system

Abstract

Provide a system. 【Solution means】 Information processing means for collecting and centrally managing past legal and security-related data accumulated within the company, Data processing means for preprocessing the acquired data, removing noise data, and standardizing the text, Learning means for retraining based on the preprocessed data using a generated AI model and generating responses using organization-specific knowledge, Risk detection means for analyzing newly acquired contract document data, automatically detecting potential risks, and performing tagging, Result generation means for visualizing the risk detection results and outputting them in the form of a report, Feedback acquisition means for acquiring feedback from the user and using it to improve the accuracy of the AI model, Image processing means for analyzing image data from multiple input devices and performing optical character recognition, Display output means for performing visual display based on the analyzed text data, A system including...

Description

Technical Field

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

Background Art

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

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] In legal and security-related consulting services, it often requires a great deal of time and labor manually, and it is difficult to make quick decisions, which is an issue. Also, when there is no standard procedure to ensure the accuracy of the work, there is a problem in the accuracy and consistency of risk detection. Conventional methods cannot effectively utilize past consultation data, and there is a risk that the same problems will be repeated.

Means for Solving the Problems

[0005] This invention comprises an information processing means for collecting and centrally managing past legal and security-related data accumulated within the company; a data processing means for pre-processing and standardizing the data; a learning means using a generated AI model; a risk detection means for analyzing newly acquired contract documents and automatically detecting risks; a result generation means for visualizing and outputting the detection results; and a feedback acquisition means for obtaining user feedback and using it to improve the accuracy of the model. This enables faster and more accurate operations, as well as efficient problem-solving that leverages the organization's unique knowledge.

[0006] "Information processing means" refers to a system element that has the function of collecting and centrally managing past legal and security-related data accumulated within the company.

[0007] "Data processing means" refers to a system element that has the function of pre-processing acquired legal and security data, removing noise data, and standardizing the text.

[0008] A "learning tool" is a system element that uses a generative AI model to retrain based on pre-processed data and generate responses using organization-specific knowledge.

[0009] A "risk detection tool" is a system element that analyzes newly acquired contract document data, automatically detects potential risks, and tags them.

[0010] "Result generation means" refers to a system element that has the function of visualizing the results of risk detection or outputting the results in a report format.

[0011] A "feedback acquisition method" is a system element that has the function of acquiring feedback from users and using that information to improve the accuracy of the AI ​​model. [Brief explanation of the drawing]

[0012] [Figure 1]This is a conceptual diagram showing an example of the configuration of a data processing system according to the first embodiment. [Figure 2] This is a conceptual diagram showing an example of the essential functions of a data processing device and a smart device according to the first embodiment. [Figure 3] This is a conceptual diagram showing an example of the configuration of a data processing system according to the second embodiment. [Figure 4] This is a conceptual diagram showing an example of the main functions of a data processing device and smart glasses according to the second embodiment. [Figure 5] This is a conceptual diagram showing an example of the configuration of a data processing system according to the third embodiment. [Figure 6] This is a conceptual diagram showing an example of the main functions of a data processing device and a headset-type terminal according to the third embodiment. [Figure 7] This is a conceptual diagram showing an example of the configuration of a data processing system according to the fourth embodiment. [Figure 8] This is a conceptual diagram showing an example of the main functions of a data processing device and a robot according to the fourth embodiment. [Figure 9] This shows an emotion map where multiple emotions are mapped. [Figure 10] This shows an emotion map where multiple emotions are mapped. [Figure 11] This is a sequence diagram showing the processing flow of the data processing system in Example 1. [Figure 12] This is a sequence diagram showing the processing flow of the data processing system in Application Example 1. [Figure 13] This is a sequence diagram showing the processing flow of the data processing system in Example 2, which incorporates an emotion engine. [Figure 14] This is a sequence diagram showing the processing flow of the data processing system in Application Example 2, which combines an emotion engine. [Modes for carrying out the invention]

[0013] Hereinafter, an example of an embodiment of the system relating to the technology of this disclosure will be described with reference to the attached drawings.

[0014] First, the terms used in the following description will be explained.

[0015] In the following embodiments, the labeled processor (hereinafter simply referred to as "processor") may be a single arithmetic unit or a combination of multiple arithmetic units. Also, the processor may be a single type of arithmetic unit or a combination of multiple types of arithmetic units. Examples of arithmetic units include a CPU (Central Processing Unit), a GPU (Graphics Processing Unit), a GPGPU (General-Purpose computing on Graphics Processing Units), an APU (Accelerated Processing Unit), and the like.

[0016] In the following embodiments, the labeled RAM (Random Access Memory) is a memory in which information is temporarily stored and is used as a work memory by the processor.

[0017] In the following embodiments, the labeled storage is one or more non-volatile storage devices that store various programs and various parameters, etc. Examples of non-volatile storage devices include flash memory (SSD (Solid State Drive)), magnetic disks (e.g., hard disks), or magnetic tapes, and the like.

[0018] In the following embodiments, the labeled communication I / F (Interface) is an interface including a communication processor and an antenna, etc. The communication I / F controls communication between multiple computers. Examples of communication standards applied to the communication I / F include wireless communication standards including 5G (5th Generation Mobile Communication System), Wi-Fi (registered trademark), or Bluetooth (registered trademark), and the like.

[0019] In the following embodiments, "A and / or B" is synonymous with "at least one of A and B." That is, "A and / or B" means that it may be A alone, or B alone, or a combination of A and B. Furthermore, in this specification, the same concept as "A and / or B" applies when expressing three or more things linked by "and / or."

[0020] [First Embodiment]

[0021] Figure 1 shows an example of the configuration of the data processing system 10 according to the first embodiment.

[0022] As shown in Figure 1, the data processing system 10 includes a data processing device 12 and a smart device 14. An example of the data processing device 12 is a server.

[0023] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0024] The smart device 14 comprises a computer 36, a reception device 38, an output device 40, a camera 42, and a communication interface 44. The computer 36 comprises a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The reception device 38, output device 40, and camera 42 are also connected to the bus 52.

[0025] The reception device 38 is equipped with a touch panel 38A and a microphone 38B, etc., and receives user input. The touch panel 38A receives user input by detecting contact with an object (e.g., a pen or finger). The microphone 38B receives user input by detecting the user's voice. The control unit 46A transmits data indicating the user input received by the touch panel 38A and microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the data indicating the user input.

[0026] The output device 40 includes a display 40A and a speaker 40B, and presents data to the user 20 by outputting the data in a form perceptible to the user 20 (e.g., audio and / or text). The display 40A displays visible information such as text and images according to instructions from the processor 46. The speaker 40B outputs audio according to instructions from the processor 46. The camera 42 is a small digital camera equipped with an optical system such as a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor.

[0027] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various types of information between processor 46 and processor 28 via network 54.

[0028] Figure 2 shows an example of the main functions of the data processing device 12 and the smart device 14.

[0029] As shown in Figure 2, in the data processing device 12, a specific processing is performed by the processor 28. A specific processing program 56 is stored in the storage 32. The specific processing program 56 is an example of a "program" related to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 according to the specific processing program 56 executed on the RAM 30.

[0030] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0031] In the smart device 14, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The reception output program 60 is used in conjunction with a specific processing program 56 by the data processing system 10. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0032] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0033] This invention is configured as a system for improving the efficiency of legal and security operations. Specific embodiments of this system are described below.

[0034] Data collection and management

[0035] The server collects legal and security-related data accumulated within the company over time. This data includes contracts, past consultation histories, and risk assessment reports. This data is centrally managed in a database.

[0036] Data preprocessing

[0037] The server removes unnecessary noise from the collected data and standardizes the text. This allows the AI ​​model to learn more efficiently. For example, it removes unnecessary document formatting information and extracts pure text data.

[0038] AI model training

[0039] The server uses a generative AI model that learns from organized data. This AI model learns organization-specific legal knowledge and industry-specific risk factors to achieve optimal analytical capabilities.

[0040] Automated contract review and risk detection

[0041] The server uses a trained AI model to automatically review newly submitted contracts. This review process can automatically detect potential risks hidden within the contract and tag them according to their type. For example, it can identify deficiencies in contract terms and potential legal risks.

[0042] Presentation of results and collection of feedback

[0043] The device presents the user with risks and review points detected by the AI. Each risk is accompanied by relevant information and recommended actions, allowing the user to make decisions based on this information. Simultaneously, the user provides feedback on the review results, which is sent to the server for model improvement.

[0044] This system aims to streamline contract review and risk management, supporting users in making quick and accurate decisions. Through its implementation methods, it improves the accuracy and speed of legal and security-related tasks.

[0045] The following describes the processing flow.

[0046] Step 1:

[0047] The server collects historical legal and security-related data from various sources within the company and stores it centrally in a database. This ensures that the necessary data is efficiently accessible.

[0048] Step 2:

[0049] The server performs preprocessing to remove noise from the collected data. Specifically, it standardizes the data by removing unnecessary headers, footers, and email signatures, and by unifying the text format.

[0050] Step 3:

[0051] The server retrains the generative AI model using pre-processed data. In this process, it learns past patterns and risk factors specific to the organization, improving the accuracy and effectiveness of the AI ​​model.

[0052] Step 4:

[0053] The server inputs the text data of the submitted new contract into an AI model for automatic analysis. Here, it detects deficiencies in the contract terms and legal risks, and tags the relevant sections with risk tags.

[0054] Step 5:

[0055] The server compiles the analysis results and uses them to create a detailed risk report. This report includes details of each detected risk and recommended actions for addressing them.

[0056] Step 6:

[0057] The terminal displays the generated risk report to the user. Based on this information, the user can make decisions regarding contract modifications and risk management.

[0058] Step 7:

[0059] Users review the presented report and provide feedback. This feedback is sent to the server and used to retrain the AI ​​model to further improve its accuracy.

[0060] (Example 1)

[0061] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0062] In modern organizations, legal and security-related operations rely on vast amounts of data. While efficiently managing and analyzing this data is crucial, manual processes are time-consuming, costly, and prone to overlooking risks and making misjudgments. A system is needed to address these challenges and improve the accuracy and speed of operations.

[0063] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0064] In this invention, the server includes information processing means for collecting and centrally managing information, data preprocessing means for standardizing acquired data, and learning means for relearning knowledge from the preprocessed data using a generative model. This enables highly accurate and efficient data analysis and automated risk detection compared to conventional manual work.

[0065] "Information processing means" refers to technologies used to effectively collect and centrally manage data gathered within an organization.

