw data acquired from databases and sensors, and the output is a data set ready for analysis.

[0679] Step 3:

[0680] The server links the old-generation model with the new-generation information processing model. This means setting up a data communication channel and creating a state where data can be exchanged between the two models. The input is the old-generation model and the new-generation model, and the output is the communicative, linked state.

[0681] Step 4:

[0682] The server trains a new generation information processing model using a prepared data set. During training, the new generation model self-adjusts by referencing data patterns obtained from the old generation model. The input is a data set in a specific domain, and the output is the adjusted new generation model.

[0683] Step 5:

[0684] The server utilizes an auto-tuning function to continuously monitor and readjust the model to ensure optimal performance. Here, performance metrics are checked in real time, and parameters are adjusted as needed. The input is the model state during training, and the output is the optimized model state.

[0685] Step 6:

[0686] Users receive improvement suggestions from the server and use them to optimize their production processes. Based on the information provided, users create concrete action plans and apply them to the manufacturing line. The input is suggestions from the new generation model, and the output is a concrete action plan.

[0687] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0688] This invention relates to a system that links older generation large-scale language models with newer generation models, and further combines them with an emotion engine that recognizes user emotions.

[0689] First, the server selects a large-scale language model from the older generation and loads it from storage into memory to prepare it for reuse. This step prepares the older model as a preliminary step before retraining.

[0690] Next, the server links the new generation large-scale language model with the old generation model via a data communication channel. This makes the insights derived from the old generation model available to the new generation model.

[0691] Subsequently, the server retrieves datasets related to a specific field from the database and provides them to the next-generation model. For example, in the medical field, the latest research data and patient information on diseases would be used as datasets.

[0692] Next, the emotion engine collects user emotion data through the device. This emotion data is obtained from user interactions and interpreted using natural language processing. As a result, the next generation of models receive emotion information as additional adaptive feedback, improving the accuracy of their learning.

[0693] During the retuning process, the server provides the new generation model with specialized datasets and sentiment data, allowing the model to be readjusted to deepen its understanding of specific domains. The auto-tuning function monitors the entire process and efficiently adjusts parameters.

[0694] Finally, users utilize retrained, specialized models to respond to specific domain needs and tasks. At this stage, the addition of sentiment-based analysis provides more personalized responses and support to the user.

[0695] As a concrete example, in the field of education, we can consider a system that allows teachers to evaluate students' participation based on their reactions and use that information to improve individualized instruction. This system uses an emotion engine to analyze students' responses in real time and adjust the educational content using that information, thereby enabling the delivery of more effective education.

[0696] The following describes the processing flow.

[0697] Step 1:

[0698] The server selects a large-scale language model from the previous generation and loads it from storage into memory. This makes the data and parameters of the older model available.

[0699] Step 2:

[0700] The server establishes a data communication channel to link the new generation of large-scale language models with the older generation models. This link allows the new generation models to leverage the knowledge of the older generation models.

[0701] Step 3:

[0702] The server retrieves datasets related to a specific field from a database and supplies them to the next-generation model. For example, it prepares specific medical data to enable the model to learn field-specific knowledge.

[0703] Step 4:

[0704] The emotion engine built into the device collects emotional data through interaction with the user. This includes voice tone and facial recognition data, which are then analyzed through natural language processing.

[0705] Step 5:

[0706] The server provides the next-generation model with domain-specific datasets and sentiment data collected from users, and performs retuning. The sentiment data is used as learning feedback and contributes to the model's refinement.

[0707] Step 6:

[0708] The server monitors the entire learning process using an auto-tuning function. It evaluates the model's learning effectiveness in real time and optimizes parameters as needed. As a result, effective model readjustment is achieved.

[0709] Step 7:

[0710] Users can utilize retrained, specialized models to obtain responses to specific challenges and needs in particular fields. For example, it becomes possible to provide personalized medical advice or educational content based on the user's emotions.

[0711] (Example 2)

[0712] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0713] To meet the evolving needs of information processing models and diverse application fields, there is a demand for personalized responses that reflect user sentiment while leveraging the knowledge gained from older models. Furthermore, the readjustment of information processing models for efficient operation in specific application fields remains a challenge.

