Information processing device and information processing method

Abstract

[Problem] Conventional operating systems (OS) run on the basis of static algorithms and predefined rules and thus cannot sufficiently adapt to dynamic environmental changes or complex system commands. Additionally, conventional technologies present problems such as inefficient energy management, security vulnerability, and system scalability limitations. [Solution] In the present invention, an OS is composed of a coordinator AI 210 and a plurality of module AIs, and implements system-wide resource management, process scheduling, memory optimization, security management, improved energy efficiency, and the like. In particular, the present invention comprises an advanced AI management function for supporting the safe and efficient operation of artificial general intelligence (AGI) and artificial superintelligence (ASI).

Description

Information processing device and information processing method

[0001] The present invention relates to a new type of operating system (hereinafter referred to as "AI / OS") that replaces conventional operating systems (OSs) and performs system control and resource management based on artificial intelligence (AI) technology. In particular, AI / OSs belong to the technical field of using AI to automate and optimize the roles previously played by conventional OSs, such as system resource management, process scheduling, memory allocation, file system control, and device management. Furthermore, AI / OSs improve the efficiency and stability of the entire system by using dynamic, learnable AI algorithms instead of the static management methods used in conventional OSs, and are particularly related to fundamental technologies for the safe and efficient operation of highly intelligent AGIs (artificial general intelligence) and ASIs (artificial superintelligence).

[0002] The main roles of conventional operating systems (OSs) are managing hardware resources, scheduling processes, allocating memory, controlling file systems, and managing devices. However, these OSs basically operate based on static algorithms and predefined rules, making it difficult to adequately adapt to dynamic environmental changes and complex system requirements.

[0003] Existing operating systems (OS) have several significant design challenges. First, they suffer from inefficient resource management. Existing OSes manage resources such as CPU, memory, and storage by relying on static allocations or predefined algorithms. However, this approach may not be flexible enough to handle dynamic load fluctuations and unexpected resource demands. This can result in poor system performance, excessive resource consumption, and system instability.

[0004] Second, there are security vulnerabilities. Many operating systems rely on fixed security protocols and signature-based malware protection, making it difficult to respond quickly to new threats and unknown attacks. This situation makes the system more vulnerable to security risks, which can lead to serious problems such as data loss and system destruction.

[0005] Furthermore, system scalability is also a major issue for existing operating systems. Scaling out or scaling up, especially in cloud-based distributed systems or high-performance computing environments, requires extensive manual configuration, which increases operational costs. Traditional operating systems lack the flexibility to accommodate system expansions and changes, which reduces the overall system efficiency.

[0006] Another problem with existing operating systems is the lack of consistency in the user experience. Inconsistent operations across different applications and services can lead to poor user experience and negatively impact productivity. This inconsistency makes it difficult to provide a consistent user experience, making improving usability a challenge.

[0007] Finally, there is the complexity of system management. Existing operating systems require system administrators to perform complex configuration and maintenance manually, which increases the risk of errors and increases the workload. This increases the cost of running the system and makes efficient management difficult.

[0008] These challenges are particularly pronounced when implementing highly intelligent artificial intelligence (AGI or ASI). Conventional operating systems lack the flexibility and adaptability to handle the complex and dynamic environments required by these advanced AIs, making efficient and safe operation difficult. To solve these problems, a new operating system based on AI technology, i.e., an AI / OS, is needed. The present invention aims to provide an A / OS that overcomes the limitations of these conventional operating systems and enables more advanced, flexible, and safe system management.

[0009] The information processing device of the first invention is an operating system executed on a computer, comprising a coordinator AI that monitors the operating status of the entire system and controls the operation of multiple functional modules in an integrated manner, and multiple modules AI that analyze the usage status of system resources for each function instructed by the coordinator AI and set allocation or usage restrictions of the system resources based on the analysis results, characterized in that the coordinator AI obtains the analysis results from the multiple modules AI and dynamically changes the allocation or usage restrictions of the system resources based on the obtained analysis results.

[0010] The information processing device of the second invention is characterized in that, in the first invention, the plurality of modules AI include at least one functional module AI selected from the group consisting of a memory management module AI, a process management module AI, a file system management module AI, a device management module AI, and a battery management module AI.

[0011] The information processing device of the third invention is characterized in that, in the first invention, the plurality of modules AI include at least one functional module AI selected from the group consisting of a security access control module AI, an energy resource optimization module AI, and a network and communication management module AI.

[0012] The information processing device of the fourth invention is characterized in that, in the first invention, the plurality of modules AI include at least one functional module AI selected from the group consisting of a high-level task management module AI, an ethics and compliance management module AI, a self-learning and self-evolution module AI, and a social interface module AI.

[0013] The information processing device of the fifth invention is characterized in that, in the first invention, the coordinator AI optimizes the resource management policy of the entire system based on the internal cognitive state of the energy / resource optimization module AI recognized by the module AI.

[0014] The information processing device of the sixth invention is characterized in that, in the first invention, the multiple functional module AI calculates an emotional salience score based on the user's emotions or the degree of goal achievement, and transmits it to the coordinator AI, and the coordinator AI optimizes process priority, memory allocation, and network bandwidth in real time according to the emotional salience score.

[0015] The information processing device of the seventh invention is characterized in that, in the first invention, the plurality of module AIs maximize the sensor recording parameters via the coordinator AI when the emotional salience score exceeds a threshold.

[0016] An information processing device according to an eighth aspect of the present invention is characterized in that, in the first aspect, the plurality of functional modules AI include a security access control module AI, which embeds contextual information including at least time, location, and event tags and an emotional salience score into the data for multimodal information obtained by active perception, thereby concealing the information.

[0017] The information processing device of the ninth invention is characterized in that, in the eighth invention, the security access control module AI autonomously updates the algorithms and internal models of each module AI, and records all important OS-level decisions and state transitions, including self-improvement processes, in a blockchain-based distributed ledger, ensuring immutability, traceability, and auditability.

[0018] An information processing device according to a tenth aspect of the present invention is the information processing device of the first aspect, further comprising a user emotion information compatible application that generates emotion information based on biometric data or input data of a subject, acquires first content based on the generated emotion information, integrates the emotion information with the first content, generates second content based on the integrated information, and outputs the generated second content, wherein the user emotion information compatible application transmits the generated emotion information to the coordinator AI, and the coordinator AI dynamically changes the allocation or usage restrictions of the system resources based on the emotion information.

[0019] An information processing device according to an eleventh aspect of the present invention is characterized in that, in the first aspect, the information processing device further includes a user interaction application that acquires input content, analyzes the acquired input content to acquire information on user characteristics, intentions, and background, generates a response based on the input content, the user characteristics information, and the information on intentions and background, and outputs the generated response, and the user interaction application transmits the acquired information on user characteristics, intentions, and background to the coordinator AI, and the coordinator AI dynamically changes the allocation or usage restrictions of the system resources based on the information.

[0020] The information processing method of the twelfth invention is an information processing method by an operating system executed on a computer, characterized in that it includes the steps of monitoring the operating status of the entire system and integrally controlling the operation of multiple functional modules, analyzing the usage status of system resources for each function instructed in the integral control step and executing multiple module AIs that set allocation or usage restrictions of the system resources based on the analysis results, and the integral control step further includes the steps of acquiring the analysis results in the step of executing the multiple module AIs and dynamically changing the allocation or usage restrictions of the system resources based on the acquired analysis results.

[0021] The present invention provides the following advantages. First, an OS kernel composed of a coordinator AI and multiple module AIs achieves dynamic and adaptive resource management compared to conventional static resource management. The coordinator AI integrates the entire system, and each module AI is responsible for specialized processing, thereby optimizing resource allocation and improving the efficiency of the entire system.

[0022] Furthermore, the system's flexibility is greatly improved by the structure in which each module AI operates independently while the coordinator AI centrally manages it. Each module AI can make optimal decisions based on real-time data in its own field of expertise and self-improve as needed. This structure makes it easy to expand the system and add new modules, allowing the system to continue evolving over the long term.

[0023] In addition, the introduction of additional AI modules, such as the security and access control module AI and the network and communication management module AI, will improve the security and reliability of the system, allowing for a rapid and effective response to unknown threats and greatly enhancing the safety of the system.

[0024] Furthermore, the AI / OS based on this invention provides a foundation for the efficient and safe operation of highly intelligent AI systems such as AGI and ASI. Cooperation between the coordinator AI and module AI enables system management with a high degree of adaptability and efficiency that was not possible with previous OSs, and can flexibly respond to the enormous resource demands and complex tasks of AGI and ASI.

[0025] As described above, the AI / OS provided by the present invention solves the problems faced by conventional OSs and can demonstrate excellent performance as the foundation for next-generation AI technology.

[0026] FIG. 1 is a diagram showing a basic hierarchical structure in a computer system according to a first embodiment. FIG. 2 is a diagram illustrating a hardware configuration of an information processing device according to the first embodiment. FIG. 3 is a flowchart illustrating execution and processing of the operating system 21 according to the first embodiment. FIG. 4 is a diagram illustrating AI and other AIs constituting the kernel of the operating system 21 according to the second embodiment. FIG. 5 is a flowchart illustrating execution processing of the operating system 21 according to the second embodiment. FIG. 6 is a diagram illustrating an overview of a user emotion information responsive application according to a third embodiment. FIG. 7 is a diagram illustrating a functional block configuration of a user emotion information responsive application according to the third embodiment. FIG. 8 is a flowchart illustrating processing of a user emotion information responsive application according to the third embodiment. FIG. 9 is a diagram illustrating an example of prompt and answer content of a user emotion information responsive application according to the third embodiment. FIG. 10 is a diagram illustrating a functional block configuration of a terminal 50 of an information processing device to which a user interaction responsive application according to the fourth embodiment is applied. FIG. 11 is a flowchart illustrating processing of a terminal 50 of an information processing device to which a user interaction responsive application according to the fourth embodiment is applied. FIG. 12 is a diagram illustrating an example of a dialogue by response generation by an information processing device to which a user interaction responsive application according to the fourth embodiment is applied.

[0027] Hereinafter, embodiments of the present invention will be described in detail with reference to the drawings. In the following, the same or corresponding parts in the drawings are denoted by the same reference numerals, and their description will not be repeated in principle.

[0028] (Embodiment 1) Embodiment 1 is based on a typical computer configuration. FIG. 1 is a diagram showing the basic hierarchical structure of a computer system according to embodiment 1. Referring to FIG. 1, the computer system is composed of three main layers. The lowest physical layer 1 represents the hardware and includes the physical components of the computer. The middle layer 2 is composed of device drivers 22 and an operating system 21. The operating system 21 forms the core of the middle layer 2 and functions as a coordinator for the entire system, managing hardware resources, controlling processes, managing memory, and providing a file system. It also acts as a bridge between the application layer 3 and the hardware 1, providing a standardized interface to facilitate application development. The device drivers 22 are responsible for controlling specific hardware devices and enabling communication between the operating system 21 and the hardware 1. The top application layer 3 includes software programs that are directly operated by the user and operates using services provided by the operating system 21. This hierarchical structure maintains the independence of each layer and ensures the flexibility and scalability of the system.

[0029] 2 is a diagram illustrating the configuration of hardware 1 of the information processing device according to embodiment 1. The hardware 1 includes a control unit 10, an input device 11, a camera 12, a microphone 13, a sensor 14, an output device 15, a storage device 16, a communication I / F 17, and an internal bus connecting the various units.

[0030] Furthermore, the information processing device hardware 1 is connected via a network NW to a terminal 50, an information processing device 100, a large-scale language model system 200, and a distributed ledger server 500 equipped with a blockchain function.

[0031] The control unit 10 includes a CPU (Central Processing Unit), RAM (Random Access Memory), and ROM (Read Only Memory). The input device 11 accepts input data (input information) from a user. The input device 11 includes, for example, a known mouse or keyboard. The input device 11 may use, for example, a voice recognition module, and may convert voice input from the user into text data before accepting the input. The input device 11 may accept input data and voice data from the user via, for example, a microphone 13 or a known microphone separate from the microphone 13.

[0032] The camera 12 acquires image data (image information) of the target object. However, other information may be acquired. For example, an infrared camera may be used to acquire surface temperature information of the target object.

[0033] The sensors 14 are used for various purposes, such as collecting environmental information, navigation, object recognition, and ensuring safety. An appropriate sensor is selected depending on the application and environment, such as an optical sensor, acoustic sensor, motion sensor, pressure sensor, temperature sensor, chemical sensor, magnetic sensor, proximity sensor, or biosensor, and it is common to use multiple sensors in combination.

[0034] The output device 15 plays a role in displaying or outputting information and data processed within the computer to the outside in a format that can be understood by humans. Examples of the output device 15 include a display device such as a monitor that outputs text data, image data, video data, etc., and a playback device such as a speaker that outputs audio data, etc.

[0035] The storage device 16 stores various software programs including an operating system.

[0036] The communication I / F 17 is connected to the network NW and executes the transmission and reception of data (information) between the network NW and external devices.

[0037] The terminal 50 is, for example, a personal computer (PC) owned by the target learner (user), and may be portable or fixed. Various applications are installed on the terminal 50, or the terminal 50 is remotely connected to other servers, terminals, etc. via a network NW, allowing various applications (such as user emotion information-enabled applications and user interaction-enabled applications, which will be described later) to be used online, and allows the user of the application to learn programming. The terminal 50 also displays a learning screen 51 for learning programming, and the learning screen 51 may include, for example, a character 52.