[0066] "Data preprocessing means" refers to processes that perform noise reduction and standardization in order to improve the quality of acquired data.

[0067] A "generative model" is an artificial intelligence model that learns from large amounts of data and generates new information or responses.

[0068] "Learning methods" are techniques used to enable models to acquire new knowledge using collected data, thereby improving the accuracy and usefulness of their responses.

[0069] "Analysis means" refers to a process for automatically analyzing newly acquired document data to detect and classify potential risks.

[0070] "Result generation means" refers to technology for visually displaying analyzed data in an easy-to-understand format and outputting it in document format.

[0071] A "feedback mechanism" is a system that collects opinions and evaluations from users and uses them to improve the system or model.

[0072] A "user interface" is a visual interface that users use to interact with a system, obtain information, and perform operations.

[0073] This invention relates to a system aimed at improving the efficiency of legal and security operations. This system operates primarily through a server to effectively collect and manage information.

[0074] The server accesses an internal database to collect legal and security-related data accumulated to date. This data includes contract documents, past consultation history, and risk assessment records. At this stage, database software is used to efficiently centralize and manage the data.

[0075] Next, the server performs data preprocessing. Here, OCR (Optical Character Recognition) software is used to extract text from image files, and the text is de-noised and standardized. This process converts the data into a format suitable for training the generative AI model.

[0076] Subsequently, the server launches a generative AI model and trains it using the formatted data. This process utilizes machine learning frameworks such as Tensorflow® and PyTorch. By learning organization-specific knowledge and industry-specific risk factors, the AI ​​model enables highly accurate data analysis.

[0077] For example, when reviewing a new contract using an AI model, risks are automatically detected and specific risks are tagged. As a result, deficiencies in contract terms and legal risks are identified in advance, enabling prompt countermeasures.

[0078] Users can use their devices to review risks and areas for improvement detected by the AI. The information is displayed visually, and relevant recommended actions are shown, allowing users to make informed decisions. Users can also provide feedback, which is sent to the server to further improve the accuracy of the AI ​​model.

[0079] As a concrete example, one might input the following prompt into the AI ​​model: "Use past contract data to learn risk patterns and detect risks under specific conditions."

[0080] This system improves the accuracy and speed of legal and security operations, and strengthens the organization's overall risk management capabilities.

[0081] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0082] Step 1:

[0083] The server connects to an internal database to collect legal and security-related data. It uses contracts and consultation histories stored within the company as input. Because this data requires centralized management, it is saved in a specific folder via database software and then converted into a format usable for subsequent processing.

[0084] Step 2:

[0085] The server uses OCR software to extract text from scanned images and performs data preprocessing, including noise reduction and formatting standardization. Using scanned image data as input, it generates clean, standardized text data as output. This process makes the text suitable for AI models.

[0086] Step 3:

[0087] The server supplies the formatted data to the generating AI model and starts the learning process. Machine learning frameworks such as TensorFlow are used. Using clean text data as input, the output is an AI model that has learned organization-specific risk knowledge. The AI ​​model can understand various risk factors and perform highly accurate analysis.

[0088] Step 4:

[0089] The server receives newly submitted contract documents and automatically reviews them using a pre-trained AI model. It receives the new contract as input. The AI ​​analyzes potential risks from the contract document and tags any anomalies. It generates a list of tagged risks as output and sends it to the next step.

[0090] Step 5:

[0091] The terminal displays risk information analyzed by the server to the user. It receives risk information sent from the server as input and displays it on the user interface in a visually organized format. As output, the user can view detailed information about the risk and recommended actions, enabling quick decision-making.

[0092] Step 6:

[0093] Users review the information provided by the system through their terminals and send feedback to the server. They refer to the risk information and recommendations displayed on their terminals as input, and submit opinions and modifications. The feedback for improvement is sent to the server as output and used to improve the accuracy of the model.

[0094] (Application Example 1)

[0095] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart device 14 will be referred to as the "terminal."

[0096] In modern society, streamlining legal and security operations is a critical challenge. However, traditional methods require significant time and effort to process vast amounts of document data and identify risks. Furthermore, the slow pace of risk detection and presentation to users leads to delays in decision-making. Moreover, the lack of mechanisms for continuous model improvement utilizing feedback limits the accuracy of existing systems.

[0097] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0098] In this invention, the server includes information processing means for collecting and centrally managing past legal and security-related data accumulated within the company; data processing means for preprocessing the acquired data, removing noise data, and standardizing the text; and learning means for using a generation AI model to retrain based on the preprocessed data and generate responses using organization-specific knowledge. This enables automated review of contract documents and automated risk detection.

[0099] "Information processing means" refers to a device or system for collecting and centrally managing past legal and security-related data accumulated within a company.

[0100] "Data processing means" refers to a device or system for preprocessing acquired data, removing noise data, and standardizing text data.

[0101] A "generative AI model" refers to artificial intelligence technology that learns from data and generates responses to specific tasks.

[0102] A "learning tool" is a device or system that uses a generative AI model to retrain itself based on pre-processed data and generate responses using organization-specific knowledge.

[0103] A "risk detection means" is a device or system that analyzes newly acquired contract document data, automatically detects potential risks, and tags them.

[0104] "Result generation means" refers to a device or system for visualizing risk detection results and outputting them in report format.

[0105] A "feedback acquisition method" is a device or system that acquires feedback from users and uses that information to improve the accuracy of an AI model.

[0106] "Image processing means" refers to a device or system for analyzing image data from multiple input devices and performing optical character recognition.

[0107] "Display output means" refers to a device or system for visually displaying results based on analyzed text data.

[0108] The system for carrying out this invention consists of information processing means, data processing means, learning means using a generated AI model, risk detection means, result generation means, feedback acquisition means, image processing means, and display output means. Each of these means is described in detail below.

[0109] First, the server collects legal and security-related data accumulated within the company and uses an information processing system that centrally manages it in a single database. At this stage, more efficient data management becomes possible.

[0110] Next, the server preprocesses the acquired data, using data processing techniques to remove noise and standardize the text. This facilitates analysis by the generative AI model.

[0111] Generative AI model-based learning methods learn organization-specific knowledge from pre-processed data to prepare for future contract reviews. At this stage, tools such as OpenAI® and Google® AI Platform can be utilized.

[0112] For newly acquired contract documents, the image data is analyzed using image processing tools and converted into text data through optical character recognition. Image processing libraries such as OpenCV are often used for this purpose.

[0113] The risk detection system uses a generative AI model to analyze contract documents, identify potential risks, and tag them. The tagged risk information is then visualized and output in report format by the results generation system.

[0114] The device presents the results to the user, visually displaying detected risks and recommended actions. A user-friendly interface can be provided by using front-end frameworks such as Flutter® or React.

[0115] Furthermore, user feedback is collected through feedback acquisition methods and used to continuously improve the accuracy of the AI ​​model.

[0116] A concrete example of its application is a process where, for instance, when a company creates a new internal policy contract, it uses "Secure Legal Review" technology to photograph the contract and then performs AI-powered risk detection. The AI ​​automatically detects potential problems within the contract and notifies the user along with recommended actions.

[0117] An example of a prompt for a generative AI model would be: "Detect potential risks within the contract and determine whether the clauses related to 'payment terms' are appropriate."

[0118] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0119] Step 1:

[0120] The server collects legal and security-related data accumulated within the company. This data includes contracts, consultation histories, and risk assessment reports. The entered data is centrally managed by information processing equipment and stored in a database.

[0121] Step 2:

[0122] The server preprocesses the acquired data. It uses data processing techniques to remove noise and standardize the text. Specifically, it removes unnecessary document formatting information and extracts pure text data. This results in data that is suitable for analysis by the generative AI model.

[0123] Step 3:

[0124] The server uses a generative AI model to learn organization-specific knowledge from pre-processed data. This AI model is built using frameworks such as TensorFlow and generates responses that can be used in future contract reviews throughout the learning process. The output of this step is the trained model.

[0125] Step 4:

[0126] The user inputs an image of a newly acquired contract document into the terminal. The terminal analyzes the image data using image processing equipment and performs OCR to convert it into text data. Using image processing libraries such as OpenCV, it extracts document information from the contract and generates the necessary text data.

[0127] Step 5:

[0128] The server analyzes the OCR-processed text data using risk detection methods. A generative AI model is used to automatically detect potential risks and tag them. As a result of the analysis, data classified according to the type of risk is output.

[0129] Step 6:

[0130] The server visualizes the risk detection results and outputs them in report format using a results generation mechanism. The output report includes the detected risk information and recommended actions based on it.

[0131] Step 7:

[0132] The device visually presents the user with detected risk information and recommended actions. Frontend frameworks such as Flutter and React are used to display the information clearly on the user interface.

[0133] Step 8:

[0134] Users provide feedback based on the results via their devices. This feedback is sent to a server via a feedback acquisition system and used to improve the accuracy of the AI ​​model.

[0135] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0136] This invention combines a system designed to improve efficiency in legal and security operations with an emotion engine that recognizes user emotions. This system provides efficient and user-friendly support through the following process.

[0137] Data collection and management

[0138] The server collects legal and security-related data from various internal sources and stores it in a database. This centralized management ensures data availability and integrity.

[0139] Data preprocessing

[0140] The server preprocesses the accumulated data, removing noise and standardizing the text data. This process makes the data suitable for efficient training by AI models.

[0141] AI model training

[0142] The server uses organized data to train a generative AI model. In particular, by acquiring organization-specific knowledge and industry-specific risk patterns, it becomes possible to respond in a way that is tailored to the organization's needs.

[0143] Automated contract review and risk detection

[0144] The server analyzes newly acquired contract data using an AI model to automatically detect potential risks. Risk points are tagged, and analysis is performed according to their importance. For example, it can identify ambiguity in payment terms and deficiencies in legal liability.

[0145] Emotional engine integration

[0146] The server analyzes user feedback using an emotion engine. This retrieves user emotion data at the time of feedback, which is then used to improve the system's response.

[0147] Presentation of results and feedback processing

[0148] The terminal presents the analysis results to the user, offering recommended actions along with detailed risk information. The presentation of results takes the user's emotions into consideration, adjusting the priority and format of information accordingly. For example, if the user expresses negative emotions, the server will make the explanation more helpful and detailed.

[0149] Improving model accuracy

[0150] User feedback and corresponding emotional data are stored on the server and used to improve the accuracy of the AI ​​model. This continuous improvement process enables the system to provide even greater value in the user experience.