[0714] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0715] In this invention, the server includes means for selecting and loading an older generation information processing model into memory; means for linking a new information processing model to the older generation information processing model using a communication path; means for acquiring a set of knowledge related to a specific application field and providing it to the new information processing model; means for collecting user emotional information through a terminal and interpreting it using natural language processing; means for incorporating the emotional information into the new information processing model and readjusting it; and means for monitoring the readjustment process and efficiently adjusting it by utilizing an automatic adjustment function. This makes it possible to provide personalized responses in a specific application field.

[0716] An "older generation information processing model" is a collection of information processing algorithms developed in previous generations, and it possesses more fundamental knowledge and data processing processes compared to models based on new technologies.

[0717] A "new information processing model" is an information processing system that employs the latest technologies and integrates knowledge from previous generation models to possess the ability to perform advanced data analysis and inference suitable for the application field.

[0718] A "communication path" is a physical or logical connection means for transmitting data and insights between information processing models, enabling the sending and receiving of information.

[0719] A "knowledge set" is a collection of data and information relating to a specific application field, including the latest case studies and research results related to that application field.

[0720] "Automatic adjustment function" is a technology that monitors information processing processes in real time and autonomously makes adjustments necessary to maintain optimal performance.

[0721] "User emotional information" refers to data about emotional responses and states obtained through interaction with the user, and is interpreted using natural language processing.

[0722] "Personalized responses" refer to information provision and actions tailored to the individual user's needs and circumstances, enabling effective support in specific application areas.

[0723] This invention is a system that provides personalized responses by efficiently combining older and newer information processing models, and further utilizing user emotional information.

[0724] First, the server selects an older generation information processing model and loads it from storage into memory. This prepares the older generation model for linking with the newer information processing model. Next, the server links the newer information processing model with the older generation model using a data communication path. This link allows the newer model to leverage the insights from the older generation model. The hardware used includes a database server and a communication network. The software employs machine learning algorithms and natural language processing tools.

[0725] The server also retrieves knowledge sets relevant to specific application areas from the database and provides them to a new information processing model. This is a step for the new model to learn information specific to the application area. For example, in the medical field, the latest research data and patient health information are used as the knowledge set.

[0726] The device plays a role in collecting user emotional information. This involves analyzing the user's facial expressions and voice using the camera and microphone, and interpreting the emotions through natural language processing. This process provides data that helps understand the user's current situation and needs, contributing to improving the learning accuracy of new information processing models.

[0727] Users receive personalized responses through the system. For example, in the field of education, teachers can analyze student responses and optimize learning plans based on that data. Examples of specific prompts include, "Please tell me about recent advances in medical technology," or "Please suggest ways to improve students' learning progress."

[0728] In this way, combining emotional information with older and newer information processing models enables effective and flexible responses in specific application areas.

[0729] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0730] Step 1:

[0731] The server selects an older generation information processing model from storage and loads it into memory. It uses the identifier and associated file path of the older generation model as input and outputs the model data loaded into memory. Specifically, the server executes the process of reading the model file from storage and placing it into memory.

[0732] Step 2:

[0733] The server links the new information processing model to the old generation information processing model using a data communication path. It takes specific information about the old and new generation models as input and outputs a state where the communication path between the models has been established. Specifically, the server connects to both models via an API and sets up a state where data can be exchanged.

[0734] Step 3:

[0735] The server retrieves a set of knowledge relevant to a specific application domain from a database and provides it to a new information processing model. It uses application domain query data as input and provides the model with appropriate knowledge information as output. Specifically, the server executes database queries and passes the extracted data to the model's learning system.

[0736] Step 4:

[0737] The device collects user emotional information. It uses signal data of the user's facial expressions and voice, acquired through the camera and microphone, as input, and obtains analyzed emotional data as output. Specifically, this involves an emotion analysis engine built into the device processing the signal data, identifying emotional patterns, and transmitting them as numerical data.

[0738] Step 5:

[0739] The server incorporates emotional data into the new information processing model and readjusts it. It uses user emotional data and a specialized knowledge set as input and designs the adjusted model parameters as output. Specifically, the server generates a feedback loop using the emotional data to optimize the model so that it outputs a response that is more appropriate for the application field.

[0740] Step 6:

[0741] The user receives personalized responses from the system. The user sends prompts as input and receives optimized response data as output. Specifically, the process involves the user communicating questions and requests to the system via a terminal, and the model returning information tailored to those requests.