[0038] The information processing device 100 is configured to be able to communicate with the large-scale language model system 200 via a network NW. The information processing device 100 may be connected to a terminal 50 to provide various applications or to return processing results in response to instructions from the terminal 50.

[0039] The large-scale language model system 200 includes a question receiving unit 201 that receives questions from users, an answer generating unit 202 that generates answers to the questions, and a known trained model 204 that can learn from large amounts of text data and generate natural-looking sentences that sound like they were written by a human, and may also include various AI models (not shown), etc.

[0040] 3 is a diagram illustrating the configuration of a kernel of operating system 21 according to embodiment 1. Referring to Fig. 3, the kernel of operating system 21 is made up of a coordinator AI210, a memory management module AI211, a process management module AI212, a file system management module AI213, a device management module AI214, and a battery management module AI215.

[0041] (Coordinator AI 210) The coordinator AI 210 plays a central role in the kernel of the operating system 21. This coordinator AI 210 is responsible for managing resources across the system, allocating tasks, and maintaining system stability and efficiency. The coordinator AI 210 has the following specific functions:

[0042] Task Allocation and Management: The Coordinator AI 210 monitors all tasks running in the system and assigns them to the appropriate Module AI. This includes process scheduling, resource allocation, and task prioritization. For example, critical system services are given higher priority and given preferential resource allocation.

[0043] <Resource Management> The coordinator AI 210 monitors the usage of all resources in the system (CPU, memory, disk space, battery, etc.) and manages them efficiently. It appropriately allocates the resources required by each process and device to prevent resource conflicts and shortages. In particular, by working with the battery management module AI, it optimizes energy efficiency based on the battery status and adjusts power consumption throughout the system.

[0044] <Monitoring System Stability and Performance> The coordinator AI 210 monitors the performance metrics of the entire system (CPU usage, memory usage, disk I / O, network traffic, battery consumption, etc.) in real time. If it detects abnormal behavior or excessive resource usage, it takes appropriate action. In particular, if battery consumption increases rapidly or the battery condition deteriorates, it takes action such as activating an emergency power-saving mode.

[0045] <Access to Hardware> The coordinator AI 210 accesses all hardware devices in the system based on hardware information obtained from the BIOS and UEFI. For example, when the system starts up, it scans the PCI bus and USB bus to obtain the vendor ID and device ID of the connected devices. It also works with the battery management module AI to monitor the battery status and operating conditions in real time and maintain the battery's health.

[0046] <Information Acquisition and Communication> The coordinator AI 210 aggregates feedback and system event information from each module AI to grasp the overall system status. It also issues instructions to each module AI as needed to adjust the overall system. It also receives information from the battery management module AI and optimizes system operation by allocating resources and scheduling tasks according to the battery status.

[0047] (Memory Management Module AI211) The memory management module AI211 allocates and releases memory, manages virtual memory, and optimizes memory, efficiently managing system memory resources, allocating appropriate memory to each process and application, and ensuring efficient memory usage throughout the system. Specific functions are described below.

[0048] <Memory Allocation and Release> When a new process or application requests memory, the memory management module AI211 checks the available memory in the system and allocates the required amount of memory. When the process ends, the memory used is released and made available for reuse.

[0049] <Memory Monitoring> The memory management module AI211 monitors the memory usage of the entire system in real time. This includes the usage of used memory, free memory, and swap space. If it detects abnormal memory usage patterns or signs of memory shortage, it issues a warning and takes appropriate action.

[0050] <Paging Management> The memory management module AI211 manages the paging of virtual memory, including setting up page tables, handling page faults, and swapping pages in and out. Paging allows the system to efficiently utilize limited physical memory.

[0051] <Using System Calls> The memory management module AI211 obtains memory information through system calls, thereby managing memory resources in a safe and standardized manner and maintaining system stability.

[0052] <Memory Reallocation> When reallocating memory allocated to a specific process, the memory management module AI211 checks the current memory usage and expands or shrinks the memory area as needed, thereby improving the flexibility and efficiency of the system.

[0053] (Process Management Module AI212) The process management module AI212 is responsible for creating, managing, and terminating all processes executed within the system, and plays an important role in supporting efficient operation of the system. Specific functions of the process management module AI212 are described below.

[0054] <Process Creation and Termination> When the process management module AI212 receives a request to create a new process, it registers the necessary information in the process table and starts executing the process. When the process ends, it releases the resources it was using and deletes them from the process table. This allows for efficient management of system resources and prevents resource waste.

[0055] <Process Scheduling> The process management module AI212 manages multiple processes running simultaneously within the system and allocates appropriate CPU time to each process. Scheduling is performed based on process priority, optimizing overall system performance. In particular, more resources are allocated to high-priority tasks, ensuring that important system functions are executed without delay.

[0056] <Process State Management> The process management module AI212 tracks the state of each process (ready, running, waiting, finished, etc.) and transitions the state at the appropriate time. This ensures efficient system operation and minimizes contention between processes. The process management module AI212 optimizes process state transitions to maximize overall system throughput.

[0057] Inter-Process Communication Management The process management module AI212 manages inter-process communication (IPC) to enable processes to exchange data safely and efficiently. This includes managing IPC mechanisms such as message queues, semaphores, and shared memory. It facilitates cooperation between processes and improves the efficiency of parallel processing.

[0058] <Process Resource Management> The process management module AI212 monitors the resources required by each process (CPU time, memory, I / O resources, etc.), and allocates and reallocates resources. If resources are insufficient, adjustments are made based on the priorities and resource usage status of other processes to maintain the stability of the entire system.

[0059] Process Prioritization: The process management module AI212 prioritizes processes based on their importance and urgency, allowing important processes to have priority over other processes in terms of resource availability and ensuring that important system functions are performed reliably.

[0060] (File System Management Module AI213) The file system management module AI213 manages the entire file system of the system and is responsible for storing, accessing, and protecting data. Specific functions are described below.

[0061] <Mounting the Root File System> The file system management module AI213 recognizes the root file system when the system is started and mounts it, enabling basic system operations and data access.

[0062] <File System Integrity Checks> After the root file system is mounted, the file system management module AI213 performs the necessary file system integrity checks, using file system tools (e.g., fsck) to verify the integrity of the file system metadata and block data.

[0063] <Mounting additional file systems> Because the system has multiple file systems, the file system management module AI213 also mounts data partitions that store user data and application data, as well as network file systems (e.g., NFS and SMB), allowing for efficient data management across the entire system.

[0064] <File System Monitoring> The file system management module AI213 continuously monitors the status of the file system while the system is running. This includes monitoring disk usage, logging file read and write operations, and error handling. If an abnormality is detected, it takes appropriate action and initiates a repair process if necessary.

[0065] <Access Rights and Security Management> The file system management module AI213 manages access rights to files and directories and protects data. It sets access rights for each user and prevents unauthorized access.

[0066] <Backup and Recovery> The file system management module AI 213 manages regular data backup and recovery, which enables rapid recovery in the event of data loss.

[0067] (Device Management Module AI214) The device management module AI214 is responsible for managing and initializing all hardware devices connected to the system. Specific functions are as follows:

[0068] <Device Recognition and Driver Loading> When the system starts up, the device management module AI 214 scans all connected devices and obtains the vendor ID and device ID of each device. Based on this, it selects and loads the appropriate device driver, allowing each device to communicate properly with the system.

[0069] <Device Initialization> The device management module AI 214 initializes each device using the loaded driver. For example, it sets the resolution and refresh rate for a display device, and checks and sets the input signal for a keyboard or mouse.

[0070] <Network Interface Configuration> The device management module AI 214 recognizes network adapters, loads the corresponding drivers, configures IP addresses, and establishes network connections, allowing the system to communicate with external networks.

[0071] <Device Monitoring and Management> After initialization is complete, the device management module AI 214 continues to monitor the status of each device. It records device performance and performs appropriate error handling if an abnormality is detected. This ensures device stability and reliability.

[0072] <Firmware Update> The device management module AI 214 also manages firmware updates for devices, which makes it easier to upgrade device functionality and apply security patches.

[0073] (Battery Management Module AI215) The battery management module AI215 monitors the system's battery status in real time and plays an important role in optimizing the energy efficiency of the entire system. The battery management module AI215 provides the following specific functions to extend the battery life and stabilize the system operation:

[0074] Battery Monitoring and Status Reporting: The battery management module AI215 monitors battery data such as charge level, health status (deterioration), temperature, and number of charge / discharge cycles in real time. This allows the battery status to be constantly monitored and appropriate status information provided to the entire system. It also issues warnings and responds promptly to abnormal battery operation or sudden power consumption.

[0075] <Power-saving mode management> The battery management module AI215 automatically applies power-saving modes according to the remaining battery level and usage status. This includes adjusting screen brightness, limiting background applications, adjusting the CPU clock, etc. When the battery level is low, it reduces power consumption across the entire system to extend battery life.

[0076] Charging Optimization: The battery management module AI215 optimizes the charging process to maximize battery life. This includes adjusting the charging rate and maintaining a charge level of approximately 80% to prevent battery degradation. It can also pause charging based on battery health or limit charging under certain conditions.

[0077] <Optimization of energy consumption> The battery management module AI215 monitors the energy consumption of all processes and devices in the system and performs optimization to prevent unnecessary consumption. It maximizes battery efficiency by terminating unnecessary processes and applications and allocating resources appropriately for operation in low power states.

[0078] Alerts and Notifications: The battery management module AI 215 provides important notifications and alerts to the user regarding the battery status. These include warnings when the battery is low, when the battery temperature is abnormally high, and when the battery needs to be replaced. This allows the user to stay aware of the battery status and take appropriate action.

[0079] Learning and predicting battery usage patterns: The battery management module AI 215 learns system usage patterns and predicts battery consumption based on user activity, enabling smarter battery management, such as proactively applying power-saving measures to prepare for situations where the battery is likely to be drained.

[0080] Thermal Management: The battery management module AI215 constantly monitors the battery temperature and takes steps to cool the system if the temperature gets too high. Thermal management is essential to maintaining the long-term health of the battery, as excessive temperature can cause battery degradation and shorten its lifespan.

[0081] <Cooperation with the System> The battery management module AI 215 cooperates with the coordinator AI and other module AIs to integrate energy management for the entire system, thereby adjusting the operating mode of the entire system according to the battery status and minimizing battery consumption.

[0082] FIG. 4 is a flow diagram illustrating the execution and processing of the operating system 21 according to the first embodiment. Referring to FIG. 4, when the power is turned on, the computer system goes through a series of processes. Eventually, the terminal starts up, and the user is able to give commands to the AI ​​by text or voice. The cooperation between the coordinator AI 210 and each module AI in this series of processes will be described in detail with reference to the flow diagram.

[0083] <<Power On (Step 0)>> Initially, when power is applied to a computer system, the system's hardware begins to initialize. During this stage, the POST (Power-On Self Test) process runs, performing basic hardware checks. Once power is applied, the BIOS / UEFI boots and is ready to proceed to the next step. At this early stage, the AI ​​has not yet booted, and all operations are performed by the hardware itself and the BIOS / UEFI.

[0084] <<BIOS / UEFI Execution (Step S1)>> The BIOS or UEFI starts up, performing system initialization and hardware initialization. The BIOS / UEFI is responsible for collecting system hardware configuration information and loading the OS boot loader into memory. At this point, the hardware information is collected by the BIOS / UEFI and passed to the coordinator AI when the BIOS / UEFI boot loader starts. At the BIOS / UEFI stage, a dedicated firmware program controls the hardware. The BIOS / UEFI loads the BIOS / UEFI boot loader into memory and executes it. The boot loader loads the operating system 21 into system memory and allocates the coordinator AI 210 and other module AIs that make up the kernel in memory. At this stage, the boot loader passes the hardware information provided by the BIOS / UEFI to the operating system 21, and the coordinator AI 210 starts the initialization process based on that information. The boot loader itself operates according to a pre-programmed sequence of operations to ensure that the operating system 21 is properly started.

[0085] <<Coordinator AI Execution (Step S2)>> When the boot loader places the kernel in memory, the coordinator AI 210 starts up. The coordinator AI 210 acquires hardware information provided by the BIOS / UEFI and the boot loader, and begins controlling the entire system. Specifically, the coordinator AI 210 accesses memory, CPU, storage devices, etc., and performs their initial settings. The coordinator AI 210 directly accesses hardware registers via the system bus as necessary, and performs initial settings and resource allocation. At this time, the operating system 21 communicates with the hardware in cooperation with dedicated drivers and firmware.

[0086] <Multiple Modules AI> The multiple modules AI are functional modules having individual functions, such as a memory management module AI211, a process management module AI212, a file system management module AI213, a device management module AI214, a battery management module AI215, a security and access control module AI216, an energy and resource optimization module AI217, a network and communication management module AI218, a higher-level task management module AI219, an ethics and compliance management module AI220, a self-learning and self-evolution module AI221, and a social interface module AI222, and are configured from multiple groups.

[0087] The multiple AI modules also calculate emotional salience scores based on the user's emotions or their own goal achievement levels. The multiple AI modules acquire information about the user's emotions, for example, via the device driver 22 in the intermediate layer 2. Furthermore, target values ​​for their own function execution are set, and the goal achievement levels are calculated from the process performance of each AI module in real time or over a predetermined period. The multiple AI modules calculate emotional salience scores indicating the degree of the user's emotions according to the acquired user emotions or the calculated goal achievement levels. The emotional salience scores may be displayed, for example, as several ranks according to the calculated numerical values.