[0151] This configuration improves the efficiency of contract review and risk management, while also enabling flexible responses that take user sentiment into consideration.

[0152] The following describes the processing flow.

[0153] Step 1:

[0154] The server collects historical legal and security-related data from the company's information systems and databases. The collected data is centralized and prepared as a foundation for subsequent processing.

[0155] Step 2:

[0156] The server preprocesses the collected data. Specifically, it removes unnecessary information and noise and converts the text data into a standard format. Improving the quality of the data at this stage enhances the accuracy of analysis by the AI ​​model.

[0157] Step 3:

[0158] The server uses the formatted text data to retrain the generative AI model. This process strengthens the system's knowledge base by learning industry-specific risk patterns and legal terminology.

[0159] Step 4:

[0160] The server analyzes newly submitted contract data through an AI model. While scanning the contract content, it detects potential risks and tags them according to their importance. For example, it identifies ambiguous clauses and potential legal issues.

[0161] Step 5:

[0162] The server uses a results generation mechanism to compile the risk detection results into a report. The report includes recommended actions for the user and details of the risks, providing a basis for future decision-making.

[0163] Step 6:

[0164] The terminal presents the user with a risk report. The user interface visually organizes the information and prompts the user to take necessary actions to aid their understanding.

[0165] Step 7:

[0166] The device and the emotion engine work together to analyze the user's emotions when they review risk reports. This data provides valuable feedback for improving the user experience.

[0167] Step 8:

[0168] Users provide feedback. The emotional information obtained here is sent to the server and used to adjust and improve the AI ​​model. This feedback loop continuously improves the system's accuracy and user satisfaction.

[0169] (Example 2)

[0170] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart device 14 as the "terminal".

[0171] While there is a need for increased efficiency and accuracy in legal and security management operations, conventional systems struggle to collect and manage large amounts of data, making it difficult to quickly identify new risk factors. Furthermore, there is a lack of systems that reflect user feedback in real time, hindering the implementation of flexible responses based on user sentiment.

[0172] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0173] In this invention, the server includes information gathering means for collecting and integrating a wide range of legal and security-related information from information sources; data processing means for removing interfering information by preprocessing the acquired information and arranging the information in a unified format; and a learning device that uses a generated machine learning model to perform further learning based on the processed information and generate responses using the organization's unique knowledge. This enables efficient and accurate risk identification and quick, flexible responses while considering user sentiment.

[0174] "Information sources" refer to means of providing diverse data related to legal and security management, including internal document management systems and communication systems.

[0175] An "information gathering method" is a system for centrally collecting and integrating data from various information sources.

[0176] "Data processing means" refers to methods for removing noise and unnecessary information from collected data and preparing the information in a format that can be analyzed.

[0177] A "machine learning model" is a framework that includes algorithms that use artificial intelligence technology to learn patterns and rules from data and generate responses and predictions based on that learning.

[0178] A "learning device" is a device that uses machine learning models to retrain based on processed data and generate responses using organization-specific knowledge.

[0179] A "risk detection device" is a device that analyzes newly acquired document data and automatically detects and assigns attributes to potential risks.

[0180] A "results generation device" is a device that visually displays the results of detected risks and outputs them in a report format.

[0181] A "response acquisition device" is a device that effectively acquires user feedback and uses that feedback to improve the accuracy of machine learning models.

[0182] "Emotional analysis tools" are means of analyzing user feedback and obtaining emotional information at that time.

[0183] This invention consists of a system that performs a series of processes including data collection, preprocessing, analysis, result presentation, and improvement based on feedback. Specifically, the server, terminal, and user elements each play their respective roles, enabling efficient support for legal and security operations.

[0184] First, the server automatically collects a wide range of legal and security-related information from internal sources. This utilizes hardware and software such as document management systems and mail servers over the network. The collected data is then processed using data processing tools to standardize the text, remove noise, and organize it into a unified format.

[0185] Next, the server uses the pre-processed data to train a generative AI model. This training process utilizes pattern recognition techniques to incorporate industry-specific hazard patterns and specific legal terminology into the model, enabling responses that leverage the organization's unique knowledge. The AI ​​model analyzes newly acquired agreement document information and automatically identifies potential risks using hazard detection devices.

[0186] The analysis results are presented to the user via a terminal. The terminal's interface, through a results generation device, visualizes risks and outputs them in report format. Information prioritization and presentation are adjusted according to the user's emotions, enabling more effective decision-making. For example, if the user expresses negative emotions, the server presents the information in a more user-friendly and easy-to-understand manner.

[0187] Finally, user feedback is received by the server's response acquisition device, and the user's emotional information is extracted by emotion analysis. This feedback information is used to improve the accuracy of the AI ​​model, and the model is continuously improved.

[0188] For example, by using a prompt such as, "Review the new contract, analyze potential risks, and create a prompt to consider improvements based on the feedback," more detailed and accurate analysis results can be obtained. In this way, the system continues to provide fast and reliable support in legal and security management.

[0189] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0190] Step 1:

[0191] The server collects legal and security-related information from internal sources. Inputs include data from multiple databases, document management systems, and mail servers. The server automatically integrates this data to create a centralized dataset. The output is a unified database. Specifically, scheduled tasks are set to periodically collect data, ensuring that the most up-to-date information is accumulated.

[0192] Step 2:

[0193] The server preprocesses the collected data. The database integrated in step 1 is used as input. Denoising and standardization are performed to normalize text and numerical data with inconsistent formats. The output is a processed, clean dataset. Specifically, this involves text filtering using regular expressions and imputation of missing values.

[0194] Step 3:

[0195] The server trains a generative AI model using preprocessed data. The input is the clean data obtained in step 2. The processed data is fed to the AI ​​model, and it performs learning for pattern recognition and risk prediction. The output is an AI model that incorporates organization-specific knowledge. Specifically, it uses supervised learning methods to extract meaningful patterns from past data.

[0196] Step 4:

[0197] The server analyzes new contract data using an AI model to automatically detect potential risks. The input consists of new document data such as agreements and contracts. The AI ​​model analyzes these documents, identifies and tags risk factors. The output is an analysis result with visualized risks. Specifically, it classifies and prioritizes risks according to their importance.

[0198] Step 5:

[0199] The terminal presents the analysis results to the user. The risk analysis results obtained in step 4 are used as input. The terminal provides the user with visualized information via its display and presents recommended actions. The output provides information in a format that is easy for the user to understand. Specifically, text and graphs are displayed via the interface, and links to detailed information are provided as needed.

[0200] Step 6:

[0201] The user provides feedback based on the information presented. The input is the analysis information provided in step 5. Through the feedback, the user inputs evaluations and impressions into the server, which are used to improve the system. The output is the feedback data sent to the server. Specifically, information is entered using a feedback form.

[0202] Step 7:

[0203] The server receives feedback from users and uses it to improve the accuracy of the AI ​​model. The input is the feedback data obtained in step 6. Sentiment analysis is used to extract user emotion information and utilize it to improve the AI ​​model. The output is an updated AI model. Specifically, continuous training data is added and the model is retrained.

[0204] (Application Example 2)

[0205] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as a "server" and the smart device 14 as a "terminal".

[0206] In on-site security operations, real-time detection of risks and immediate response are required. However, conventional systems suffer from fragmented and inefficient information gathering and analysis, often relying on human judgment under stressful circumstances. As a result, the accuracy of risk management decreases, and the mental burden on users increases.

[0207] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0208] In this invention, the server includes data processing means for collecting and centrally managing historical information-related data accumulated within the company; information processing means for pre-processing the acquired data, removing unnecessary data, and standardizing the information; and learning means for retraining a generation AI model based on the pre-processed data to generate responses using organization-specific knowledge. This enables real-time automatic detection of risks and the presentation of appropriate countermeasures, and further improves work efficiency and reduces mental burden by providing support that takes into account the user's emotional data.

[0209] "Data processing means" refers to a device or method for collecting and centrally managing historical information-related data accumulated within a company.

[0210] "Information processing means" refers to an apparatus or method for pre-processing acquired data, removing unnecessary data, and organizing the information into a standardized format.

[0211] A "learning tool" is a device or method for generating responses using organization-specific knowledge by retraining a generative AI model based on pre-processed data.

[0212] A "risk detection means" is a device or method for analyzing newly acquired information document data, automatically detecting potential hazards, and labeling them.

[0213] A "risk assessment tool" is a device or method for analyzing video and audio data collected in real time to evaluate the safety of a site.

[0214] "Emotion analysis means" refers to a device or method for acquiring user feedback and user emotion data and using them to improve the accuracy of an AI model.

[0215] The system of this invention is designed to provide real-time risk detection and user support in security operations. In this system, a server primarily handles information processing, while terminals such as smart devices provide on-site data collection and interfaces.

[0216] The server collects and centrally manages historical information data accumulated within the company using data processing tools. Next, information processing tools preprocess the acquired data to remove unnecessary data and arrange it in a standardized format. Based on this standardized data, a learning tool retrains using a generation AI model, utilizing organization-specific knowledge in its responses. Furthermore, a risk detection tool analyzes new information document data, automatically detects potential risks, and labels them.

[0217] The terminal collects video and audio data in real time at the site using risk assessment tools and transmits it to the server. The server analyzes this data, assesses the safety of the site, and proposes immediate countermeasures. The sentiment analysis tool acquires user feedback and sentiment data and uses it to continuously improve the accuracy of the AI ​​model.

[0218] For example, suppose a security staff member wearing smart glasses is patrolling a facility when a sudden unusual sound is detected. The server analyzes the anomaly using risk assessment tools and determines that it "may be an animal or a small moving object," then issues instructions to the on-site staff. At the same time, it analyzes the staff member's stress level from their voice and gestures using emotion analysis tools and suggests, "Take a three-second deep breath" to encourage calm behavior.

[0219] An example of a prompt is, "Explain how this security system detects on-site hazards and supports the mental health of personnel." This prompt allows the AI ​​model to learn how to respond in specific scenarios.

[0220] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0221] Step 1:

[0222] The terminal acquires video and audio in real time as on-site monitoring data and transmits it to the server. It receives raw data from devices such as cameras and microphones as input and converts it into a processable data format as output.

[0223] Step 2:

[0224] The server processes the received video and audio data as input to a risk assessment system. This process utilizes a generative AI model to analyze specific movements and sounds. The output identifies potential hazards, and labels are applied as needed. Specific actions include detecting suspicious behavior and unusual sounds.