[0742] (Application Example 2)

[0743] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0744] In modern information processing technology, providing personalized information that responds to user emotions is a crucial challenge. However, conventional technologies have struggled to accurately analyze user emotions and recommend optimized information based on that data. Furthermore, there is a need for a solution that effectively links and reuses both old and new information processing technologies.

[0745] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0746] In this invention, the server includes means for selecting and loading older generation information processing technologies linked to new generation information processing technologies in order to reuse existing information processing technologies; means for the new generation information processing technologies to readjust the older generation information processing technologies using a set of information related to a specific field; and means for generating personalized information using sentiment analysis technology. This enables precise information recommendations based on the user's emotions.

[0747] "Information processing technology" refers to the technology of processing, analyzing, transforming, or optimizing digital data to serve a specific purpose.

[0748] An "information set" is a collection of data and information related to a specific field, which is then analyzed and processed.

[0749] "Emotional analysis technology" is a technology that analyzes users' emotions and sensitivities and optimizes data processing and service provision based on the results.

[0750] "Communication technology" refers to technologies for transmitting, receiving, and exchanging data and information, and enables the coordination of different technologies and systems.

[0751] "Personalized information" refers to information customized based on the individual user's characteristics and preferences, aiming to provide more appropriate and efficient information.

[0752] The system of the present invention links older and newer information processing technologies to provide personalized information based on user sentiment data. The server first selects and loads older information processing technologies to reuse existing ones. Next, the newer information processing technologies readjust the older technologies using a set of information relevant to a specific field.

[0753] This process uses emotion analysis technology to acquire user emotion data and optimize the service. Users can provide their emotional information to the system through devices such as smartphones and smart glasses, and the server processes this data to provide personalized information.

[0754] Specifically, the server monitors the readjustment process using an auto-tuning function, completing the learning process at the optimal time. In this process, older and newer generation information processing technologies are effectively coordinated using communication technology. As a result, precise information recommendations based on user emotions are realized.

[0755] For example, an entertainment information application can recommend movies that match the user's current mood when they choose a film. If the user is feeling sad, the system will recommend a more cheerful comedy film, providing a service that is considerate of the user's feelings.

[0756] An example of a prompt message might be: "Recommend a movie using the emotions the user is currently feeling. If the user is sad, choose a comedy movie to cheer them up."

[0757] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0758] Step 1:

[0759] The server selects older generation information processing technologies and loads them from storage into memory. The input here is the identification information of the older generation technology, and the output is an instance of the loaded technology. This loading operation prepares the older generation technology for reuse.

[0760] Step 2:

[0761] The server uses communication technology to link new-generation information processing technology with older-generation technology. The input to this step is information from both the old and new generations of technology, and the output is the established link. This link allows the new generation technology to utilize the knowledge of the old generation.

[0762] Step 3:

[0763] The server retrieves a set of information related to a specific field from a database. The input is a request related to that field, and the output is the retrieved information set. The server uses this information to prepare for the re-engineering of next-generation technologies.

[0764] Step 4:

[0765] The device collects user emotional information using emotion analysis technology and sends it to a server. The input is data of the user's facial expressions and voice, and the output is the analyzed emotional information. This provides data based on the user's emotions.

[0766] Step 5:

[0767] The server uses emotional information and data sets to retune next-generation technologies. The input is emotional information and data sets, and the output is the parameters of the updated technologies. This allows the system to provide users with more relevant information.

[0768] Step 6:

[0769] The server monitors the readjustment process using an auto-tuning function and completes it at the optimal time. The input is the readjustment progress data, and the output is the readjustment completion time. This enables an efficient learning process.

[0770] Step 7:

[0771] The server uses tuned information processing technology to provide users with personalized information. The input consists of the user's request and the tuned technology, while the output is information optimized for the user. As a result, users can receive information that resonates with their own emotions.

[0772] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0773] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0774] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0775] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0776] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0777] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0778] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0779] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0780] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0781] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0782] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0783] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0784] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0785] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0786] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0787] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0788] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0789] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0790] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0791] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0792] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted to be incorporated by reference.

[0793] The following is further disclosed regarding the embodiments described above.

[0794] (Claim 1)

[0795] In order to reuse existing large-scale language models, a means for selecting and loading older generation large-scale language models to be linked to new generation large-scale language models,

[0796] A means for the new generation of large-scale language models to readjust the old generation of large-scale language models using datasets related to a specific field,

[0797] A means of monitoring the readjustment process using an auto-tuning function and terminating the learning process at the optimal time,

[0798] A system that includes this.