[0088] Furthermore, the multiple functional modules AI transmit their calculated emotional salience scores to the coordinator AI 210. The coordinator AI 210 optimizes the priority, memory allocation, and network bandwidth of each process in real time according to the emotional salience scores transmitted from the functional modules AI. Furthermore, the multiple functional modules AI may reconfigure the sensor recording parameters via the coordinator AI 210 to maximize them when the set emotional salience score exceeds a threshold.

[0089] <<Execution of Memory Management Module AI (Step S3)>> When the coordinator AI 210 starts the memory management module AI 211, memory management begins. The memory management module AI 211 accesses the system memory and creates an overall memory map. The memory management module AI 211 accesses the physical memory via the memory controller and sets and allocates virtual memory. Memory management here is performed by the memory management module AI 211 directly accessing and controlling the memory, as well as by utilizing memory management routines programmed in the kernel. Once memory management is complete, the coordinator AI 210 begins control to initialize the CPU and system timer. The coordinator AI 210 accesses the CPU registers and performs initial settings. This includes setting the CPU clock and enabling the cache. Furthermore, to initialize the system timer, the coordinator AI 210 accesses the timer controller and performs settings to ensure accurate system clock operation. This operation is performed by the coordinator AI 210 directly manipulating the hardware registers.

[0090] <<Execution of Process Management Module AI (Step S4)>> Once the initialization of the CPU and system timer is complete, the process management module AI212 is launched and process management within the system begins. The process management module AI212 creates a process table and manages information about each process. The process management module AI212 accesses information about each process via the system bus and controls the creation, scheduling, and termination of processes. The AI ​​also allocates resources and prioritizes processes according to a scheduling algorithm programmed into it.

[0091] <<Executing the File System Management Module AI (Step S5)>> Once the process management setup is complete, the file system management module AI213 is started and the root file system is mounted. The file system management module AI213 directly accesses the storage device and performs file system consistency checks and mounting operations. This is done using a file system driver, and the file system management module AI213 controls it using programmed routines related to file access. The file system management module AI213 continuously monitors the status of the file system and repairs or reconfigures it as necessary.

[0092] <<Executing Device Management Module AI (Step S6)>> Once the file system initialization is complete, the coordinator AI 210 starts the device management module AI 214, which initializes all hardware devices in the system. The device management module AI 214 accesses each device and loads its respective driver. This includes scanning the PCI bus and USB bus to detect connected devices. The device management module AI 214 communicates with the hardware through the driver to configure and manage the device. The driver enables the device management module AI 214 to operate according to programmed control logic and appropriately control the operation of each device.

[0093] <<Battery Management Module AI Execution (Step S7)>> After device management is complete, the battery management module AI215 is activated. The battery management module AI215 accesses the battery controller and monitors the battery status in real time. This includes monitoring the charge level, temperature, and health state. The battery management module AI215 works in conjunction with the coordinator AI210 and other modules AI to implement power saving modes and optimize charging. Battery management control is achieved by the battery management module AI215 directly accessing the battery controller and is controlled according to a programmed energy management algorithm.

[0094] <<Starting System Services (Step S8)>> Once battery management is complete, the coordinator AI 210 instructs the system services to be started. At this stage, system services such as system monitoring, log management, and security management are started in sequence. Each service accesses necessary information through the system bus and processes various events. These services are executed by the coordinator AI 210 using programmed monitoring and management routines to ensure the stability and security of the entire system.

[0095] <<Starting the Terminal Program (Step S9)>> After going through the steps up to this point, the terminal program is finally started. The terminal program provides an interface through which the user can input commands. At this stage, the terminal program is configured to work with the coordinator AI 210 and each module AI to appropriately process tasks instructed by the user. The terminal program uses the interface logic programmed into the coordinator AI 210 to analyze the user input and transmit appropriate instructions to each module AI.

[0096] (Second Embodiment) In the second embodiment, an additional module is added to the configuration of the first embodiment in order to create an operating system for executing advanced AI such as AGI or ASI.

[0097] 5 is a diagram illustrating the configuration of a kernel of the operating system 21 according to the second embodiment. The operating system 21 comprises a coordinator AI 210, a memory management module AI 211, a process management module AI 212, a file system management module AI 213, a device management module AI 214, a battery management module AI 215, as well as a security access control module AI 216, an energy and resource optimization module AI 217, and a network and communication management module AI 218, which constitute an operating system kernel for executing advanced AI such as AGI and ASI.

[0098] (Security Access Control Module AI216) The security access control module AI216 plays an essential role in ensuring the security of the entire system. This security access control module AI216 controls the resources that users and processes can access through access control list management, blocking unauthorized access. It also has an anomaly detection function that monitors communication patterns and operations inside and outside the system to detect abnormal behavior. This makes it possible to detect and respond to attempts at unauthorized access and unauthorized data movement in real time. It also has encryption and authentication functions that encrypt system data communications and strengthen security. It also performs regular vulnerability management to maintain a high level of system security at all times.

[0099] The security access control module AI 216 is included in or linked to multiple functional modules AI. The security access control module AI 216 acquires multimodal information, which is perception actively obtained by the user (active perception), for example, via the input device 11, the camera 12, the microphone 13, the sensor 14, or various other sensors (not shown). The security access control module AI 216 embeds contextual information, including at least the time, location, and event tag at which the multiple functional modules AI functioned, and a selected emotional salience score into the acquired multimodal information, thereby concealing the data. The security access control module AI 216 may, for example, partially or entirely embed or conceal the contextual information and the emotional salience score into the acquired multimodal information using known data embedding or concealment techniques.

[0100] Furthermore, the security access control module AI 216 may autonomously update the algorithms and internal models of the multiple module AIs. The security access control module AI 216 references a predetermined database, for example, to check whether process information is set for each functional module AI, and obtains information on all important OS-level decisions and state transitions, including self-improvement processes, within the target module. The security access control module AI 216 sequentially records the obtained information in, for example, a distributed ledger on the blockchain-based distributed ledger server 500. This allows each of the multiple module AIs to ensure the update, traceability, and auditability of their respective algorithms and internal models.

[0101] (Energy and Resource Optimization Module AI217) The energy and resource optimization module AI217 minimizes energy consumption throughout the system and promotes efficient resource utilization. This energy and resource optimization module AI217 monitors energy consumption throughout the system in real time and detects abnormal energy consumption. It optimizes energy consumption by dynamically adjusting resources (CPU, memory, disk I / O, etc.) according to the system operating status and turning off unused resources.

[0102] The system may also have a function to automatically apply a power saving mode depending on the battery status and the supply status of an external power source. Furthermore, as described above, the energy and resource optimization module AI217 monitors the internal cognitive states of multiple module AIs in real time, and the coordinator AI210 optimizes the resource management policy of the entire system based on the internal cognitive states of the module AIs detected by the energy and resource optimization module AI217. This maximizes the energy efficiency of the system and enables sustainable operation.

[0103] (Network and Communications Management Module AI218) The network and communications management module AI218 is responsible for efficiently and securely managing communications both inside and outside the system. This network and communications management module AI218 monitors network bandwidth usage in real time and routes traffic efficiently to avoid communications bottlenecks. It also achieves efficient data transfer by automatically selecting the optimal communications protocol depending on the communication status and environment. It also provides a function to encrypt communications and protect communications inside and outside the system. This module supports remote system management and monitoring, and also enables system updates and diagnostics via the network.

[0104] Although it does not provide the basic functions of a kernel, by adding a high-level task management module AI219, an ethics and compliance management module AI220, a self-learning and self-evolution module AI221, and a social interface module AI222, it can be made into an operating system for running advanced AI such as AGI and ASI.

[0105] (High-Level Task Management Module AI219) The high-level task management module AI219 manages complex, multi-stage tasks and ensures that the entire system operates efficiently toward long-term goals. This high-level task management module AI219 manages tasks hierarchically based on priority and importance, and adjusts resource allocation so that important tasks receive priority. It also sets system goals and manages progress in real time, analyzing the cause if a goal is not achieved and making necessary adjustments. It also divides complex tasks, schedules them to be executed in the appropriate order, and optimizes parallel processing. This enables efficient resource allocation and improves the flexibility and efficiency of the entire system.

[0106] (Ethics and Compliance Management Module AI220) The ethics and compliance management module AI220 is provided to ensure that AGI and ASI behave ethically and legally appropriately. This ethics and compliance management module AI220 constantly monitors whether the system is operating in accordance with pre-defined ethical standards and frameworks, and ensures that the system's behavior is appropriate. Specifically, it performs behavioral simulations that take into account ethical judgment criteria and priorities, preventing illegal acts and ethical issues in advance. It also performs compliance checks to ensure strict compliance with laws and regulations, and ensures that all actions and decisions comply with laws and regulations in accordance with the legal environment in which the system operates. Furthermore, the system simulates expected behavior, and if ethical or legal issues are detected, it is equipped with a function to issue warnings and restrict system operation.

[0107] (Self-learning and self-evolution module AI221) The self-learning and self-evolution module AI221 is a module that allows the system to self-learn while in operation and achieve self-improvement according to the environment and usage conditions. It collects daily operational data and uses machine learning algorithms to learn from that data. This allows it to identify system behavior patterns and optimization points and evolve the model. It also creates a feedback loop and reflects feedback from users and other modules in its learning, improving the accuracy and efficiency of the entire system. It also has a self-evaluation function, allowing the system to evaluate its own performance and make autonomous improvements as needed.

[0108] (Social Interface Module AI222) The social interface module AI222 is a module for realizing natural interactions with humans. This social interface module AI222 analyzes user input via text or voice and generates appropriate responses. This includes semantic analysis and context understanding, making interactions with users more human-like and intuitive. Furthermore, it recognizes the user's emotions and adjusts the content and tone of the response to provide more appropriate and empathetic interactions. This module also has the ability to understand the social rules and context within a conversation and select appropriate actions and statements, allowing the system to provide socially appropriate responses.

[0109] (Embodiment 3: User emotion information response application 223) Here, an overview of the user emotion information response application 223 will be described with reference to Figs. 2 and 7. The user emotion information response application 223 provides learning support to learners (users). Specifically, the user emotion information response application 223 functions on, for example, a terminal 50, and is connected to the network NW, the information processing device 100, and the large-scale language model system 200.

[0110] The terminal 50 is, for example, a personal computer (PC) owned by a learner (user), who is the target person, and may be portable or fixed. The terminal 50 is configured to be able to communicate with the information processing device 100 via a network NW. The information processing device 100 is configured to be able to communicate with the large-scale language model system 200 via the network NW. Communication can be either wireless or wired. The user is learning programming using the terminal 50. A learning screen 51 for learning programming is displayed on the terminal 50, and the learning screen 51 shows, for example, a case in which a character 52 is displayed.

[0111] The information processing device 100 communicates with a large-scale language model system 200 to receive content for increasing a learner's motivation to learn programming. The large-scale language model (LLM) system 200 is a natural language processing system that performs question and answer sessions. The LLM system 200 is a natural language processing model trained using a large amount of text data, and receives and outputs sentences as input. When the LLM system 200 is applied to a system that performs question and answer sessions, when a question is queried to the LLM system 200, answer content is output. In the user emotion information response application 223, the information processing device 100, for example, queries the LLM system 200 about questions related to learning and transmits answer content from the LLM system 200 to the terminal 50.

[0112] The terminal 50 receives answer content (messages) related to learning from the LLM system 200. The terminal 50 outputs the messages related to the user's learning received on the learning screen 51 of the terminal 50 during programming learning by voice output or by displaying a speech bubble or the like. As an example, the terminal 50 may use a character 52 to output the messages.

[0113] The configuration of the user emotion information response application 223 will be described with reference to Fig. 2. First, the terminal 50 includes a control unit 10, a camera 12, a communication I / F 17, an input device 11, an output device 15, a microphone 13, a storage device 16 as a storage unit, and an internal bus connecting each of the units.

[0114] The control unit 10 includes a CPU (Central Processing Unit), RAM (Random Access Memory), and ROM (Read Only Memory). The control unit 10 executes the application program to generate a learning screen for programming learning and displays it on a display device such as a monitor. The input device 11 includes a mouse and a keyboard. The camera 12 (sensor 14) acquires image data of the user's facial expression as biometric data of the user (subject). In this example, image data of the user's facial expression, etc. is described, but other information may be acquired as biometric data. For example, an infrared camera may be used to acquire information about the user's pulse wave. The microphone 13 acquires user voice data as biometric data of the user. The sensor 14 is any of various known sensors connected to the terminal 50. The output device 15 includes a display device such as a monitor, a speaker, etc. The storage device 16 includes various application programs, etc., for the user to learn programming, for example. The communication I / F 17 is connected to the network NW, the information processing device 100, the LLM system 200, and the distributed ledger server 500, and performs data exchange with external devices.

[0115] The terminal 50 acquires biometric data (for example, image data and voice data) and transmits it to the information processing device 100 via the communication I / F 17. The terminal 50 acquires input data and transmits it to the information processing device 100 via the communication I / F 17. The input data in this example is, for example, input data from a keyboard, mouse, etc. that a user uses when learning programming.

[0116] The information processing device 100 includes a control unit 10, a memory unit (storage device 16), a communication I / F 17, and an internal bus connecting each unit. The control unit 10 includes a CPU, RAM, and ROM. The communication I / F 17 is connected to a network NW and transmits and receives data to and from external devices. The memory unit includes various application programs, etc. For example, the memory unit stores application programs, etc. for supporting the user's learning. The control unit 10 executes the application programs to realize various processes. The information processing device 100 receives biometric data, etc. transmitted from the terminal 50, and generates emotional information about the user based on the biometric data, etc. The information processing device 100 queries the LLM system 200 about questions related to learning that correspond to the generated emotional information.