[0225] Step 3:

[0226] The server uses information processing tools to apply standardized data from existing databases to perform matching with new data and anomaly detection. It receives pre-processed data as input and generates information about potential risks as output.

[0227] Step 4:

[0228] The server uses emotion analysis tools to analyze user input data, such as voice tone and facial expression data. Based on this input, a generative AI model performs analysis, and the user's emotional state data is generated as output. Specifically, stress levels and anxiety levels are determined.

[0229] Step 5:

[0230] The server prepares feedback for the user based on the analysis results. It uses risk assessment and emotional state data as input and constructs specific action suggestions and guidance for the user as output. Specific examples include behavioral instructions tailored to the security situation and relaxation advice.

[0231] Step 6:

[0232] The terminal receives feedback from the server and presents the results visually or audibly through the user interface. It receives feedback data based on input and generates screen displays or audio messages as output.

[0233] The specific processing unit 290 transmits the result of the specific processing to the smart device 14. In the smart device 14, the control unit 46A causes the output device 40 to output the result of the specific processing. The microphone 38B acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 38B to the data processing device 12. In the data processing device 12, the specific processing unit 290 acquires the audio data.

[0234] Data generation model 58 is a so-called generative AI (Artificial Intelligence). An example of data generation model 58 is ChatGPT (registered trademark) (Internet search).<URL: https: / / openai.com / blog / chatgpt> ), Gemini (registered trademark) (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0235] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart device 14.

[0236] [Second Embodiment]

[0237] Figure 3 shows an example of the configuration of the data processing system 210 according to the second embodiment.

[0238] As shown in Figure 3, the data processing system 210 includes a data processing device 12 and smart glasses 214. An example of the data processing device 12 is a server.

[0239] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0240] The smart glasses 214 include a computer 36, a microphone 238, a speaker 240, a camera 42, and a communication interface 44. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, and camera 42 are also connected to the bus 52.

[0241] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0242] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0243] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0244] Figure 4 shows an example of the main functions of the data processing device 12 and the smart glasses 214. As shown in Figure 4, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0245] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0246] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0247] In the smart glasses 214, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0248] Next, the identification processing performed by the identification processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0249] This invention is configured as a system for improving the efficiency of legal and security operations. Specific embodiments of this system are described below.

[0250] Data collection and management

[0251] The server collects legal and security-related data accumulated within the company over time. This data includes contracts, past consultation histories, and risk assessment reports. This data is centrally managed in a database.

[0252] Data preprocessing

[0253] The server removes unnecessary noise from the collected data and standardizes the text. This allows the AI ​​model to learn more efficiently. For example, it removes unnecessary document formatting information and extracts pure text data.

[0254] AI model training

[0255] The server uses a generative AI model that learns from organized data. This AI model learns organization-specific legal knowledge and industry-specific risk factors to achieve optimal analytical capabilities.

[0256] Automated contract review and risk detection

[0257] The server uses a trained AI model to automatically review newly submitted contracts. This review process can automatically detect potential risks hidden within the contract and tag them according to their type. For example, it can identify deficiencies in contract terms and potential legal risks.

[0258] Presentation of results and collection of feedback

[0259] The device presents the user with risks and review points detected by the AI. Each risk is accompanied by relevant information and recommended actions, allowing the user to make decisions based on this information. Simultaneously, the user provides feedback on the review results, which is sent to the server for model improvement.

[0260] This system aims to streamline contract review and risk management, supporting users in making quick and accurate decisions. Through its implementation methods, it improves the accuracy and speed of legal and security-related tasks.

[0261] The following describes the processing flow.

[0262] Step 1:

[0263] The server collects historical legal and security-related data from various sources within the company and stores it centrally in a database. This ensures that the necessary data is efficiently accessible.

[0264] Step 2:

[0265] The server performs preprocessing to remove noise from the collected data. Specifically, it standardizes the data by removing unnecessary headers, footers, and email signatures, and by unifying the text format.

[0266] Step 3:

[0267] The server retrains the generative AI model using pre-processed data. In this process, it learns past patterns and risk factors specific to the organization, improving the accuracy and effectiveness of the AI ​​model.

[0268] Step 4:

[0269] The server inputs the text data of the submitted new contract into an AI model for automatic analysis. Here, it detects deficiencies in the contract terms and legal risks, and tags the relevant sections with risk tags.

[0270] Step 5:

[0271] The server compiles the analysis results and uses them to create a detailed risk report. This report includes details of each detected risk and recommended actions for addressing them.

[0272] Step 6:

[0273] The terminal displays the generated risk report to the user. Based on this information, the user can make decisions regarding contract modifications and risk management.

[0274] Step 7:

[0275] Users review the presented report and provide feedback. This feedback is sent to the server and used to retrain the AI ​​model to further improve its accuracy.

[0276] (Example 1)

[0277] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0278] In modern organizations, legal and security-related operations rely on vast amounts of data. While efficiently managing and analyzing this data is crucial, manual processes are time-consuming, costly, and prone to overlooking risks and making misjudgments. A system is needed to address these challenges and improve the accuracy and speed of operations.

[0279] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0280] In this invention, the server includes information processing means for collecting and centrally managing information, data preprocessing means for standardizing acquired data, and learning means for relearning knowledge from the preprocessed data using a generative model. This enables highly accurate and efficient data analysis and automated risk detection compared to conventional manual work.

[0281] "Information processing means" refers to technologies used to effectively collect and centrally manage data gathered within an organization.

[0282] "Data preprocessing means" refers to processes that perform noise reduction and standardization in order to improve the quality of acquired data.

[0283] A "generative model" is an artificial intelligence model that learns from large amounts of data and generates new information or responses.

[0284] The "learning means" is a technology for enabling a model to acquire new knowledge using the collected data and improving the accuracy and usefulness of responses.

[0285] The "analysis means" is a process for automatically analyzing newly acquired document data and detecting and classifying potential risks.

[0286] The "result generation means" is a technology for visually and clearly displaying the analyzed data and outputting it in document form.

[0287] The "feedback means" is a mechanism for collecting opinions and evaluations from users and using them to improve the system or model.

[0288] The "user interface" is a visual interface used by users to interact with the system, obtain information, and perform operations.

[0289] The present invention relates to a system aimed at improving the operational efficiency of legal and security operations. This system operates centered around a server in order to effectively accumulate and manage information.

[0290] The server accesses an internal database and collects legal and security-related data accumulated so far. This data includes contract documents, past consultation histories, and risk assessment records. At this stage, database software is used to efficiently manage the data centrally.

[0291] Next, the server performs data preprocessing. Here, OCR (Optical Character Recognition) software is used to extract text from image files and perform text noise removal and standardization. Through this process, the data is converted into a form suitable for learning by the generative AI model.

[0292] Subsequently, the server launches a generative AI model and trains it using the formatted data. This process utilizes machine learning frameworks such as TensorFlow or PyTorch. By learning organization-specific knowledge and industry-specific risk factors, the AI ​​model becomes capable of highly accurate data analysis.

[0293] For example, when reviewing a new contract using an AI model, risks are automatically detected and specific risks are tagged. As a result, deficiencies in contract terms and legal risks are identified in advance, enabling prompt countermeasures.

[0294] Users can use their devices to review risks and areas for improvement detected by the AI. The information is displayed visually, and relevant recommended actions are shown, allowing users to make informed decisions. Users can also provide feedback, which is sent to the server to further improve the accuracy of the AI ​​model.

[0295] As a concrete example, one might input the following prompt into the AI ​​model: "Use past contract data to learn risk patterns and detect risks under specific conditions."

[0296] This system improves the accuracy and speed of legal and security operations, and strengthens the organization's overall risk management capabilities.

[0297] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0298] Step 1:

[0299] The server connects to an internal database to collect legal and security-related data. It uses contracts and consultation histories stored within the company as input. Because this data requires centralized management, it is saved in a specific folder via database software and then converted into a format usable for subsequent processing.

[0300] Step 2:

[0301] The server uses OCR software to extract text from the scanned image and performs data preprocessing. This includes noise removal and format standardization. Using the scanned image data as input, it generates clean and standardized text data as output. Through this process, the text is in a form suitable for the AI model.

[0302] Step 3:

[0303] The server supplies the formatted data to the AI model for generating and starts the learning process. A machine learning framework such as TensorFlow is used. Using clean text data as input, it obtains an AI model that has learned the organization-specific risk knowledge as output. The AI model can understand various risk factors and enable high-precision analysis.

[0304] Step 4:

[0305] The server receives the newly submitted contract document and automatically reviews it using the pre-trained AI model. It receives the new contract as input. The AI analyzes potential risks from the contract document and tags the abnormal parts. As output, it generates a list of tagged risks and sends it to the next step.

[0306] Step 5:

[0307] The terminal displays the risk information analyzed by the server to the user. It receives the risk information sent from the server as input and displays it on the user interface in a visually organized format. As output, the user can check the detailed information and recommended actions regarding the risks, enabling quick decision-making.

[0308] Step 6:

[0309] Users review the information provided by the system through their terminals and send feedback to the server. They refer to the risk information and recommendations displayed on their terminals as input, and submit opinions and modifications. The feedback for improvement is sent to the server as output and used to improve the accuracy of the model.

[0310] (Application Example 1)

[0311] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0312] In modern society, streamlining legal and security operations is a critical challenge. However, traditional methods require significant time and effort to process vast amounts of document data and identify risks. Furthermore, the slow pace of risk detection and presentation to users leads to delays in decision-making. Moreover, the lack of mechanisms for continuous model improvement utilizing feedback limits the accuracy of existing systems.

[0313] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0314] In this invention, the server includes information processing means for collecting and centrally managing past legal and security-related data accumulated within the company; data processing means for preprocessing the acquired data, removing noise data, and standardizing the text; and learning means for using a generation AI model to retrain based on the preprocessed data and generate responses using organization-specific knowledge. This enables automated review of contract documents and automated risk detection.

[0315] "Information processing means" refers to a device or system for collecting and centrally managing past legal and security-related data accumulated within a company.

[0316] "Data processing means" refers to a device or system for preprocessing acquired data, removing noise data, and standardizing text data.

[0317] A "generative AI model" refers to artificial intelligence technology that learns from data and generates responses to specific tasks.

[0318] A "learning tool" is a device or system that uses a generative AI model to retrain itself based on pre-processed data and generate responses using organization-specific knowledge.

[0319] A "risk detection means" is a device or system that analyzes newly acquired contract document data, automatically detects potential risks, and tags them.