[0799] (Claim 2)

[0800] The system according to claim 1, wherein the aforementioned specific field is the medical field.

[0801] (Claim 3)

[0802] The system according to claim 1, wherein the link between the older generation large-scale language model and the newer generation large-scale language model is realized as a data communication channel.

[0803] "Example 1"

[0804] (Claim 1)

[0805] A means for selecting and loading information from a digital data storage device as an information format for an older generation information processing device,

[0806] A means for connecting a new generation of information processing equipment with an older generation of information processing equipment to enable information exchange,

[0807] A means by which the new generation information processing device readjusts the old generation information processing device using a collection of information related to a specific industrial field,

[0808] A means of monitoring the readjustment procedure using a self-optimization function and adjusting the learning procedure in a variable manner,

[0809] A system that includes this.

[0810] (Claim 2)

[0811] The system according to claim 1, wherein the aforementioned specific industrial field is the medical field.

[0812] (Claim 3)

[0813] The system according to claim 1, wherein the combination of the older generation information processing device and the new generation information processing device is realized as an information exchange path.

[0814] "Application Example 1"

[0815] (Claim 1)

[0816] In order to reuse existing data processing models, a means for selecting and loading older generation information processing models to be linked to new generation information processing models,

[0817] A means by which the new generation information processing model readjusts the old generation information processing model using a data set related to a specific domain,

[0818] A means of monitoring the readjustment process using an automatic adjustment function and ending the learning process at the optimal time,

[0819] A means of acquiring information from measuring devices and image acquisition devices, comparing it with past data patterns derived from previous generation models to detect problems, and proposing improvement measures,

[0820] A system that includes this.

[0821] (Claim 2)

[0822] The system according to claim 1, wherein the specified domain is a manufacturing operation.

[0823] (Claim 3)

[0824] The system according to claim 1, wherein the link between the older generation information processing model and the new generation information processing model is realized as an information transmission path.

[0825] "Example 2 of combining an emotion engine"

[0826] (Claim 1)

[0827] A means of selecting an older generation information processing model, loading it into memory, and preparing it for reuse,

[0828] A means of linking a new information processing model to an older generation information processing model using a communication path, and utilizing the knowledge gained from the older generation model,

[0829] A means of acquiring a set of knowledge related to a specific application field and providing it to a new information processing model,

[0830] A means of collecting user emotional information through a device and interpreting it using natural language processing,

[0831] By incorporating emotional information into new information processing models and readjusting them, we can deepen our understanding of application fields.

[0832] A means of efficiently performing adjustments by utilizing the automatic adjustment function to monitor the readjustment process,

[0833] A means of providing personalized responses to users,

[0834] A system that includes this.

[0835] (Claim 2)

[0836] The system according to claim 1, wherein the aforementioned specific application field is the field of education.

[0837] (Claim 3)

[0838] The system according to claim 1, wherein the link between the older generation information processing model and the new information processing model is realized as a communication path.

[0839] "Application example 2 when combining with an emotional engine"

[0840] (Claim 1)

[0841] In order to reuse existing information processing technologies, a means for selecting and loading older generation information processing technologies to be linked to new generation information processing technologies,

[0842] A means by which the new generation of information processing technology readjusts the old generation of information processing technology using a set of information related to a specific field,

[0843] A means of generating personalized information using emotion analysis technology,

[0844] A means of monitoring the readjustment process using an auto-tuning function and terminating the learning process at the optimal time,

[0845] A system that includes this.

[0846] (Claim 2)

[0847] The system according to claim 1, wherein the specified field is entertainment information.

[0848] (Claim 3)

[0849] The system according to claim 1, wherein the link between the older generation information processing technology and the newer generation information processing technology is realized as a communication technology. [Explanation of Symbols]

[0850] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. In order to reuse existing large-scale language models, a means for selecting and loading older generation large-scale language models to be linked to new generation large-scale language models, A means for the new generation of large-scale language models to readjust the old generation of large-scale language models using datasets related to a specific field, A means of monitoring the readjustment process using an auto-tuning function and terminating the learning process at the optimal time, A system that includes this.

2. The system according to claim 1, wherein the aforementioned specific field is the medical field.

3. The system according to claim 1, wherein the link between the older generation large-scale language model and the new generation large-scale language model is realized as a data communication channel.

Images