[0117] The LLM system 200 includes a question receiving unit 201, an answer generating unit 202, and a trained model 204. The question receiving unit 201 receives a question inquiry related to learning from the information processing device 100. The answer generating unit 202 analyzes the received inquiry and, based on the analysis result, uses the trained model 204 to generate answer content (a message, as an example) related to learning in response to the question inquiry and transmits it to the information processing device 100. The trained model 204 is a learning model trained using a text database (DB) or the like having a large amount of text data for generating an answer.

[0118] The information processing device 100 receives answer content (e.g., a message) related to learning from the LLM system 200 and transmits it to the terminal 50. The terminal 50 outputs the received content (e.g., a message) related to learning on the learning screen 51 of the terminal 50 during programming learning. As an example, the terminal 50 may output the answer content (message) using a character 52 provided on the learning screen 51. In this example, the LLM system 200 is described as being provided separately from the information processing device 100. However, it is also possible to provide functions equivalent to those of the LLM system 200 within the information processing device 100. Alternatively, the LLM system 200 may perform the functions of generating emotional information and generating content. Furthermore, a configuration in which the functions of the information processing device 100 and the LLM system 200 are integrated into the terminal 50 may be adopted. The terminal 50 itself may generate emotional information for the user and also generate answer content (e.g., a message) related to learning, and output the content directly to the learning screen 51 of the terminal 50. This configuration eliminates the need to communicate with external systems over a network, shortening response times and improving processing efficiency on terminal 50, and also strengthening user privacy protection since the user's emotional data and other personal information is processed without being transmitted outside the terminal.

[0119] FIG. 8 is a diagram illustrating the functional block configuration of the information processing device 100 that implements the user emotion information response application 223. Referring to FIG. 8 , the control unit 10 of the information processing device 100 realizes various functional blocks by executing application programs stored in the storage unit. Specifically, the information processing device 100 includes a biometric data acquisition unit 110, an emotion analysis unit 112, an input data acquisition unit 114, a prompt generation unit 116, an input state analysis unit 117, an output control unit 118, an integration unit 120, a content acquisition unit 122, and a recording unit 124. The biometric data acquisition unit 110 acquires biometric data (image data, audio data, etc.) transmitted from the terminal 50 via the communication I / F 17. The acquired biometric data is stored in the storage unit.

[0120] The emotion analysis unit 112 generates emotion information by estimating the user's emotion (psychology) based on the acquired biometric data (image data, audio data, etc.). The emotion analysis unit 112 estimates (analyzes) emotions such as joy, anger, sadness, and fun based on the biometric data. The emotion analysis unit 112 may estimate emotions such as calmness, surprise, satisfaction, boredom, disappointment, fear, relief, and anxiety, without being limited to these. Specifically, the emotion analysis unit 112 can estimate the emotion by processing the acquired image data and the user's facial expressions, such as eyelid opening, gaze, eyebrow movement, presence or absence of nose wrinkles, mouth movement, mouth opening, and pupil opening. The emotion analysis unit 112 can also estimate (analyze) the emotion using audio data, including the content of the user's voice, sighs, breathing sounds, etc. Note that the emotion analysis unit 112 may estimate the emotion using only one type of data, or may estimate the emotion using a combination of data. Furthermore, in this example, a case has been described in which biometric data (image data) from the camera 12 is used, but the present invention is not limited to this, and biometric data may be acquired using a wearable device.

[0121] The input data acquisition unit 114 acquires input data (input key data, etc.) transmitted from the terminal 50 via the communication I / F 17. The acquired input data is stored in the storage unit. The input state analysis unit 117 analyzes the user's input state based on the acquired input data (input key data, etc.). Specifically, the input state analysis unit 117 estimates the user's input state by analyzing the user's input speed using a keyboard, etc., the number of input errors, etc. The content acquisition unit 122 acquires content (a message, for example) previously output to the user via the output control unit 118, which is stored in the storage unit based on the emotion information generated by the emotion analysis unit 112 (first content).

[0122] The integration unit 120 integrates the emotion information generated by the emotion analysis unit 112 with the first content acquired by the content acquisition unit 122. The prompt generation unit 116 generates a sentence to ask the LLM system 200 a question related to learning based on the information integrated by the integration unit 120. The prompt generation unit 116 outputs the generated sentence to the LLM system 200. The output control unit 118 outputs information such as answer content from the LLM system 200 to the terminal 50. The recording unit 124 associates the answer content (second content) from the LLM system 200 with the emotion information and records it in the recording unit 124.

[0123] (Fourth Embodiment: User-Interaction-Compatible Application 224) Here, with reference to FIGS. 2 and 11 , an overview of the user-interaction-compatible application 224 will be described. The user-interaction-compatible application 224 cooperates with a learner (user) and multiple AIs (such as the coordinator AI 210). Specifically, the user-interaction-compatible application 224 functions on, for example, a terminal 50 and connects to the network NW, the information processing device 100, and the LLM system 200. The terminal 50 is, for example, a personal computer (PC) owned by the target learner (user), and may be portable or fixed. The terminal 50 may be equipped with input devices such as a microphone and a camera, various generative AI models compatible with multimodality, a device for recording information, sensors for acquiring information such as position, temperature, tilt, and acceleration, a device for outputting information, etc., and is used for dialogue between the user and the AI.

[0124] The control unit 10 includes a processing unit (CPU, GPU, NPU), RAM, and ROM. The input device 11 includes general input by the user via touch, mouse, or keyboard. The camera 12 acquires the user's biometric data and image data to be input to the AI. Without being limited to this, for example, an infrared camera may be used to acquire the user's environmental information. The microphone 13 acquires the user's voice data and the user's environmental information. The sensor 14 is a known sensor connected to the terminal 50. The output device 15 includes an information display device such as a display and a speaker. The storage device 16 stores various application programs, various generated AI models, profiles such as user hobbies and preferences, records of interactions between the user and the AI, and content such as music, images, and videos. The communication I / F 17 communicates with external devices.

[0125] The user interaction application 224 allows the user to interact with the AI ​​without going through a network, shortening response times and improving processing efficiency on the terminal 50, and also strengthening user privacy protection since the user's emotional data and other personal information is processed without being transmitted outside the terminal.

[0126] 11 shows the configuration of a terminal 50 on which the user interaction-enabled application 224 functions. Referring to Fig. 11, the terminal 50 on which the user interaction-enabled application 224 functions includes an input information integration unit 310, a response generation unit 320, a recording unit 330, and an output unit 340. The input information integration unit 310 includes an input information acquisition unit 311, a user profile acquisition unit 312, and an interaction record acquisition unit 313. The response generation unit 320 includes a user emotion estimation unit 321, a dynamic interaction navigation analysis unit 322, a personalized response generation unit 323, and a learning / evolution unit 324.

[0127] Specifically, the input information integration unit 310 acquires user input from the camera, microphone, touch (keyboard) operation, etc. of the terminal 50 using the input information acquisition unit 311, acquires content recorded in the storage device 16, and acquires environmental information from various sensors. Furthermore, the user profile acquisition unit 312 acquires the user profile recorded in the storage device 16 from the recording unit 330, and the dialogue record acquisition unit 313 acquires the most recent dialogue record stored in the storage device 16 from the recording unit 330, analyzes the dialogue content, extracts dialogue records suitable for the dialogue content from the storage device 16, and acquires them via the recording unit 330 (input content).

[0128] The response generation unit 320 is a multimodal AI system in which individual generative AI models or functions are individually modularized and integrated into one. The response generation unit 320 may be a single generative AI model or a group of multiple models trained with specialized knowledge. In this case, it may be possible to select which model to use to generate a response based on the content of the dialogue. This allows each model to be lightweight and compact, making it possible to run it on a terminal with limited resources. Furthermore, by connecting to a network via the communication I / F 17, it is possible to search for information on the Internet and include it in the generation of a response.

[0129] Furthermore, if there is information that the AI ​​has not learned during a dialogue between the user and the target AI, the user dialogue application 224 may communicate with the AI ​​of another device to generate a response regarding the information requested by the user. In this case, to prevent the user's personal information or the content of the dialogue from leaking to an external network, information, data, etc. related to the user's privacy may be deleted or masked. Furthermore, if it is necessary to request information from an external AI (for example, the content of the dialogue or some personal information needs to be sent), the user may be notified in advance and permission may be obtained.

[0130] The user emotion analysis unit 321 analyzes input content, such as text acquired by the input information integration unit 310, voice tone acquired from a microphone, facial expressions acquired from a camera, a user profile, and past conversations, to acquire user characteristic information, such as the user's emotional state, strength, psychological state, psychological tendency, psychological pattern, and behavioral pattern. Specifically, emotions can be estimated by analyzing text content, the user's voice, sighs, breathing sounds, and the user's facial expressions, such as eyelid opening, gaze, eyebrow movement, presence or absence of nose wrinkles, mouth movement, mouth opening, and pupil dilation. The emotions and psychological states that can be estimated include joy, anger, sadness, joy, calmness, surprise, satisfaction, boredom, disappointment, fear, relief, and anxiety. While the emotion analysis unit 321 can estimate emotions using only one piece of information, combining multiple pieces of information allows for more accurate estimation of the user's emotions. Furthermore, emotions may be estimated by acquiring biometric data using a wearable device. The dynamic dialogue navigation analysis unit 322 may use natural language processing technology to extract the intention contained in the user's words from the input content, identify the user's needs, and obtain information about the intention and background of the dialogue that will determine the direction of the dialogue.

[0131] Specifically, the dynamic dialogue navigation analyzer 322 performs semantic analysis of text and audio data, integrating language patterns, contextual clues, and the user's previous dialogue to identify the user's unique intent. In the initial analysis stage, the input data is tokenized and relevant entities and phrases are extracted to extract the basic elements of the user's question or request.

[0132] Next, deep learning models or other methods may be used to extract high-level contextual relevance from these elements and map them to specific intents and needs. This stage evaluates not only the meaning of the user's words but also how they relate to previous interactions and the user's profile. Depending on the understanding of intents and needs, the dynamic interaction navigation analyzer 322 may derive an optimal interaction route from the user's current request and previous behavioral patterns.

[0133] The personalized response generation unit 323 generates information and suggestions customized to the user's needs based on the analyzed direction of the dialogue, and further adjusts the response using personalized data obtained from the user's past dialogue history. The personalized response generation unit 323 determines the momentum of the dialogue from the strength of the user's emotions, and generates an appropriate response based on the determination result.

[0134] The learning and evolution unit 324 collects feedback from users on responses to improve the quality of responses, and performs AI learning at appropriate times at the end of a dialogue, for example, during periods when the system is inactive.

[0135] The recording unit 330 acquires and stores data such as a user profile obtained in advance from a questionnaire or the like when starting to use the system, records of the content of conversations, audio, images, and video from the storage device 16 of the terminal 50.

[0136] The output unit 340 may render responses to the user as text, images, or videos on a display, output them through a speaker, or project them onto a wall. Furthermore, by outputting information to control the robot's motors, it is possible to operate the robot, and responses may be expressed in sign language to hearing-impaired users through the robot's movements. In this case, by using AI that has learned sign language and acquiring the user's sign language expressions as input from a camera, even hearing-impaired people can interact with the AI.

[0137] 6 is a flow diagram illustrating the execution and processing of the operating system 21 according to the first embodiment, as well as a flow diagram illustrating the execution processing of the operating system 21 for executing advanced AI such as AGI or ASI according to the second embodiment. Note that the execution steps of the coordinator AI210, memory management module AI211, process management module AI212, file system management module AI213, device management module AI214, and battery management module AI215 described in the first embodiment are omitted to avoid duplication.

[0138] (First Embodiment) The first embodiment is based on a general computer configuration.

[0139] <<Power-on (Step S0)>> The power is turned on.

[0140] <<BIOS / UEFI Execution (Step S1)>> The BIOS / UEFI starts up and performs system initialization and hardware initialization. The BIOS / UEFI collects system hardware configuration information and loads the OS boot loader into memory. The BIOS / UEFI loads the boot loader into memory and executes it. The boot loader loads the operating system 21 into system memory and places the coordinator AI 210 and other module AIs that make up the kernel in memory. The coordinator AI 210 starts the initialization process based on that information.

[0141] <<Coordinator AI Execution (Step S2)>> The coordinator AI 210 starts up, acquires hardware information provided by the BIOS / UEFI and boot loader, and begins control of the entire system. The coordinator AI 210 accesses the memory, CPU, storage devices, etc., performs their initial settings, and directly accesses hardware registers via the system bus as necessary to perform initial settings and allocate resources.

[0142] <<Execution of Security Access Control Module AI (Step S3)>> As soon as the coordinator AI210 starts up and the basic components of the system are initialized, the security access control module AI216 is started. The security access control module AI216 controls access to resources based on access control lists (ACLs) and security policies. The security access control module AI216 uses interfaces provided by the memory management module AI211 and the process management module AI212 to check the access rights of users and processes. The security access control module AI216 collects security information in real time through system logs and security event monitoring functions, and takes immediate action if an abnormality is detected.

[0143] <<Execution of Energy and Resource Optimization Module AI (Step S4)>> Next, once the security and access control module AI216 has completed its startup, the energy and resource optimization module AI217 is launched. The energy and resource optimization module AI217 is responsible for optimizing system resources and efficiently managing power consumption. Specifically, it acquires information from sensors and power management units (PMUs) within the system in real time and adjusts resource allocation and power consumption based on this information. This access is performed using standardized protocols such as ACPI (Advanced Configuration and Power Interface) and smart battery systems. Based on this information, the energy and resource optimization module AI217 performs operations such as changing device power modes and shutting down unused resources as necessary. This minimizes system power consumption and improves energy efficiency.