[0320] "Result generation means" refers to a device or system for visualizing risk detection results and outputting them in report format.

[0321] A "feedback acquisition method" is a device or system that acquires feedback from users and uses that information to improve the accuracy of an AI model.

[0322] "Image processing means" refers to a device or system for analyzing image data from multiple input devices and performing optical character recognition.

[0323] "Display output means" refers to a device or system for visually displaying results based on analyzed text data.

[0324] The system for carrying out this invention consists of information processing means, data processing means, learning means using a generated AI model, risk detection means, result generation means, feedback acquisition means, image processing means, and display output means. Each of these means is described in detail below.

[0325] First, the server collects legal and security-related data accumulated within the company and uses an information processing system that centrally manages it in a single database. At this stage, more efficient data management becomes possible.

[0326] Next, the server preprocesses the acquired data, using data processing techniques to remove noise and standardize the text. This facilitates analysis by the generative AI model.

[0327] Generative AI models are used to learn organization-specific knowledge from pre-processed data, preparing for future contract reviews. OpenAI and Google AI Platform can be used at this stage.

[0328] For newly acquired contract documents, the image data is analyzed using image processing tools and converted into text data through optical character recognition. Image processing libraries such as OpenCV are often used for this purpose.

[0329] The risk detection system uses a generative AI model to analyze contract documents, identify potential risks, and tag them. The tagged risk information is then visualized and output in report format by the results generation system.

[0330] The device presents the results to the user, visually displaying detected risks and recommended actions. A user-friendly interface can be provided by using front-end frameworks such as Flutter or React.

[0331] Furthermore, user feedback is collected through feedback acquisition methods and used to continuously improve the accuracy of the AI ​​model.

[0332] A concrete example of its application is a process where, for instance, when a company creates a new internal policy contract, it uses "Secure Legal Review" technology to photograph the contract and then performs AI-powered risk detection. The AI ​​automatically detects potential problems within the contract and notifies the user along with recommended actions.

[0333] An example of a prompt for a generative AI model would be: "Detect potential risks within the contract and determine whether the clauses related to 'payment terms' are appropriate."

[0334] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0335] Step 1:

[0336] The server collects legal and security-related data accumulated within the company. This data includes contracts, consultation histories, and risk assessment reports. The entered data is centrally managed by information processing equipment and stored in a database.

[0337] Step 2:

[0338] The server preprocesses the acquired data. It uses data processing techniques to remove noise and standardize the text. Specifically, it removes unnecessary document formatting information and extracts pure text data. This results in data that is suitable for analysis by the generative AI model.

[0339] Step 3:

[0340] The server uses a generative AI model to learn organization-specific knowledge from pre-processed data. This AI model is built using frameworks such as TensorFlow and generates responses that can be used in future contract reviews throughout the learning process. The output of this step is the trained model.

[0341] Step 4:

[0342] The user inputs an image of a newly acquired contract document into the terminal. The terminal analyzes the image data using image processing equipment and performs OCR to convert it into text data. Using image processing libraries such as OpenCV, it extracts document information from the contract and generates the necessary text data.

[0343] Step 5:

[0344] The server analyzes the OCR-processed text data using risk detection methods. A generative AI model is used to automatically detect potential risks and tag them. As a result of the analysis, data classified according to the type of risk is output.

[0345] Step 6:

[0346] The server visualizes the risk detection results and outputs them in report format using a results generation mechanism. The output report includes the detected risk information and recommended actions based on it.

[0347] Step 7:

[0348] The device visually presents the user with detected risk information and recommended actions. Frontend frameworks such as Flutter and React are used to display the information clearly on the user interface.

[0349] Step 8:

[0350] Users provide feedback based on the results via their devices. This feedback is sent to a server via a feedback acquisition system and used to improve the accuracy of the AI ​​model.

[0351] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0352] This invention combines a system designed to improve efficiency in legal and security operations with an emotion engine that recognizes user emotions. This system provides efficient and user-friendly support through the following process.

[0353] Data collection and management

[0354] The server collects legal and security-related data from various internal sources and stores it in a database. This centralized management ensures data availability and integrity.

[0355] Data preprocessing

[0356] The server preprocesses the accumulated data, removing noise and standardizing the text data. This process makes the data suitable for efficient training by AI models.

[0357] AI model training

[0358] The server uses organized data to train a generative AI model. In particular, by acquiring organization-specific knowledge and industry-specific risk patterns, it becomes possible to respond in a way that is tailored to the organization's needs.

[0359] Automated contract review and risk detection

[0360] The server analyzes newly acquired contract data using an AI model to automatically detect potential risks. Risk points are tagged, and analysis is performed according to their importance. For example, it can identify ambiguity in payment terms and deficiencies in legal liability.

[0361] Emotional engine integration

[0362] The server analyzes user feedback using an emotion engine. This retrieves user emotion data at the time of feedback, which is then used to improve the system's response.

[0363] Presentation of results and feedback processing

[0364] The terminal presents the analysis results to the user, offering recommended actions along with detailed risk information. The presentation of results takes the user's emotions into consideration, adjusting the priority and format of information accordingly. For example, if the user expresses negative emotions, the server will make the explanation more helpful and detailed.

[0365] Improving model accuracy

[0366] User feedback and corresponding emotional data are stored on the server and used to improve the accuracy of the AI ​​model. This continuous improvement process enables the system to provide even greater value in the user experience.

[0367] This configuration improves the efficiency of contract review and risk management, while also enabling flexible responses that take user sentiment into consideration.

[0368] The following describes the processing flow.

[0369] Step 1:

[0370] The server collects historical legal and security-related data from the company's information systems and databases. The collected data is centralized and prepared as a foundation for subsequent processing.

[0371] Step 2:

[0372] The server preprocesses the collected data. Specifically, it removes unnecessary information and noise and converts the text data into a standard format. Improving the quality of the data at this stage enhances the accuracy of analysis by the AI ​​model.

[0373] Step 3:

[0374] The server uses the formatted text data to retrain the generative AI model. This process strengthens the system's knowledge base by learning industry-specific risk patterns and legal terminology.

[0375] Step 4:

[0376] The server analyzes newly submitted contract data through an AI model. While scanning the contract content, it detects potential risks and tags them according to their importance. For example, it identifies ambiguous clauses and potential legal issues.

[0377] Step 5:

[0378] The server uses a results generation mechanism to compile the risk detection results into a report. The report includes recommended actions for the user and details of the risks, providing a basis for future decision-making.

[0379] Step 6:

[0380] The terminal presents the user with a risk report. The user interface visually organizes the information and prompts the user to take necessary actions to aid their understanding.

[0381] Step 7:

[0382] The device and the emotion engine work together to analyze the user's emotions when they review risk reports. This data provides valuable feedback for improving the user experience.

[0383] Step 8:

[0384] Users provide feedback. The emotional information obtained here is sent to the server and used to adjust and improve the AI ​​model. This feedback loop continuously improves the system's accuracy and user satisfaction.

[0385] (Example 2)

[0386] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the smart glasses 214 will be referred to as the "terminal".

[0387] While there is a need for increased efficiency and accuracy in legal and security management operations, conventional systems struggle to collect and manage large amounts of data, making it difficult to quickly identify new risk factors. Furthermore, there is a lack of systems that reflect user feedback in real time, hindering the implementation of flexible responses based on user sentiment.

[0388] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0389] In this invention, the server includes information gathering means for collecting and integrating a wide range of legal and security-related information from information sources; data processing means for removing interfering information by preprocessing the acquired information and arranging the information in a unified format; and a learning device that uses a generated machine learning model to perform further learning based on the processed information and generate responses using the organization's unique knowledge. This enables efficient and accurate risk identification and quick, flexible responses while considering user sentiment.

[0390] "Information sources" refer to means of providing diverse data related to legal and security management, including internal document management systems and communication systems.

[0391] An "information gathering method" is a system for centrally collecting and integrating data from various information sources.

[0392] "Data processing means" refers to methods for removing noise and unnecessary information from collected data and preparing the information in a format that can be analyzed.

[0393] A "machine learning model" is a framework that includes algorithms that use artificial intelligence technology to learn patterns and rules from data and generate responses and predictions based on that learning.

[0394] A "learning device" is a device that uses machine learning models to retrain based on processed data and generate responses using organization-specific knowledge.

[0395] A "risk detection device" is a device that analyzes newly acquired document data and automatically detects and assigns attributes to potential risks.

[0396] A "results generation device" is a device that visually displays the results of detected risks and outputs them in a report format.

[0397] A "response acquisition device" is a device that effectively acquires user feedback and uses that feedback to improve the accuracy of machine learning models.

[0398] "Emotional analysis tools" are means of analyzing user feedback and obtaining emotional information at that time.

[0399] This invention consists of a system that performs a series of processes including data collection, preprocessing, analysis, result presentation, and improvement based on feedback. Specifically, the server, terminal, and user elements each play their respective roles, enabling efficient support for legal and security operations.

[0400] First, the server automatically collects a wide range of legal and security-related information from internal sources. This utilizes hardware and software such as document management systems and mail servers over the network. The collected data is then processed using data processing tools to standardize the text, remove noise, and organize it into a unified format.

[0401] Next, the server uses the pre-processed data to train a generative AI model. This training process utilizes pattern recognition techniques to incorporate industry-specific hazard patterns and specific legal terminology into the model, enabling responses that leverage the organization's unique knowledge. The AI ​​model analyzes newly acquired agreement document information and automatically identifies potential risks using hazard detection devices.

[0402] The analysis results are presented to the user via a terminal. The terminal's interface, through a results generation device, visualizes risks and outputs them in report format. Information prioritization and presentation are adjusted according to the user's emotions, enabling more effective decision-making. For example, if the user expresses negative emotions, the server presents the information in a more user-friendly and easy-to-understand manner.

[0403] Finally, user feedback is received by the server's response acquisition device, and the user's emotional information is extracted by emotion analysis. This feedback information is used to improve the accuracy of the AI ​​model, and the model is continuously improved.

[0404] For example, by using a prompt such as, "Review the new contract, analyze potential risks, and create a prompt to consider improvements based on the feedback," more detailed and accurate analysis results can be obtained. In this way, the system continues to provide fast and reliable support in legal and security management.

[0405] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0406] Step 1:

[0407] The server collects legal and security-related information from internal sources. Inputs include data from multiple databases, document management systems, and mail servers. The server automatically integrates this data to create a centralized dataset. The output is a unified database. Specifically, scheduled tasks are set to periodically collect data, ensuring that the most up-to-date information is accumulated.

[0408] Step 2:

[0409] The server preprocesses the collected data. The database integrated in step 1 is used as input. Denoising and standardization are performed to normalize text and numerical data with inconsistent formats. The output is a processed, clean dataset. Specifically, this involves text filtering using regular expressions and imputation of missing values.

[0410] Step 3:

[0411] The server trains a generative AI model using preprocessed data. The input is the clean data obtained in step 2. The processed data is fed to the AI ​​model, and it performs learning for pattern recognition and risk prediction. The output is an AI model that incorporates organization-specific knowledge. Specifically, it uses supervised learning methods to extract meaningful patterns from past data.

[0412] Step 4:

[0413] The server analyzes new contract data using an AI model to automatically detect potential risks. The input consists of new document data such as agreements and contracts. The AI ​​model analyzes these documents, identifies and tags risk factors. The output is an analysis result with visualized risks. Specifically, it classifies and prioritizes risks according to their importance.

[0414] Step 5:

[0415] The terminal presents the analysis results to the user. The risk analysis results obtained in step 4 are used as input. The terminal provides the user with visualized information via its display and presents recommended actions. The output provides information in a format that is easy for the user to understand. Specifically, text and graphs are displayed via the interface, and links to detailed information are provided as needed.

[0416] Step 6:

[0417] The user provides feedback based on the information presented. The input is the analysis information provided in step 5. Through the feedback, the user inputs evaluations and impressions into the server, which are used to improve the system. The output is the feedback data sent to the server. Specifically, information is entered using a feedback form.

[0418] Step 7:

[0419] The server receives feedback from users and uses it to improve the accuracy of the AI ​​model. The input is the feedback data obtained in step 6. Sentiment analysis is used to extract user emotion information and utilize it to improve the AI ​​model. The output is an updated AI model. Specifically, continuous training data is added and the model is retrained.

[0420] (Application Example 2)

[0421] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the smart glasses 214 will be referred to as the "terminal."

[0422] In on-site security operations, real-time detection of risks and immediate response are required. However, conventional systems suffer from fragmented and inefficient information gathering and analysis, often relying on human judgment under stressful circumstances. As a result, the accuracy of risk management decreases, and the mental burden on users increases.

[0423] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0424] In this invention, the server includes data processing means for collecting and centrally managing historical information-related data accumulated within the company; information processing means for pre-processing the acquired data, removing unnecessary data, and standardizing the information; and learning means for retraining a generation AI model based on the pre-processed data to generate responses using organization-specific knowledge. This enables real-time automatic detection of risks and the presentation of appropriate countermeasures, and further improves work efficiency and reduces mental burden by providing support that takes into account the user's emotional data.

[0425] "Data processing means" refers to a device or method for collecting and centrally managing historical information-related data accumulated within a company.

[0426] "Information processing means" refers to an apparatus or method for pre-processing acquired data, removing unnecessary data, and organizing the information into a standardized format.

[0427] A "learning tool" is a device or method for generating responses using organization-specific knowledge by retraining a generative AI model based on pre-processed data.

[0428] A "risk detection means" is a device or method for analyzing newly acquired information document data, automatically detecting potential hazards, and labeling them.

[0429] A "risk assessment tool" is a device or method for analyzing video and audio data collected in real time to evaluate the safety of a site.

[0430] "Emotion analysis means" refers to a device or method for acquiring user feedback and user emotion data and using them to improve the accuracy of an AI model.

[0431] The system of this invention is designed to provide real-time risk detection and user support in security operations. In this system, a server primarily handles information processing, while terminals such as smart devices provide on-site data collection and interfaces.

[0432] The server collects and centrally manages historical information data accumulated within the company using data processing tools. Next, information processing tools preprocess the acquired data to remove unnecessary data and arrange it in a standardized format. Based on this standardized data, a learning tool retrains using a generation AI model, utilizing organization-specific knowledge in its responses. Furthermore, a risk detection tool analyzes new information document data, automatically detects potential risks, and labels them.

[0433] The terminal collects video and audio data in real time at the site using risk assessment tools and transmits it to the server. The server analyzes this data, assesses the safety of the site, and proposes immediate countermeasures. The sentiment analysis tool acquires user feedback and sentiment data and uses it to continuously improve the accuracy of the AI ​​model.

[0434] For example, suppose a security staff member wearing smart glasses is patrolling a facility when a sudden unusual sound is detected. The server analyzes the anomaly using risk assessment tools and determines that it "may be an animal or a small moving object," then issues instructions to the on-site staff. At the same time, it analyzes the staff member's stress level from their voice and gestures using emotion analysis tools and suggests, "Take a three-second deep breath" to encourage calm behavior.

[0435] An example of a prompt is, "Explain how this security system detects on-site hazards and supports the mental health of personnel." This prompt allows the AI ​​model to learn how to respond in specific scenarios.

[0436] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0437] Step 1:

[0438] The terminal acquires video and audio in real time as on-site monitoring data and transmits it to the server. It receives raw data from devices such as cameras and microphones as input and converts it into a processable data format as output.

[0439] Step 2:

[0440] The server processes the received video and audio data as input to a risk assessment system. This process utilizes a generative AI model to analyze specific movements and sounds. The output identifies potential hazards, and labels are applied as needed. Specific actions include detecting suspicious behavior and unusual sounds.

[0441] Step 3:

[0442] The server uses information processing tools to apply standardized data from existing databases to perform matching with new data and anomaly detection. It receives pre-processed data as input and generates information about potential risks as output.

[0443] Step 4:

[0444] The server uses emotion analysis tools to analyze user input data, such as voice tone and facial expression data. Based on this input, a generative AI model performs analysis, and the user's emotional state data is generated as output. Specifically, stress levels and anxiety levels are determined.

[0445] Step 5:

[0446] The server prepares feedback for the user based on the analysis results. It uses risk assessment and emotional state data as input and constructs specific action suggestions and guidance for the user as output. Specific examples include behavioral instructions tailored to the security situation and relaxation advice.

[0447] Step 6:

[0448] The terminal receives feedback from the server and presents the results visually or audibly through the user interface. It receives feedback data based on input and generates screen displays or audio messages as output.

[0449] The specific processing unit 290 transmits the result of the specific processing to the smart glasses 214. In the smart glasses 214, the control unit 46A causes the speaker 240 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0450] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0451] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the smart glasses 214.

[0452] [Third Embodiment]

[0453] Figure 5 shows an example of the configuration of the data processing system 310 according to the third embodiment.

[0454] As shown in Figure 5, the data processing system 310 includes a data processing device 12 and a headset terminal 314. An example of the data processing device 12 is a server.

[0455] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0456] The headset terminal 314 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a display 343. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and display 343 are also connected to the bus 52.

[0457] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0458] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0459] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0460] Figure 6 shows an example of the main functions of the data processing device 12 and the headset terminal 314. As shown in Figure 6, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0461] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0462] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0463] In the headset terminal 314, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0464] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the headset terminal 314 will be referred to as the "terminal".

[0465] This invention is configured as a system for improving the efficiency of legal and security operations. Specific embodiments of this system are described below.

[0466] Data collection and management

[0467] The server collects legal and security-related data accumulated within the company over time. This data includes contracts, past consultation histories, and risk assessment reports. This data is centrally managed in a database.

[0468] Data preprocessing

[0469] The server removes unnecessary noise from the collected data and standardizes the text. This allows the AI ​​model to learn more efficiently. For example, it removes unnecessary document formatting information and extracts pure text data.

[0470] AI model training

[0471] The server uses a generative AI model that learns from organized data. This AI model learns organization-specific legal knowledge and industry-specific risk factors to achieve optimal analytical capabilities.

[0472] Automated contract review and risk detection

[0473] The server uses a trained AI model to automatically review newly submitted contracts. This review process can automatically detect potential risks hidden within the contract and tag them according to their type. For example, it can identify deficiencies in contract terms and potential legal risks.

[0474] Presentation of results and collection of feedback

[0475] The device presents the user with risks and review points detected by the AI. Each risk is accompanied by relevant information and recommended actions, allowing the user to make decisions based on this information. Simultaneously, the user provides feedback on the review results, which is sent to the server for model improvement.

[0476] This system aims to streamline contract review and risk management, supporting users in making quick and accurate decisions. Through its implementation methods, it improves the accuracy and speed of legal and security-related tasks.

[0477] The following describes the processing flow.

[0478] Step 1:

[0479] The server collects historical legal and security-related data from various sources within the company and stores it centrally in a database. This ensures that the necessary data is efficiently accessible.

[0480] Step 2:

[0481] The server performs preprocessing to remove noise from the collected data. Specifically, it standardizes the data by removing unnecessary headers, footers, and email signatures, and by unifying the text format.

[0482] Step 3:

[0483] The server retrains the generative AI model using pre-processed data. In this process, it learns past patterns and risk factors specific to the organization, improving the accuracy and effectiveness of the AI ​​model.

[0484] Step 4:

[0485] The server inputs the text data of the submitted new contract into an AI model for automatic analysis. Here, it detects deficiencies in the contract terms and legal risks, and tags the relevant sections with risk tags.

[0486] Step 5:

[0487] The server compiles the analysis results and uses them to create a detailed risk report. This report includes details of each detected risk and recommended actions for addressing them.

[0488] Step 6:

[0489] The terminal displays the generated risk report to the user. Based on this information, the user can make decisions regarding contract modifications and risk management.

[0490] Step 7:

[0491] Users review the presented report and provide feedback. This feedback is sent to the server and used to retrain the AI ​​model to further improve its accuracy.

[0492] (Example 1)

[0493] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0494] In modern organizations, legal and security-related operations rely on vast amounts of data. While efficiently managing and analyzing this data is crucial, manual processes are time-consuming, costly, and prone to overlooking risks and making misjudgments. A system is needed to address these challenges and improve the accuracy and speed of operations.

[0495] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0496] In this invention, the server includes information processing means for collecting and centrally managing information, data preprocessing means for standardizing acquired data, and learning means for relearning knowledge from the preprocessed data using a generative model. This enables highly accurate and efficient data analysis and automated risk detection compared to conventional manual work.

[0497] "Information processing means" refers to technologies used to effectively collect and centrally manage data gathered within an organization.

[0498] "Data preprocessing means" refers to processes that perform noise reduction and standardization in order to improve the quality of acquired data.

[0499] A "generative model" is an artificial intelligence model that learns from large amounts of data and generates new information or responses.