[0144] <<Battery Management Module AI Execution (Step S5)>> The battery management module AI 215 starts up and accesses the battery controller to monitor the battery status in real time. This includes monitoring the charge level, temperature, and health state. The battery management module AI 215 works in conjunction with the coordinator AI 210 and other module AIs to implement power saving modes and optimize charging. The battery management control is performed by the battery management module AI 215 directly accessing the battery controller and following the programmed energy management algorithm.

[0145] <<Execution of Memory Management Module AI (Step S6)>> The memory management module AI211 is started and memory management begins. The memory management module AI211 accesses the system memory and creates an overall memory map. The memory management module AI211 accesses the physical memory via the memory controller and sets and allocates virtual memory. The memory management module AI211 directly accesses and controls the memory, as well as using memory management routines programmed in the kernel.

[0146] <<Execution of Process Management Module AI (Step S7)>> The process management module AI212 starts up and begins managing processes within the system. The process management module AI212 creates a process table and manages information about each process. The process management module AI212 accesses information about each process via the system bus and controls the creation, scheduling, and termination of processes, as well as allocating resources and prioritizing processes according to a scheduling algorithm programmed by the AI.

[0147] <<Execution of High-Level Task Management Module AI (Step S8)>> After the process management module AI212 is started and basic task management within the system is established, the high-level task management module AI219 is started. The high-level task management module AI219 manages complex, multi-stage tasks and supports the system in achieving its long-term goals. The high-level task management module AI219 sets task priorities and optimizes task scheduling based on task information and the status of system resources provided by the coordinator AI210. Access to information is via the process management module AI212 and memory management module AI211, and the execution status and resource usage status of each process are monitored in real time. This enables efficient processing of tasks across the entire system and optimal resource allocation.

[0148] <<Executing File System Management Module AI (Step S9)>> The file system management module AI213 starts up and mounts the root file system. The file system management module AI213 directly accesses the storage device and performs file system consistency checks and mounting operations. The file system management module AI213 performs control using programmed routines related to file access. The file system management module AI213 continuously monitors the status of the file system and repairs or reconfigures it as necessary.

[0149] <<Executing Device Management Module AI (Step S10)>> The device management module AI 214 starts up and initializes all hardware devices in the system. The device management module AI 214 accesses each device, loads the respective drivers, and enables appropriate control of the operation of each device.

[0150] <<Execution of Network and Communication Management Module AI (Step S11)>> When the device management module AI214 initializes the network interface, the network and communication management module AI218 is launched. The network and communication management module AI218 performs the management required for the system to communicate with external devices safely and efficiently. Specifically, it manages communications with network interface cards (NICs) and routers and monitors network bandwidth usage in real time. This access is performed through a standard network protocol stack (e.g., a TCP / IP stack), optimizing communication traffic and dynamically switching communication protocols. This eliminates communication bottlenecks and improves the efficiency of the system's interaction with external networks.

[0151] <<System Service Execution (Step S12)>> System service execution involves sequential activation of system monitoring, log management, security management, etc. Each service accesses necessary information through the system bus and processes various events. These services are executed by the coordinator AI 210 using programmed monitoring and management routines to ensure the stability and security of the entire system.

[0152] <<Execution of Ethics and Compliance Management Module AI (Step S13)>> After the system service is started, the ethics and compliance management module AI220 is started. The ethics and compliance management module AI220 plays a role in verifying and maintaining that the operation of the entire system is ethically and legally appropriate. The ethics and compliance management module AI220 applies pre-set ethical standards and compliance requirements to the data processed and tasks performed by the system, and monitors whether these are being observed. Specifically, it analyzes system operation logs and user interaction data and issues warnings as necessary. Furthermore, if a violation is detected, it restricts system operation. Access to information is via APIs and interfaces provided by the system service, and the operation of the entire system is monitored in real time.

[0153] <<Execution of Self-Learning and Self-Evolution Module AI (Step S14)>> After the high-level task management module AI219 is launched, the self-learning and self-evolution module AI221 is launched. The self-learning and self-evolution module AI221 performs self-learning based on data obtained during system operation and plays a role in continuously improving system performance. Specifically, it collects data on processes and user interactions within the system and analyzes it using machine learning algorithms. This information is obtained through APIs provided by the coordinator AI210 and other modules AI. The self-learning and self-evolution module AI221 identifies system optimization points based on the collected data and automatically adjusts system settings as necessary. This allows the system to continue evolving while adapting to changes in the environment and usage conditions.

[0154] <<Executing the Social Interface Module AI (Step S15)>> Finally, just before the terminal program is launched, the social interface module AI222 is launched. The social interface module AI222 provides functionality for realizing natural interaction with the user. This module analyzes user input and manages text and voice communication. Data from input devices (keyboard, microphone, etc.) is acquired through the coordinator AI210, and the social interface module AI222 analyzes and generates responses. The social interface module AI222 properly handles user interactions and ensures that the system is user-friendly.

[0155] <<Executing the Terminal Program (Step S16)>> The terminal program starts and provides an interface through which the user can input commands. The terminal program works with the coordinator AI 210 and each module AI to appropriately process tasks instructed by the user. The terminal program uses interface logic programmed into the coordinator AI 210 to analyze the user input and transmit appropriate instructions to each module AI.

[0156] As such, the coordinator AI 210 is typically responsible for managing and controlling the entire system, issuing instructions to each module AI and adjusting resource allocation. However, there are situations where module AIs need to directly collaborate and exchange information. This improves the system's real-time capabilities and efficiency, allowing specific processes to be performed more quickly and appropriately. For example, the security and access control module AI 216 and the network and communication management module AI 218 may collaborate. This is to encrypt communication data and perform access control in real time when security regarding network communications is important. Specifically, when an anomaly in network traffic is detected, the security and access control module AI 216 immediately sends feedback to the network and communication management module AI 218, instructing it to change the communication protocol or block specific communications as necessary.

[0157] In some cases, the energy and resource optimization module AI217 and the battery management module AI215 work together. In particular, if the battery status changes rapidly, the battery management module AI215 may directly provide data on the remaining battery level and usage trends to the energy and resource optimization module AI217, which may then use that data to optimize energy allocation. This is important for streamlining system power consumption and maximizing the operating time of battery-powered devices.

[0158] Furthermore, the high-level task management module AI219 and the process management module AI212 may operate in cooperation with each other. This is because, when complex task scheduling and management are required, the high-level task management module AI219 can directly issue instructions to the process management module AI212 regarding process prioritization and schedule adjustment, thereby improving system efficiency. This cooperation allows for better management of tasks across the system and minimizes resource waste.

[0159] Furthermore, the self-learning and self-evolving module AI221 cooperates with other modules AI to collect information. For example, the self-learning and self-evolving module AI221 can learn energy consumption patterns based on data from the energy and resource optimization module AI217 and use the results to improve energy management of the entire system. In this way, the self-learning and self-evolving module AI221 can directly collect data and feedback from other modules and use it for learning, thereby continuously improving the performance of the entire system.

[0160] In this way, the module AIs can directly exchange information and cooperate with each other, improving the real-time performance and efficiency of the entire system and enabling module AIs with specialized knowledge to work together to perform optimal processing. This allows the system to meet the requirements of users and the system while responding to advanced processing demands.

[0161] In constructing an AI-based operating system (A / OS), the selection of its underlying model is a crucial factor. The selection of this underlying model significantly influences the AI's processing power, the efficiency of resource management, and the flexibility and scalability of the system. Below, we will examine the underlying models for A / OS and consider optimal combinations. We will also examine the underlying models for operating systems that run advanced AI, particularly AGI (artificial general intelligence) and ASI (artificial superintelligence), and consider optimal combinations.

[0162] First, when considering the underlying model of an AI / OS, a comparison can be made between a monolithic kernel and a microkernel. A monolithic kernel has the advantage of achieving high performance and extremely fast inter-process communication because all kernel services are built into a single, large kernel. However, its disadvantage is that a crash of one component increases the likelihood that the entire system will be affected. On the other hand, a microkernel is an approach that minimizes kernel functionality and provides the remaining functions as user-space services. This promotes modularization and clearly separates each service, improving system flexibility and security. Because AI / OS requires system modularization and flexibility, a microkernel is considered suitable. This allows each AI module to operate independently, which is expected to improve the reliability of the entire system.

[0163] Next, we compare distributed systems with centralized systems. In distributed systems, resources and processes are distributed across multiple machines and operate in coordination, resulting in high scalability and availability, and improved fault tolerance. On the other hand, centralized systems manage all resources and processes on a single machine, which has the advantage of simple management and low communication overhead, but has limited scalability and a higher risk in the event of a failure. In AI / OS, flexibility and scalability are important, so distributed systems are suitable. This allows multiple AIs to process in coordination, which is expected to improve the performance and fault tolerance of the entire system.

[0164] Furthermore, when comparing batch processing and real-time processing as data processing models, batch processing processes data at fixed time intervals, allowing for efficient processing of large volumes of data, but has the disadvantage of low real-time performance. In contrast, real-time processing processes data as soon as it is generated, allowing for high real-time performance and time-critical processing, but it requires high processing costs and complex resource management. AI / OS requires user interaction and real-time system control, making the real-time processing model more suitable. This makes it possible to achieve instant feedback and advanced interaction.

[0165] Operating systems for running advanced AI such as AGI and ASI require greater scalability and distributed processing. Because AGI and ASI require enormous computing power and data processing capabilities, a distributed microkernel architecture is the ideal infrastructure model to support them. In this architecture, each AI module operates independently in a distributed environment while communicating via a network. This creates a flexible and scalable system, maximizing the performance of the entire system.

[0166] Furthermore, AGI and ASI require self-learning and adaptability. To achieve this, a flexible architecture is required that can instantly reflect learned data and models in the system. By adopting an architecture based on reinforcement learning, each module AI within the system can self-learn and evolve to take optimal actions in response to the environment and requirements. This enables AGI and ASI to continuously update their knowledge and skills and provide optimal performance.

[0167] Furthermore, security and safety are essential for AGI and ASI. This includes ensuring the security of communications between modules and defending against external attacks. By adopting a secure distributed kernel architecture, communications between each AI module are encrypted and access control is strictly implemented. This design allows the entire system to maintain its functionality even if some modules are attacked.

[0168] Therefore, a distributed microkernel architecture is ideal for an operating system that runs AGI or ASI. By integrating this with reinforcement learning-based self-learning capabilities and advanced security, a flexible and secure system can be built. This combination will enable an operating system that meets the requirements of future advanced AI systems and continues to evolve.

[0169] Furthermore, when selecting the underlying model for the coordinator AI 210 and each module AI, factors such as processing efficiency, scalability and flexibility, real-time performance, learning ability and adaptability, and safety and security must be considered. When selecting the optimal underlying model for each AI module based on these factors, a combination of deep neural networks (DNN) and reinforcement learning (RL) is optimal for the coordinator AI 210. This allows for efficient resource management and task scheduling for the entire system, while also providing the ability to self-learn while adapting to the environment.

[0170] The optimal base model can also be considered for each module AI. A deep neural network is suitable for the memory management module AI211, and a combination of a graph neural network (GNN) and reinforcement learning is suitable for the process management module AI212. A generative model (GAN / VAE) is optimal for the file system management module AI213 and device management module AI214. A graph neural network is suitable for the network / communication management module AI218, reinforcement learning is suitable for the battery management module AI215, and a transformer model is suitable for the security management module AI216 and the social interface module AI222. Adopting these base models enables the AI / OS to function flexibly and efficiently, making it possible to build a powerful system that can meet the advanced requirements of AGI and ASI in particular.

[0171] This AI operating system (A / OS) can be applied to a wide range of computers, from computers used in ordinary homes to commercial servers and even computers installed in robots. The A / OS design emphasizes flexibility and scalability, and is built to be adaptable to different hardware environments and uses.

[0172] In personal computers used in the home, AI / OS can improve the user experience by optimizing the interface and resource management according to user needs. AI capabilities can be used to streamline everyday tasks, enhance security, and provide personalized user support. For example, AI can learn user usage patterns, optimize performance, and dynamically allocate required resources to provide a comfortable operating environment.

[0173] For business servers, the advanced resource management functions and scalability of AI / OS play an important role. In server environments, efficient resource allocation and load balancing are required because many users and services access the system simultaneously. AI / OS is based on a distributed system, and provides high availability and fault tolerance by adjusting resources across multiple servers. In addition, security management using AI strengthens defenses against cyber attacks and unauthorized access.

[0174] Furthermore, A / OS can also be applied to computers installed in robots. Because robots require real-time data processing and quick decision-making, the real-time processing capabilities of A / OS are particularly useful. To adapt to the environment and situation in which a robot operates, A / OS utilizes self-learning and adaptive functions to help the robot select optimal actions. It also maximizes robot performance by efficiently processing sensor information and control data to ensure smooth operation.

[0175] These characteristics give A / OS versatility that allows it to flexibly adapt to a variety of applications and devices, and it can be introduced in a wide range of areas, from home computers to commercial servers and robot control systems. A / OS, with its system scalability, security, real-time processing capabilities, and self-learning capabilities, can function as an important foundation in the next-generation computer environment.