[0500] "Learning methods" are techniques used to enable models to acquire new knowledge using collected data, thereby improving the accuracy and usefulness of their responses.

[0501] "Analysis means" refers to a process for automatically analyzing newly acquired document data to detect and classify potential risks.

[0502] "Result generation means" refers to technology for visually displaying analyzed data in an easy-to-understand format and outputting it in document format.

[0503] A "feedback mechanism" is a system that collects opinions and evaluations from users and uses them to improve the system or model.

[0504] A "user interface" is a visual interface that users use to interact with a system, obtain information, and perform operations.

[0505] This invention relates to a system aimed at improving the efficiency of legal and security operations. This system operates primarily through a server to effectively collect and manage information.

[0506] The server accesses an internal database to collect legal and security-related data accumulated to date. This data includes contract documents, past consultation history, and risk assessment records. At this stage, database software is used to efficiently centralize and manage the data.

[0507] Next, the server performs data preprocessing. Here, OCR (Optical Character Recognition) software is used to extract text from image files, and the text is de-noised and standardized. This process converts the data into a format suitable for training the generative AI model.

[0508] Subsequently, the server launches a generative AI model and trains it using the formatted data. This process utilizes machine learning frameworks such as TensorFlow or PyTorch. By learning organization-specific knowledge and industry-specific risk factors, the AI ​​model becomes capable of highly accurate data analysis.

[0509] For example, when reviewing a new contract using an AI model, risks are automatically detected and specific risks are tagged. As a result, deficiencies in contract terms and legal risks are identified in advance, enabling prompt countermeasures.

[0510] Users can use their devices to review risks and areas for improvement detected by the AI. The information is displayed visually, and relevant recommended actions are shown, allowing users to make informed decisions. Users can also provide feedback, which is sent to the server to further improve the accuracy of the AI ​​model.

[0511] As a concrete example, one might input the following prompt into the AI ​​model: "Use past contract data to learn risk patterns and detect risks under specific conditions."

[0512] This system improves the accuracy and speed of legal and security operations, and strengthens the organization's overall risk management capabilities.

[0513] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0514] Step 1:

[0515] The server connects to an internal database to collect legal and security-related data. It uses contracts and consultation histories stored within the company as input. Because this data requires centralized management, it is saved in a specific folder via database software and then converted into a format usable for subsequent processing.

[0516] Step 2:

[0517] The server uses OCR software to extract text from scanned images and performs data preprocessing, including noise reduction and formatting standardization. Using scanned image data as input, it generates clean, standardized text data as output. This process makes the text suitable for AI models.

[0518] Step 3:

[0519] The server supplies the formatted data to the generating AI model and starts the learning process. Machine learning frameworks such as TensorFlow are used. Using clean text data as input, the output is an AI model that has learned organization-specific risk knowledge. The AI ​​model can understand various risk factors and perform highly accurate analysis.

[0520] Step 4:

[0521] The server receives newly submitted contract documents and automatically reviews them using a pre-trained AI model. It receives the new contract as input. The AI ​​analyzes potential risks from the contract document and tags any anomalies. It generates a list of tagged risks as output and sends it to the next step.

[0522] Step 5:

[0523] The terminal displays risk information analyzed by the server to the user. It receives risk information sent from the server as input and displays it on the user interface in a visually organized format. As output, the user can view detailed information about the risk and recommended actions, enabling quick decision-making.

[0524] Step 6:

[0525] Users review the information provided by the system through their terminals and send feedback to the server. They refer to the risk information and recommendations displayed on their terminals as input, and submit opinions and modifications. The feedback for improvement is sent to the server as output and used to improve the accuracy of the model.

[0526] (Application Example 1)

[0527] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0528] In modern society, streamlining legal and security operations is a critical challenge. However, traditional methods require significant time and effort to process vast amounts of document data and identify risks. Furthermore, the slow pace of risk detection and presentation to users leads to delays in decision-making. Moreover, the lack of mechanisms for continuous model improvement utilizing feedback limits the accuracy of existing systems.

[0529] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0530] In this invention, the server includes information processing means for collecting and centrally managing past legal and security-related data accumulated within the company; data processing means for preprocessing the acquired data, removing noise data, and standardizing the text; and learning means for using a generation AI model to retrain based on the preprocessed data and generate responses using organization-specific knowledge. This enables automated review of contract documents and automated risk detection.

[0531] "Information processing means" refers to a device or system for collecting and centrally managing past legal and security-related data accumulated within a company.

[0532] "Data processing means" refers to a device or system for preprocessing acquired data, removing noise data, and standardizing text data.

[0533] A "generative AI model" refers to artificial intelligence technology that learns from data and generates responses to specific tasks.

[0534] A "learning tool" is a device or system that uses a generative AI model to retrain itself based on pre-processed data and generate responses using organization-specific knowledge.

[0535] A "risk detection means" is a device or system that analyzes newly acquired contract document data, automatically detects potential risks, and tags them.

[0536] "Result generation means" refers to a device or system for visualizing risk detection results and outputting them in report format.

[0537] A "feedback acquisition method" is a device or system that acquires feedback from users and uses that information to improve the accuracy of an AI model.

[0538] "Image processing means" refers to a device or system for analyzing image data from multiple input devices and performing optical character recognition.

[0539] "Display output means" refers to a device or system for visually displaying results based on analyzed text data.

[0540] The system for carrying out this invention consists of information processing means, data processing means, learning means using a generated AI model, risk detection means, result generation means, feedback acquisition means, image processing means, and display output means. Each of these means is described in detail below.

[0541] First, the server collects legal and security-related data accumulated within the company and uses an information processing system that centrally manages it in a single database. At this stage, more efficient data management becomes possible.

[0542] Next, the server preprocesses the acquired data, using data processing techniques to remove noise and standardize the text. This facilitates analysis by the generative AI model.

[0543] Generative AI models are used to learn organization-specific knowledge from pre-processed data, preparing for future contract reviews. OpenAI and Google AI Platform can be used at this stage.

[0544] For newly acquired contract documents, the image data is analyzed using image processing tools and converted into text data through optical character recognition. Image processing libraries such as OpenCV are often used for this purpose.

[0545] The risk detection system uses a generative AI model to analyze contract documents, identify potential risks, and tag them. The tagged risk information is then visualized and output in report format by the results generation system.

[0546] The device presents the results to the user, visually displaying detected risks and recommended actions. A user-friendly interface can be provided by using front-end frameworks such as Flutter or React.

[0547] Furthermore, user feedback is collected through feedback acquisition methods and used to continuously improve the accuracy of the AI ​​model.

[0548] A concrete example of its application is a process where, for instance, when a company creates a new internal policy contract, it uses "Secure Legal Review" technology to photograph the contract and then performs AI-powered risk detection. The AI ​​automatically detects potential problems within the contract and notifies the user along with recommended actions.

[0549] An example of a prompt for a generative AI model would be: "Detect potential risks within the contract and determine whether the clauses related to 'payment terms' are appropriate."

[0550] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0551] Step 1:

[0552] The server collects legal and security-related data accumulated within the company. This data includes contracts, consultation histories, and risk assessment reports. The entered data is centrally managed by information processing equipment and stored in a database.

[0553] Step 2:

[0554] The server preprocesses the acquired data. It uses data processing techniques to remove noise and standardize the text. Specifically, it removes unnecessary document formatting information and extracts pure text data. This results in data that is suitable for analysis by the generative AI model.

[0555] Step 3:

[0556] The server uses a generative AI model to learn organization-specific knowledge from pre-processed data. This AI model is built using frameworks such as TensorFlow and generates responses that can be used in future contract reviews throughout the learning process. The output of this step is the trained model.

[0557] Step 4:

[0558] The user inputs an image of a newly acquired contract document into the terminal. The terminal analyzes the image data using image processing equipment and performs OCR to convert it into text data. Using image processing libraries such as OpenCV, it extracts document information from the contract and generates the necessary text data.

[0559] Step 5:

[0560] The server analyzes the OCR-processed text data using risk detection methods. A generative AI model is used to automatically detect potential risks and tag them. As a result of the analysis, data classified according to the type of risk is output.

[0561] Step 6:

[0562] The server visualizes the risk detection results and outputs them in report format using a results generation mechanism. The output report includes the detected risk information and recommended actions based on it.

[0563] Step 7:

[0564] The device visually presents the user with detected risk information and recommended actions. Frontend frameworks such as Flutter and React are used to display the information clearly on the user interface.

[0565] Step 8:

[0566] Users provide feedback based on the results via their devices. This feedback is sent to a server via a feedback acquisition system and used to improve the accuracy of the AI ​​model.

[0567] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0568] This invention combines a system designed to improve efficiency in legal and security operations with an emotion engine that recognizes user emotions. This system provides efficient and user-friendly support through the following process.

[0569] Data collection and management

[0570] The server collects legal and security-related data from various internal sources and stores it in a database. This centralized management ensures data availability and integrity.

[0571] Data preprocessing

[0572] The server preprocesses the accumulated data, removing noise and standardizing the text data. This process makes the data suitable for efficient training by AI models.

[0573] AI model training

[0574] The server uses organized data to train a generative AI model. In particular, by acquiring organization-specific knowledge and industry-specific risk patterns, it becomes possible to respond in a way that is tailored to the organization's needs.

[0575] Automated contract review and risk detection

[0576] The server analyzes newly acquired contract data using an AI model to automatically detect potential risks. Risk points are tagged, and analysis is performed according to their importance. For example, it can identify ambiguity in payment terms and deficiencies in legal liability.

[0577] Emotional engine integration

[0578] The server analyzes user feedback using an emotion engine. This retrieves user emotion data at the time of feedback, which is then used to improve the system's response.

[0579] Presentation of results and feedback processing

[0580] The terminal presents the analysis results to the user, offering recommended actions along with detailed risk information. The presentation of results takes the user's emotions into consideration, adjusting the priority and format of information accordingly. For example, if the user expresses negative emotions, the server will make the explanation more helpful and detailed.

[0581] Improving model accuracy

[0582] User feedback and corresponding emotional data are stored on the server and used to improve the accuracy of the AI ​​model. This continuous improvement process enables the system to provide even greater value in the user experience.

[0583] This configuration improves the efficiency of contract review and risk management, while also enabling flexible responses that take user sentiment into consideration.

[0584] The following describes the processing flow.

[0585] Step 1:

[0586] The server collects historical legal and security-related data from the company's information systems and databases. The collected data is centralized and prepared as a foundation for subsequent processing.