[0176] The AI / OS, which is composed of a coordinator AI 210 and multiple module AIs, is an excellent architecture that enhances the system's flexibility, scalability, and reliability. However, it can be further improved by introducing a distributed coordination model, enhancing the self-optimization capabilities of the module AIs, or utilizing blockchain technology.

[0177] Specifically, in the current design, the coordinator AI 210 centrally manages the entire system, but this poses the risk of becoming a single point of failure. To avoid this, one possible approach is to deploy multiple coordinator AIs 210 in a distributed manner, each responsible for a specific group of module AIs, or to introduce a "distributed coordination model" in which multiple coordinator AIs 210 cooperate to manage the entire system. This approach improves the redundancy and reliability of the system, making it less likely that the entire system will be affected if a specific coordinator AI 210 goes down.

[0178] Furthermore, each module AI has the ability to self-optimize and adapt independently to the environment and tasks, further improving the performance of the entire system while reducing the load on the coordinator AI 210. For example, the self-learning and self-evolving module AI 221 can be applied to other module AIs, allowing each module to learn and evolve independently, thereby improving the efficiency and adaptability of the entire system.

[0179] Furthermore, by introducing blockchain technology into communication and data management between AI modules, security and reliability will be further enhanced. In particular, in a distributed system, blockchain can be used to guarantee data integrity, making interactions between each AI module transparent and tamper-proof.

[0180] <<Execution of user emotion information response application 223 (step S17)>> Figure 9 is a flow diagram illustrating the details of the processing of the user emotion information response application 223 (step S17: steps S200 to S300). As shown in Figure 9, the information processing device 100 on which the user emotion information response application 223 is implemented acquires biometric data (step S200). Specifically, the biometric data acquisition unit 110 acquires biometric data transmitted from the terminal 50.

[0181] Next, the information processing device 100 determines whether a predetermined period has elapsed (step S210). If the information processing device 100 determines that the predetermined period has not elapsed (NO in step S210), it returns to step S200 and repeats the above process. The predetermined period can be set to any period, and can be set to five minutes as an example. If the information processing device 100 determines that the predetermined period has elapsed (YES in step S210), it proceeds to the next process.

[0182] The information processing device 100 executes a process of analyzing emotions (step S220). Specifically, the emotion analysis unit 112 generates emotion information by estimating the user's emotions (psychology) based on the biometric data (image data, audio data, etc.) stored in the storage unit 107 after a predetermined period of time has elapsed. Note that it may use some or all of the biometric data (image data, audio data, etc.) stored in the storage unit 107 within the predetermined period of time.

[0183] Next, the information processing device 100 acquires a first content based on the emotion information generated by the emotion analysis unit 112 (step S230). Specifically, the content acquisition unit 122 acquires content (first content) that was previously output via the output control unit 118 and stored in the storage unit 107 based on the emotion information generated by the emotion analysis unit 112. The content acquisition unit 122 acquires content that was previously output via the output control unit 118 and stored in the storage unit 107. For example, the content acquisition unit 122 acquires content (e.g., information about a previously output message) that was recorded in association with the same emotion information as the emotion information stored in the storage unit 107.

[0184] Next, the information processing device 100 integrates the emotion information with the acquired content (first content) (step S240). Specifically, the integration unit 120 integrates the emotion information with the acquired content (first content). Furthermore, if the content acquisition unit 122 acquires multiple pieces of content, the integration unit 120 may integrate all of the pieces of content, or may select and integrate one of the pieces of content. Alternatively, the integration unit 120 may combine multiple pieces of content as appropriate, or may integrate information that has been appropriately processed and edited by extracting only necessary parts.

[0185] Next, the information processing device 100 generates a prompt to be sent to the LLM system 200 (step S250). Specifically, the prompt generation unit 116 generates a sentence to ask the LLM system 200 a question related to the learning based on the information integrated by the integration unit 120. Next, the information processing device 100 sends the generated prompt to the LLM system 200 (step S260). Specifically, the prompt generation unit 116 sends the generated prompt, which is a sentence to ask the LLM system 200.

[0186] Next, the information processing device 100 determines whether or not answer content has been received from the LLM system 200 (step S270). Specifically, the output control unit 118 determines whether or not answer content has been received from the LLM system 200. In step S270, the information processing device 100 maintains that state until answer content is received from the LLM system 200.

[0187] When the information processing device 100 receives the answer content from the LLM system 200 (YES in step S270), the information processing device 100 outputs information such as the answer content (a message as an example) (second content) to the terminal 50 (step S280).

[0188] The output control unit 118 transmits message output information including the message to the terminal 50. When the output control unit 118 receives a plurality of answer contents from the LLM system 200, the output control unit 118 may select one of them. When the output control unit 118 receives a plurality of answer contents, the output control unit 118 may combine them as appropriate, or extract only the necessary parts and process and edit them as appropriate to generate a message that is answer content.

[0189] Next, the information processing device 100 associates the emotion information with the answer content (second content) and records them (step S290). Specifically, the recording unit 124 associates the emotion information with the answer content (a message, for example) and stores them in the storage unit 107. The recording unit 124 may also associate other related information with the answer content (a message, for example) and store them in the storage unit 107.

[0190] Next, the information processing device 100 determines whether or not to end the process (step S300). If the information processing device 100 determines in step S300 that the process should be ended, the information processing device 100 ends the process (END). Specifically, if the information processing device 100 determines that the user has ended the application program for learning programming on the terminal 50, the information processing device 100 ends the process.

[0191] Information processing device 100 may end the process when it receives a command to end the application program for learning programming from terminal 50. On the other hand, in step S300, if information processing device 100 determines not to end the process (NO in step S300), it returns to step S200 and repeats the above process.

[0192] 10A and 10B are diagrams illustrating examples of prompts and answer content in the user emotion information response application 223. FIG. 10A shows a prompt that the prompt generation unit 116 transmits to the LLM system 200 based on the information integrated by the integration unit 120. As an example, the prompt generation unit 116 generates the following prompt based on the information integrated by the integration unit 120. Specifically, the content acquisition unit 122 acquires first content based on the emotion information ("fun") generated by the emotion analysis unit 112. In this example, the content acquisition unit 122 acquires previously output answer content (e.g., a message) (first content) that is recorded in association with the emotion information ("fun"). The integration unit 120 integrates the emotion information ("fun") with the information of the previously output message.

[0193] Based on the information integrated by the integration unit 120, the prompt generation unit 116 generates the following sentence as an example: "(A) Think of something to say to a child who is in the following emotional state that will increase the child's motivation to learn. When speaking to them, use words that are appropriate for children. They are learning programming. (B) They find learning fun. (C) There are some things they don't understand and they are a little frustrated. (D) They feel it would be frustrating to give up."

[0194] Furthermore, the prompt generation unit 116 generates a sentence using, as reference information, the acquired content (for example, a message) associated with the emotional information ("fun") and stored in the storage unit 107. As an example, the prompt generation unit 116 generates a sentence such as "(E) has always been good at learning new things, and he has the strength to overcome even small difficulties. I'm sure he will be able to do well on this assignment in the While sentence."

[0195] 10(B) shows an example of response content from the LLM system 200 in response to the above prompt. As an example, the information processing device 100 receives response content such as, "Great! Programming is fun, isn't it? You did very well on the previous assignment with the While statement. But sometimes there are difficult parts, so it can be a little frustrating. Even if you don't understand something, it's a chance to gradually absorb it! Everyone had a lot of things they didn't understand at first. Let's work through those difficult parts together! Then, new and interesting things might be waiting for you!"

[0196] The information processing device 100 transmits the received answer content information to the terminal 50. The terminal 50 outputs a learning screen for programming learning and a message related to the user's learning, which is the answer content transmitted from the information processing device 100. By this processing, the information processing system analyzes the user's emotional state and adds information about previously output messages as reference information, thereby making it possible to output to the user an empathetic message that is more in line with the user's situation and takes into account the content of previously output messages.

[0197] Conventionally, each user interaction is completed with each session, and the content of previous interactions is not reflected, which means that there is a problem in outputting empathetic messages that take into account the user's past situations. However, the user emotion information response application 223 outputs messages that take into account the content of messages output in the past. Furthermore, the content of previously output messages recorded in association with emotion information is added as reference information, so that messages that are more appropriate to the emotion are output to the user. In other words, the information processing system can appropriately grasp the user's state and efficiently support the user. For example, it is possible to improve the user's learning effectiveness.

[0198] Furthermore, the user emotion information compatible application 223 can grasp the user's emotional state (joy, anger, sadness, fun, etc.) to understand the user's psychological readiness and readiness to learn at that time, and can provide a message that suits the user's emotions at the moment when the user is most receptive, thereby enabling efficient support to the user.

[0199] It can also improve the effectiveness of a user's learning. For example, when negative emotions such as frustration or anxiety are detected early on in a user's learning, appropriate support and intervention can be provided before the user gives up by outputting an appropriate message that matches that emotion. It can also increase a user's engagement in learning by outputting an appropriate message that matches that emotion when the user is in a positive emotional state.

[0200] Furthermore, by understanding the user's emotional state, a more personalized learning experience can be provided. For example, if a user is feeling sad or stressed, the learning process can be optimized and the learning efficiency can be improved by encouraging easier tasks or taking breaks. Learning in an emotionally balanced state can help with memory retention and deeper understanding.

[0201] <<<<Example of Prompt Generation>>> In the above, the prompt generation unit 116 generates a prompt based on the emotional information (“fun”), which is the information integrated by the integration unit 120, and the first content (the message associated with “fun” and recorded in the memory unit 107).

[0202] The prompt generation unit 116 can similarly generate prompts for information integrated by the integration unit 120 that includes other emotional information. For example, the prompt generation unit 116 generates a prompt based on emotional information ("anger"). For example, the prompt generation unit 116 may change sentence (B) of the above sentences (A) to (E), which are prompts, to "I feel angry about studying."

[0203] The prompt generation unit 116 may also fix the sentence (A) and change the sentences (B) to (D) as appropriate based on the emotional information. For example, the prompt generation unit 116 generates a sentence using, as reference information, the acquired content that has been associated with the emotional information ("anger") and stored in the storage unit 107. For example, the prompt generation unit 116 generates a sentence such as, "(E) You may not be good at this assignment with a while statement, but you did very well on the assignment with a for statement. So I'm sure you'll do well this time too."

[0204] In response to the prompt, the information processing device 100 receives, for example, answer content such as "Anger is an emotion that arises when you learn something new. If you take one step at a time, the anger will gradually disappear. Remember the time when you were doing the for sentence assignment. Don't miss your growth." The information processing device 100 transmits information such as the received answer content to the terminal 50.

[0205] The prompt generation unit 116 generates a prompt based on the emotional information ("sad (pitiful)"). For example, the prompt generation unit 116 may change the sentence (B) of the above sentences (A) to (E) that are prompts to "I feel sad about studying." Alternatively, the prompt generation unit 116 may keep the sentence (A) fixed, and change the sentences (B) to (D) as appropriate based on the emotional information.

[0206] Furthermore, as one example, the prompt generation unit 116 generates a sentence using, as reference information, the acquired content that has been associated with emotional information ("sadness (sorrow)") and stored in the storage unit 107. As one example, the prompt generation unit 116 generates a sentence such as "(E) You worked really hard and patiently on both this While sentence assignment and the sentence assignment. So, don't give up this time either, do your best."

[0207] In response to the prompt, the information processing device 100 receives, for example, an answer content such as "Think of sad (sorrowful) feelings as a step towards growth. It's natural to have things you don't understand, like challenges with while clauses and for clauses, but it's important to have the attitude of trying your best to overcome them." The information processing device 100 transmits information such as the received answer content to the terminal 50.

[0208] The prompt generation unit 116 generates a prompt based on emotional information ("joy"). For example, the prompt generation unit 116 may change sentence (B) of the above sentences (A) to (E), which are prompts, to "I feel joy in learning." Alternatively, the prompt generation unit 116 may keep sentence (A) fixed, and change sentences (B) to (D) as appropriate based on emotional information.

[0209] Furthermore, as an example, the prompt generation unit 116 generates a sentence using, as reference information, the acquired content that has been associated with the emotional information ("joy") and stored in the storage unit 107. As an example, the prompt generation unit 116 generates a sentence such as "(E), you really enjoyed working on the assignments with the "While" and "For" statements. Keep it up."

[0210] In response to the prompt, the information processing device 100 receives, for example, answer content such as "Your motivation for learning is great. That was clearly evident in both the while sentence assignments and the for sentence assignments. By progressing with a sense of joy, you can continue to grow." The information processing device 100 transmits information such as the received answer content to the terminal 50.

[0211] <<<<Generation of Emotion Information by Emotion Analysis Unit 112>>> The emotion analysis unit 112 may generate emotion information by estimating the user's emotion (psychology) based on input data without relying on biometric data. Specifically, the emotion information may be generated based on input data (input key data, etc.) acquired by the input data acquisition unit 114. For example, an emotion diagnostic test may be executed at the start of programming learning to diagnose the user's emotion based on their input.

[0212] The emotion analysis unit 112 may generate emotion information based on user input data for multiple questions used to diagnose the user's emotions in the emotion diagnostic test. The emotion diagnostic test may be conducted in a manner that is not particularly limited, and may prompt the user to select questions or images in a manner that is unlikely to be noticed, and emotion information may be generated based on the answers. The input data may be any data that can be used to generate emotion information, and is not limited to input key data from a keyboard or the like, but may also be, for example, user behavior history data.

[0213] Furthermore, the emotion analysis unit 112 is not limited to directly using input data, and may generate emotion information using, for example, user input state information analyzed by the input state analysis unit 117. For example, the emotion analysis unit 112 may analyze the state of actual programming input during programming learning and generate emotion information based on the input state information that is the analysis result. Alternatively, emotion information may be generated based on the analysis result of the content or context of the user's input data as input state information.

[0214] 12 is a flowchart related to the process of generating a response using dead reckoning in the terminal 50 on which the user interaction-compatible application 224 functions. An interaction starts when the terminal 50 recognizes an event such as powering on the terminal 50 or a sensor response. When the interaction starts, the user profile acquisition unit 112 of the input information integration unit 310 acquires a user profile from the recording unit 130 (step S300).

[0215] Here, the user profile is created by asking questions and taking tests to learn about the user's personality, character, and psychological patterns when the user first starts using the system, analyzing the answers to those questions, and recording them as a profile in the storage device 16 (recording unit 130).Furthermore, if the content of the dialogue with the user is analyzed and information about the user's personality is found, this information can be added to the profile and recorded, making it possible to generate more appropriate responses to the user.

[0216] Next, the dialogue record acquisition unit 313 of the input information integration unit 310 acquires past dialogue records between the AI ​​and the user (step S310). In addition to the most recent dialogue record, the dialogue record to be acquired is also acquired by analyzing the current dialogue content and extracting dialogue records that match the dialogue content. Since the dialogue records increase as the number of dialogues increases, by recording the dialogues and adding index information such as the date and time of the dialogue, emotions, and a summary of the dialogue to the dialogue content, it becomes possible to efficiently extract dialogue records that match the current dialogue content.

[0217] Next, the input information acquisition unit 311 of the input information integration unit 310 acquires input information that the user inputs to the terminal 50 (step S320). Here, in addition to input information from the camera, microphone, and various sensors of the terminal 50, content such as audio, images, and videos recorded by the user and stored in the recording unit is also acquired as input information.

[0218] As a result, the information obtained by the input information integration unit 310, which is composed of one or a combination of the user profile, the dialogue record, and the input information, is analyzed by the response generation unit 320 as multimodal input content.

[0219] The user emotion analysis unit 321 of the response generation unit 320 analyzes the input content acquired by the input information integration unit 310 to analyze the user's emotions (step S330). Based on input from the camera and microphone of the terminal 50 as well as input from various sensors, a user profile, dialogue records, and other input information, the analysis acquires user characteristic information that serves as the user's behavioral principles, such as the user's current emotional state and the strength of that emotion, the user's psychological state, and psychological patterns.

[0220] Next, an intention / needs analysis is performed (step S340). The response generation unit 320 includes a dynamic dialogue navigation analysis unit 322 (described later). The response generation unit 320 analyzes the motivations and objectives behind the user's words using, for example, known natural language processing (NLP) techniques, and obtains information about the intention and background of the analyzed dialogue. For example, when a user asks a question about a specific product, the response generation unit 320 infers what information the user is seeking from past purchase history and related questions, and generates a response to provide information that meets those needs.

[0221] In step S340, the response generator 320 predicts the next step the user is likely to take and helps the dialogue proceed smoothly. This allows the response generator 320 to not only provide or convey information, but also provide knowledge to advance the user's needs and increase satisfaction.

[0222] The response generation unit 320, in the personalized response generation unit 323, combines multimodal input content such as input information from the user, user profile, dialogue records, and various sensor information with user characteristic information obtained from the user emotion analysis process and information on intentions and background obtained from the intention / needs analysis (step S340), and generates a response optimized for each individual user (step S350).

[0223] The response generation unit 320 analyzes the questions and comments entered by the user, analyzes past trends and preferences based on the user profile, evaluates what topics the user has been interested in in the past and what responses they preferred, and applies that knowledge to the current dialogue.

[0224] The information generated by the response generator 320 is used to understand the user's current emotional state and to generate an appropriate response in the user emotion analysis step S330. For example, if the response generator 320 analyzes that the user is depressed, it generates a response that uses more careful expressions or shows empathy.

[0225] The results of the intent / needs analysis step S340 are used to understand what the user is looking for and what information would help them, which may include relevant product information, troubleshooting steps, or recommended actions. This analysis enables the user interaction application 224 to predict what information the user will seek next and generate responses to facilitate the interaction.

[0226] The personalized response generator 323 generates a response that satisfies the user's needs and provides a positive interaction experience by combining all the elements generated in the response generation step S350. The response is adjusted based on information learned from the user's past interactions, the user's current psychological state, and the purpose and goals of the interaction, and provides support to the user with appropriate suggestions and questions.

[0227] The output unit 140 then outputs the generated response (step S360). The output format depends on the format of the generated information. Specific examples include displaying text, images, videos, and the like on a display, or projecting the information onto a wall using a projector function. The output unit 140 may also output audio via a speaker, or in the form of a signal for controlling the robot's movements.

[0228] The output unit 140 obtains feedback from the user in response to the response generated in step S360 (step S370). The feedback is the user's reply to the generated response. Furthermore, for example, the user interaction application 224 may apply response generation using dead reckoning to various service fields.

[0229] The feedback may be, for example, "direct feedback" such as "customer support (conducting a satisfaction survey after an interaction and asking the user for a numerical rating or free-form opinion), educational application (assessing learning outcomes with quizzes or tests after learning and analyzing the extent to which the user has deepened their understanding of the content), entertainment (encouraging feedback after playing games or watching on a streaming service to help improve the accuracy of content recommendations)," and "indirect feedback" such as "e-commerce (using reviews and ratings after purchases as feedback to improve the accuracy of product recommendations), health care app (estimating the effectiveness of a user's health improvement progress and activity data to provide personalized health advice), automatic feedback, smart home devices (automatically detecting a user's daily usage patterns and environmental changes to help automatically adjust settings and improve services), automobiles (analyzing a driver's driving style and conditions inside and outside the car to provide safer and more comfortable driving assistance), and social feedback (social media: analyzing user posts and reactions to identify trends or suggest content tailored to individual users)."

[0230] The learning and evolution unit 324 of the response generation unit 320 then analyzes the user's feedback (step S380). The learning and evolution unit 324 applies the concept of dead reckoning to the dialogue context based on the feedback, allowing the AI ​​to self-improve and evolve. Specifically, the learning and evolution unit 324 collects not only direct user feedback from completed dialogue sessions, but also implicit feedback from the user's actions and choices, such as response times, selected dialogue options, and changes in the user's emotions during the dialogue. The data collected by the learning and evolution unit 324 is processed as learning material for the user dialogue application 224 to gain a deeper understanding of the user's dialogue style and preferences.

[0231] The user interaction application 224 then analyzes the collected data using dead reckoning to identify patterns from past interaction records and current feedback to predict future response strategies. For example, if a user consistently responds to a particular phrase or keyword, the system can recognize this pattern and generate improved responses in similar situations. Additionally, if feedback analysis reveals a change in the user's emotions, the system may intentionally delay the response. Specifically, the user interaction application 224 dynamically sets a response delay based on the user's emotional state.

[0232] In the user interaction application 224, for example, if the user expresses stress or anger, the system may be configured to delay the response for a preset period of time (e.g., 2-5 seconds). This delay may be implemented through a response timer in the user interface, temporarily holding off an AI-generated response before sending it to the user. This time lag may be configured to an appropriate length based on the user's current emotional state, for example, through updates to the emotion recognition algorithm. Specifically, this time lag is determined by an emotional state classifier that evaluates the user's emotions in real time and identifies the intensity and type of emotion (e.g., anger, sadness, joy, etc.).

[0233] A timing adjustment algorithm may then calculate the appropriate response timing based on the type and intensity of the emotion. This algorithm may set or adjust a longer lag time for more intense emotions and a shorter lag time for more neutral emotions. A response scheduler may then receive input from the timing adjustment algorithm and manage the actual sending of the response.

[0234] The scheduler operates in sync with the rest of the system, providing a timely response while maintaining a natural dialogue flow for the user. This delay may express empathy for the user and, if the user is emotionally upset, give them time to calm down, allowing a more thoughtful response to be delivered later. This approach allows the AI ​​to achieve more human-like dialogue and respond in a way that is in tune with the user's emotions.

[0235] Furthermore, the learning and evolution unit 124 fine-tunes the AI ​​model based on the feedback, continuously making small adjustments to improve the naturalness and appropriateness of the dialogue, including updating the intent recognition model and response generation algorithm. The ultimate goal of the feedback analysis process is to continuously improve the user's experience during the dialogue, resulting in a more human-like, emotionally empathetic AI dialogue. Through this process, the AI ​​can adapt to emerging user behaviors, grow over time, and better personalize the dialogue with the user.

[0236] In the record update step S390, after the series of processes are completed, the recording unit 330 updates the user profile and dialogue record (step S390). Information about the user's personality newly obtained as a result of the analysis is added to or modified in the user profile. In addition, an index is assigned to each new dialogue between the AI ​​and the user and added to the dialogue record. This makes it easy to extract and obtain from the past dialogue history those that are highly relevant to the current dialogue.

[0237] In the execution of the user-interactive application (step S18), dead reckoning is applied to the user's input to generate a response. The various flows in the execution of the user-interactive application (step S18) are not limited to the order shown. Furthermore, this series of steps can essentially be controlled by a program, but may also be self-controlled by AI. Some advanced AI systems allow machine learning models to learn from data and automatically improve their performance over time, or to fine-tune the model using user feedback. The various processing steps are self-controlled, and the AI ​​may adjust parameters independently to generate more effective responses. Furthermore, AI systems have the ability to make more complex decisions based on accumulated data and experience, enabling them to address new scenarios and unknown problems beyond the initial program design.

[0238] FIG. 13 shows an example of a dialogue in which a response is generated using dead reckoning in the user dialogue application 224. Referring to FIG. 13, the processing of a response generated by AI by applying dead reckoning to a user profile and user input will be described. Note that while only a portion of the user profile is shown in this example, more information may be collected during dialogue with the user to create a profile. Specifically, a large amount of information about the user's personality is obtained and recorded, including the user's gender and age, occupation, hobbies, topics of interest, personality, values, beliefs, knowledge level, lifestyle, dialogue preferences, decision-making style, risk response, stress tolerance, sources of motivation, learning style, interpersonal style, emotional fluctuations, reaction to the environment, and tendencies based on past behavior.

[0239] Learning by the AI ​​makes it possible to generate responses tailored to the user. Next, we will explain the process by which the AI ​​applies dead reckoning to generate a response to a user's travel-related input based on the user's profile. In this explanation, we will compare the case where the AI ​​analyzes the user's intentions and generates a response with the case where dead reckoning is not applied.

[0240] <<<<Processing (1) in User Interaction-Enabled Application 224>>> (Dialogue A) Example of user input: "Hello, I'm planning a trip to Europe next month. It's my first time, so I'd like some advice." Example of AI response: "Hello! A trip to Europe sounds great. Which countries do you plan to visit?" Analysis of user input and extraction of intent: The AI ​​analyzes the user's input using natural language processing. At this stage, the keywords "travel to Europe" and "first time" are captured, making it clear that this is the user's first visit to Europe and that they are seeking some kind of advice. Because the user says, "It's my first time, so I'd like some advice," the AI ​​infers from this expression that the user may be feeling some anxiety or anticipation. Example of generating a response based on intent: The AI ​​generates a response by asking about the user's specific destination, taking into account that the user is planning a trip and that it is their first visit. This prepares the AI ​​to provide specific advice tailored to the countries the user will visit. By asking, "Which countries do you plan to visit?", the AI ​​aims to elicit more specific information from the user. The AI ​​also responds in a friendly tone and demonstrates empathy for the user's emotions, helping the user feel at ease as the conversation progresses. ・Ensuring continuity and appropriateness of the conversation: This response shows interest in the user while paving the way for more detailed advice based on additional information the user provides. This is a strategic choice that naturally advances the conversation and is designed to make it easy for the user to deepen the topic.

[0241] <<<<Example of AI response when dead reckoning is not applied (1)>>> ・Example of AI response: "Traveling to Europe sounds fun. There are so many options." In this case, the AI's response is general and does not deeply address the user's specific needs or the context of their first trip. This response misses the opportunity to find out what information the user is looking for, which can reduce the quality of the interaction.

[0242] <<<<Processing (2) in User Interaction-Enabled Application 224>>> Dialogue B - Example of user input: "I'm thinking of France and Italy. Paris and Rome both have many attractions I want to see." - Example of AI response: "That's a great choice. Paris is famous for the Eiffel Tower and the Louvre, and Rome has the Colosseum and the Vatican Museums. What are you particularly interested in?" - Analyzing user input and extracting intent: The AI ​​analyzes the user's input using natural language processing. At this stage, the specific locations "France," "Italy," "Paris," and "Rome" are extracted. This reveals that the user is interested in the many attractions they want to see in these cities. Furthermore, if the user says, "Both have many attractions I want to see," the AI ​​infers that the user is excited and excited. - Example of generating a response based on intent: Because the user names specific cities, the AI ​​suggests famous tourist spots related to each city. The Eiffel Tower and Louvre Museum in Paris, the Colosseum and Vatican Museums in Rome, are the most iconic landmarks in their respective cities and are generally of high interest to tourists. The AI ​​uses this information to further refine the conversation by suggesting specific places that the user might be interested in, continuing to pique the user's interest. Furthermore, the AI ​​uses empathetic language, such as "That would be a great choice," to demonstrate empathy with the user's positive emotions. Deepening the conversation and exploring user interests: The AI ​​further explores the user's interest in specific types of tourist attractions and activities by asking, "What would you be particularly interested in?" This approach allows for additional information to be elicited from the user, enabling more personalized information to be provided.

[0243] <<<<Example of AI response when dead reckoning is not applied (2)>>> Example of AI response: "That will be fun. There are many beautiful places in France and Italy. Have fun." - Analyzing user input and extracting intent: In this response, the AI ​​does not ask the user what specific places they are interested in, but only provides general encouragement, missing an opportunity to deepen the dialogue.

[0244] <<<<Processing (3) in User Interaction-Enabled Application 224>>> Dialogue C - Example of user input: "I really want to see historical buildings. I also want to enjoy the local cuisine." - Example of AI response: "In that case, I recommend enjoying a croissant at a small cafe in the Marais district in Paris, and authentic Italian cuisine in Trastevere district in Rome. What is your budget?" - Example of analyzing user input and extracting intent: The AI ​​uses natural language processing to extract two key points from the user's input: "historical buildings" and "local cuisine." This indicates that the user is interested in cultural tourist attractions and local food culture. Furthermore, because the user says, "I really want to see historical buildings. I also want to enjoy the local cuisine," the AI ​​infers that the user has interest and expectations. - Example of generating a response based on intent: The AI ​​demonstrates empathy and provides specific recommendations that address the elements the user is looking forward to. Based on the user's interests, the AI ​​then suggests specific areas in Paris and Rome. The Marais and Trastevere districts are known for their cultural charm and local cuisine, respectively, and recommendations are tailored to the user's needs. The selection of restaurants and cafes here is based on the unique experiences each district offers, and identifies the best places for the user to enjoy the "local cuisine" they are looking for. This makes appropriate suggestions that meet the user's expectations and increases their satisfaction. - Example of continuing the dialogue and gathering further information: By asking, "What is your budget?", the AI ​​seeks additional information to further personalize the response. This allows for further recommendations based on the user's financial constraints, providing more useful information to the user. - Example of AI response when dead reckoning is not applied:

[0245] <<<<Example of AI response when dead reckoning is not applied (3)>>> ・Example of AI response: "Historical buildings and local cuisine can be enjoyed in any city. Have fun." ・Example of intent-based response generation: The response in this case is generic and lacks customized information based on the user's specific interests or geographical details.

[0246] <<<<Processing (4) in User Interaction-Enabled Application 224>>> Dialogue D Example of user input: "Hmm, I'd prefer it not to be too expensive. But I want to be comfortable, so I want to spend moderately." Example of AI response: "I understand. I can provide you with information on cost-effective accommodation options and reasonably priced, reputable restaurants. I can also help you make reservations." Example of analyzing user input and extracting intent: Using natural language processing technology, the AI ​​captures two key points from the user's words: "Not too expensive" and "I want to be comfortable." These phrases reveal the user's need to be mindful of their budget while still seeking a certain level of comfort. Furthermore, from the user's statement, "Hmm, I'd prefer it not to be too expensive. But I want to be comfortable, so I want to spend moderately," the AI ​​emotionally understands that the user has some concerns about the cost of travel, but also wants to ensure comfort. - Generating responses based on intent: To address user expectations and concerns, the AI ​​chooses to provide information in a way that balances budget constraints and the need for comfort. To this end, it helps users find the balance between quality and price by suggesting "cost-effective accommodation options" and "reasonable, reputable restaurants." Furthermore, it aims to increase the ease and convenience of planning by offering support for reservation arrangements. In this way, the AI ​​generates appropriate responses that are in line with the user's emotions and needs. - Concreteizing the dialogue and strengthening user assistance: The AI's responses indicate that it will provide specific services (providing information on accommodation options and restaurants, and supporting reservation arrangements), making it easier for users to take specific actions. This is expected to enable users to effectively plan their trip and further deepen their trust with the AI. <<<<Example of AI response when dead reckoning is not applied (4)>>> Example of AI response: "Europe is generally expensive, but you may be able to save some money by booking early." - Analyzing user input and extracting intent: This response does not provide a direct solution or suggestion to the user's specific needs (balancing comfort and budget).We also believe that there is a lack of concrete support for making reservations.

[0247] In the user interaction application 224, the information processing device 100 employs dead reckoning to generate responses, enabling the AI ​​to behave as if it is actively interested in the user's input. This allows the AI ​​to learn from the user's past interactions and utilize that information in the current interaction. This generates personalized responses based on the user's input, making the AI ​​appear to be "actively" responding to the user's needs and interests. For example, if the user asks a question about travel, the AI ​​can provide specific suggestions based on the user's travel style and preferences learned from past interactions. The AI's behavior of showing interest in the user's input does not simply involve generating reactive responses, but also includes asking further questions to the user to elicit more information. This allows for a deeper understanding of the user's intentions and needs, improving the quality of the interaction.

[0248] Furthermore, the user interaction application 224 allows the user to feel a sense of familiarity with the AI, similar to how a user would feel with a person. The AI ​​remembers the user's past actions and preferences and customizes responses based on them, allowing the user to feel that the AI ​​"understands" them. This allows the user to perceive the AI ​​as a conversational partner rather than just a machine, creating a sense of familiarity in personalized interactions. This allows people to naturally feel a sense of familiarity with those who understand and take an interest in them, and the AI ​​learns from past interactions and provides responses tailored to each individual user, building trust and creating a sense of familiarity with the user.

[0249] Furthermore, the user interaction application 224 facilitates interaction between the AI ​​and humans, allowing the AI ​​to provide highly relevant responses to the user. This encourages the user to further the interaction, and if the responses appropriately address the user's interests and needs, the user is more likely to ask new questions or seek more detailed information. This invigorates the interaction with the AI, providing a richer and more fulfilling interaction experience. Furthermore, it also increases user engagement and increases trust and reliance on the AI. By having the AI ​​actively show interest and continually seek dialogue, the user may feel that the AI ​​is sincerely listening to and trying to understand what they are saying, and may become more actively involved in the interaction.

[0250] Furthermore, the AI ​​may be trained to have empathetic communication skills so that the user interaction application 224 can apply dead reckoning to make the interaction with the user more natural and effective. Specifically, the AI ​​may be trained in psychology and active listening techniques. Active listening is the practice of listening carefully to what the other person is saying and empathizing with their feelings and thoughts to achieve effective communication. The AI ​​can carefully listen to what the user is saying, summarize the content, and confirm it, thereby demonstrating its interest in what the user is saying.

[0251] For example, the user interaction application 224 can be applied to services in various fields that utilize AI, leveraging a user's past behavior and needs to provide more personalized, beneficial services to the user. Specifically, in the field of education, AI typically provides feedback based on a student's current performance and can provide individualized instruction based on the student's past learning history and interests. To this end, AI can predict changes in a student's understanding and interests and suggest optimal learning materials and teaching methods. This enables education to be tailored to a student's understanding and interests, improving learning outcomes.

[0252] In the entertainment field, for example, it will be possible to recommend content based on a user's viewing history and preferences, and by predicting changes in preferences and interests based on the user's past viewing history and conversation flow, it will be possible to provide accurate recommendations and interactive dialogue, thereby personalizing the user's entertainment experience and increasing satisfaction.

[0253] In the healthcare field, for example, AI can provide advice based on a patient's current symptoms and data. By applying dead reckoning, AI can predict future health risks and needs based on a patient's medical history and current health status, and provide appropriate self-care advice and reminders. This can make patient health management more effective and improve the quality of preventative care.

[0254] In the retail and e-commerce fields, for example, it will be possible to recommend products based on a user's purchase history. This makes it possible to predict future purchasing intentions and preferences based on a user's purchase history and interests, and provide highly relevant products and promotions. This is expected to increase users' purchasing motivation and encourage repeat purchases.

[0255] In the transportation and travel sector, for example, it will be possible to create plans based on the user's current schedule and preferences, and to predict future travel destinations and activities based on the user's past travel history and preferences, thereby suggesting optimal destinations and transportation methods, thereby improving the user's travel experience and increasing satisfaction.

[0256] In the smart home field, for example, AI typically operates home appliances based on the resident's current behavioral patterns, but it can also predict the resident's future behavior and needs based on the resident's past behavioral patterns and preferences, providing more appropriate home appliance operation and home automation. This can improve the comfort and convenience of the resident and increase the value of the smart home.

[0257] Furthermore, while traditional customer support AI responds reactively based on the user's current inquiry, it can provide optimal advice and responses by referencing the user's past inquiry history and problems, predicting the user's inquiry content and the intention behind it, and proposing appropriate solutions. This allows users to have a consistent support experience and shortens the time it takes to resolve their issues.

[0258] Although the present invention has been specifically described above based on the embodiments, it is not limited to these embodiments and can be modified in various ways without departing from the spirit of the invention.

[0259] 1 Physical layer 10 Control unit 11 Input device 12 Camera 13 Microphone 14 Sensor 15 Output device 16 Storage device 17 Communication I / F 100 Information processing device 110 Biometric data acquisition unit 112 Emotion analysis unit 114 Input data acquisition unit 116 Prompt generation unit 117 Input state analysis unit 118 Output control unit 120 Integration unit 122 Content acquisition unit 124 Recording unit 2 Intermediate layer 21 Operating system 22 Device driver 200 LLM system (large-scale language model system) 201 Question receiving unit 202 Answer generation unit 204 Trained model 210 Coordinator AI 211 Memory management module AI 212 Process management module AI 213 File system management module AI 214 Device management module AI 215 Battery management module AI 216 Security and access control module AI 217 Energy and resource optimization module AI 218 Network and communication management module AI 219 High-level task management module AI 220 Ethics and compliance management module AI 221 Self-learning and self-evolution module AI 222 Social interface module AI 223 User emotion information response application 224 User dialogue response application 3Application layer 310 Input information integration unit 311 Input information acquisition unit 312 User profile acquisition unit 313 Dialogue record acquisition unit 320 Response generation unit 321 User emotion analysis unit 322 Dynamic dialogue navigation analysis unit 323 Personalized response generation unit 324 Learning and evolution unit 330 Recording unit 340 Output unit 50 Terminal 51 Learning screen 52 Character 500 Distributed ledger server

Claims

1. An information processing device comprising an operating system executed on a computer, comprising: a coordinator AI that monitors the operating status of the entire system and controls the operation of multiple functional modules in an integrated manner; and multiple modules AI that analyze the usage status of system resources for each function instructed by the coordinator AI and set allocations or usage restrictions of the system resources based on the analysis results, wherein the coordinator AI obtains the analysis results from the multiple modules AI and dynamically changes the allocation or usage restrictions of the system resources based on the obtained analysis results.

2. The information processing device according to claim 1, characterized in that the plurality of modules AI include at least one functional module AI selected from the group consisting of a memory management module AI, a process management module AI, a file system management module AI, a device management module AI, and a battery management module AI.

3. The information processing device according to claim 1, characterized in that the plurality of modules AI include at least one functional module AI selected from the group consisting of a security access control module AI, an energy resource optimization module AI, and a network and communication management module AI.

4. The information processing device of claim 1, wherein the plurality of modules AI includes at least one functional module AI selected from the group consisting of a high-level task management module AI, an ethics and compliance management module AI, a self-learning and self-evolution module AI, and a social interface module AI.

5. The information processing device according to claim 1, characterized in that the coordinator AI optimizes the resource management policy of the entire system based on the internal cognitive state of the energy / resource optimization module AI recognized by the module AI.

6. The information processing device described in claim 1, characterized in that the multiple functional module AIs calculate an emotional salience score based on the user's emotions or the degree of goal achievement, and transmit it to the coordinator AI, and the coordinator AI optimizes process priority, memory allocation, and network bandwidth in real time according to the emotional salience score.

7. The information processing device of claim 1, wherein the plurality of module AIs maximizes the sensor recording parameters via the coordinator AI when the emotional salience score exceeds a threshold.

8. The information processing device described in claim 1, characterized in that the multiple functional modules AI include a security and access control module AI, which embeds contextual information including at least time, location, and event tags and emotional salience scores into the data for multimodal information obtained by active perception, thereby concealing the information.

9. The information processing device of claim 8, wherein the security access control module AI autonomously updates the algorithms and internal models of each module AI, and records all important OS-level decisions and state transitions, including self-improvement processes, in a blockchain-based distributed ledger to ensure immutability, traceability, and auditability.

10. The information processing device according to claim 1, further comprising a user emotion information compatible application that generates emotion information based on biometric data or input data of a subject, acquires first content based on the generated emotion information, integrates the emotion information with the first content, generates second content based on the integrated information, and outputs the generated second content, wherein the user emotion information compatible application transmits the generated emotion information to the coordinator AI, and the coordinator AI dynamically changes the allocation or usage restrictions of the system resources based on the emotion information.

11. The information processing device according to claim 1, further comprising a user interaction application that acquires input content, analyzes the acquired input content to acquire information on user characteristics, intentions, and background, generates a response based on the input content, the user characteristics, and the information on intentions and background, and outputs the generated response; the user interaction application transmits the acquired information on user characteristics, intentions, and background to the coordinator AI, and the coordinator AI dynamically changes the allocation or usage restrictions of the system resources based on the information.

12. An information processing method using an operating system executed on a computer, comprising: a step of monitoring the operating status of the entire system and controlling the operation of multiple functional modules in an integrated manner; a step of executing multiple module AIs that analyze the usage status of system resources for each function instructed in the integrated control step and set allocation or usage restrictions of the system resources based on the analysis results; and a step in which the integrated control step further comprises a step of acquiring the analysis results in the step of executing the multiple module AIs and dynamically changing the allocation or usage restrictions of the system resources based on the acquired analysis results.

Images