[0587] Step 2:

[0588] The server preprocesses the collected data. Specifically, it removes unnecessary information and noise and converts the text data into a standard format. Improving the quality of the data at this stage enhances the accuracy of analysis by the AI ​​model.

[0589] Step 3:

[0590] The server uses the formatted text data to retrain the generative AI model. This process strengthens the system's knowledge base by learning industry-specific risk patterns and legal terminology.

[0591] Step 4:

[0592] The server analyzes newly submitted contract data through an AI model. While scanning the contract content, it detects potential risks and tags them according to their importance. For example, it identifies ambiguous clauses and potential legal issues.

[0593] Step 5:

[0594] The server uses a results generation mechanism to compile the risk detection results into a report. The report includes recommended actions for the user and details of the risks, providing a basis for future decision-making.

[0595] Step 6:

[0596] The terminal presents the user with a risk report. The user interface visually organizes the information and prompts the user to take necessary actions to aid their understanding.

[0597] Step 7:

[0598] The device and the emotion engine work together to analyze the user's emotions when they review risk reports. This data provides valuable feedback for improving the user experience.

[0599] Step 8:

[0600] Users provide feedback. The emotional information obtained here is sent to the server and used to adjust and improve the AI ​​model. This feedback loop continuously improves the system's accuracy and user satisfaction.

[0601] (Example 2)

[0602] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0603] While there is a need for increased efficiency and accuracy in legal and security management operations, conventional systems struggle to collect and manage large amounts of data, making it difficult to quickly identify new risk factors. Furthermore, there is a lack of systems that reflect user feedback in real time, hindering the implementation of flexible responses based on user sentiment.

[0604] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0605] In this invention, the server includes information gathering means for collecting and integrating a wide range of legal and security-related information from information sources; data processing means for removing interfering information by preprocessing the acquired information and arranging the information in a unified format; and a learning device that uses a generated machine learning model to perform further learning based on the processed information and generate responses using the organization's unique knowledge. This enables efficient and accurate risk identification and quick, flexible responses while considering user sentiment.

[0606] "Information sources" refer to means of providing diverse data related to legal and security management, including internal document management systems and communication systems.

[0607] An "information gathering method" is a system for centrally collecting and integrating data from various information sources.

[0608] "Data processing means" refers to methods for removing noise and unnecessary information from collected data and preparing the information in a format that can be analyzed.

[0609] A "machine learning model" is a framework that includes algorithms that use artificial intelligence technology to learn patterns and rules from data and generate responses and predictions based on that learning.

[0610] A "learning device" is a device that uses machine learning models to retrain based on processed data and generate responses using organization-specific knowledge.

[0611] A "risk detection device" is a device that analyzes newly acquired document data and automatically detects and assigns attributes to potential risks.

[0612] A "results generation device" is a device that visually displays the results of detected risks and outputs them in a report format.

[0613] A "response acquisition device" is a device that effectively acquires user feedback and uses that feedback to improve the accuracy of machine learning models.

[0614] "Emotional analysis tools" are means of analyzing user feedback and obtaining emotional information at that time.

[0615] This invention consists of a system that performs a series of processes including data collection, preprocessing, analysis, result presentation, and improvement based on feedback. Specifically, the server, terminal, and user elements each play their respective roles, enabling efficient support for legal and security operations.

[0616] First, the server automatically collects a wide range of legal and security-related information from internal sources. This utilizes hardware and software such as document management systems and mail servers over the network. The collected data is then processed using data processing tools to standardize the text, remove noise, and organize it into a unified format.

[0617] Next, the server uses the pre-processed data to train a generative AI model. This training process utilizes pattern recognition techniques to incorporate industry-specific hazard patterns and specific legal terminology into the model, enabling responses that leverage the organization's unique knowledge. The AI ​​model analyzes newly acquired agreement document information and automatically identifies potential risks using hazard detection devices.

[0618] The analysis results are presented to the user via a terminal. The terminal's interface, through a results generation device, visualizes risks and outputs them in report format. Information prioritization and presentation are adjusted according to the user's emotions, enabling more effective decision-making. For example, if the user expresses negative emotions, the server presents the information in a more user-friendly and easy-to-understand manner.

[0619] Finally, user feedback is received by the server's response acquisition device, and the user's emotional information is extracted by emotion analysis. This feedback information is used to improve the accuracy of the AI ​​model, and the model is continuously improved.

[0620] For example, by using a prompt such as, "Review the new contract, analyze potential risks, and create a prompt to consider improvements based on the feedback," more detailed and accurate analysis results can be obtained. In this way, the system continues to provide fast and reliable support in legal and security management.

[0621] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0622] Step 1:

[0623] The server collects legal and security-related information from internal sources. Inputs include data from multiple databases, document management systems, and mail servers. The server automatically integrates this data to create a centralized dataset. The output is a unified database. Specifically, scheduled tasks are set to periodically collect data, ensuring that the most up-to-date information is accumulated.

[0624] Step 2:

[0625] The server preprocesses the collected data. The database integrated in step 1 is used as input. Denoising and standardization are performed to normalize text and numerical data with inconsistent formats. The output is a processed, clean dataset. Specifically, this involves text filtering using regular expressions and imputation of missing values.

[0626] Step 3:

[0627] The server trains a generative AI model using preprocessed data. The input is the clean data obtained in step 2. The processed data is fed to the AI ​​model, and it performs learning for pattern recognition and risk prediction. The output is an AI model that incorporates organization-specific knowledge. Specifically, it uses supervised learning methods to extract meaningful patterns from past data.

[0628] Step 4:

[0629] The server analyzes new contract data using an AI model to automatically detect potential risks. The input consists of new document data such as agreements and contracts. The AI ​​model analyzes these documents, identifies and tags risk factors. The output is an analysis result with visualized risks. Specifically, it classifies and prioritizes risks according to their importance.

[0630] Step 5:

[0631] The terminal presents the analysis results to the user. The risk analysis results obtained in step 4 are used as input. The terminal provides the user with visualized information via its display and presents recommended actions. The output provides information in a format that is easy for the user to understand. Specifically, text and graphs are displayed via the interface, and links to detailed information are provided as needed.

[0632] Step 6:

[0633] The user provides feedback based on the information presented. The input is the analysis information provided in step 5. Through the feedback, the user inputs evaluations and impressions into the server, which are used to improve the system. The output is the feedback data sent to the server. Specifically, information is entered using a feedback form.

[0634] Step 7:

[0635] The server receives feedback from users and uses it to improve the accuracy of the AI ​​model. The input is the feedback data obtained in step 6. Sentiment analysis is used to extract user emotion information and utilize it to improve the AI ​​model. The output is an updated AI model. Specifically, continuous training data is added and the model is retrained.

[0636] (Application Example 2)

[0637] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server," and the headset-type terminal 314 will be referred to as the "terminal."

[0638] In on-site security operations, real-time detection of risks and immediate response are required. However, conventional systems suffer from fragmented and inefficient information gathering and analysis, often relying on human judgment under stressful circumstances. As a result, the accuracy of risk management decreases, and the mental burden on users increases.

[0639] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0640] In this invention, the server includes data processing means for collecting and centrally managing historical information-related data accumulated within the company; information processing means for pre-processing the acquired data, removing unnecessary data, and standardizing the information; and learning means for retraining a generation AI model based on the pre-processed data to generate responses using organization-specific knowledge. This enables real-time automatic detection of risks and the presentation of appropriate countermeasures, and further improves work efficiency and reduces mental burden by providing support that takes into account the user's emotional data.

[0641] "Data processing means" refers to a device or method for collecting and centrally managing historical information-related data accumulated within a company.

[0642] "Information processing means" refers to an apparatus or method for pre-processing acquired data, removing unnecessary data, and organizing the information into a standardized format.

[0643] A "learning tool" is a device or method for generating responses using organization-specific knowledge by retraining a generative AI model based on pre-processed data.

[0644] A "risk detection means" is a device or method for analyzing newly acquired information document data, automatically detecting potential hazards, and labeling them.

[0645] A "risk assessment tool" is a device or method for analyzing video and audio data collected in real time to evaluate the safety of a site.

[0646] "Emotion analysis means" refers to a device or method for acquiring user feedback and user emotion data and using them to improve the accuracy of an AI model.

[0647] The system of this invention is designed to provide real-time risk detection and user support in security operations. In this system, a server primarily handles information processing, while terminals such as smart devices provide on-site data collection and interfaces.

[0648] The server collects and centrally manages historical information data accumulated within the company using data processing tools. Next, information processing tools preprocess the acquired data to remove unnecessary data and arrange it in a standardized format. Based on this standardized data, a learning tool retrains using a generation AI model, utilizing organization-specific knowledge in its responses. Furthermore, a risk detection tool analyzes new information document data, automatically detects potential risks, and labels them.

[0649] The terminal collects video and audio data in real time at the site using risk assessment tools and transmits it to the server. The server analyzes this data, assesses the safety of the site, and proposes immediate countermeasures. The sentiment analysis tool acquires user feedback and sentiment data and uses it to continuously improve the accuracy of the AI ​​model.

[0650] For example, suppose a security staff member wearing smart glasses is patrolling a facility when a sudden unusual sound is detected. The server analyzes the anomaly using risk assessment tools and determines that it "may be an animal or a small moving object," then issues instructions to the on-site staff. At the same time, it analyzes the staff member's stress level from their voice and gestures using emotion analysis tools and suggests, "Take a three-second deep breath" to encourage calm behavior.

[0651] An example of a prompt is, "Explain how this security system detects on-site hazards and supports the mental health of personnel." This prompt allows the AI ​​model to learn how to respond in specific scenarios.

[0652] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0653] Step 1:

[0654] The terminal acquires video and audio in real time as on-site monitoring data and transmits it to the server. It receives raw data from devices such as cameras and microphones as input and converts it into a processable data format as output.

[0655] Step 2:

[0656] The server processes the received video and audio data as input to a risk assessment system. This process utilizes a generative AI model to analyze specific movements and sounds. The output identifies potential hazards, and labels are applied as needed. Specific actions include detecting suspicious behavior and unusual sounds.

[0657] Step 3:

[0658] The server uses information processing tools to apply standardized data from existing databases to perform matching with new data and anomaly detection. It receives pre-processed data as input and generates information about potential risks as output.

[0659] Step 4:

[0660] The server uses emotion analysis tools to analyze user input data, such as voice tone and facial expression data. Based on this input, a generative AI model performs analysis, and the user's emotional state data is generated as output. Specifically, stress levels and anxiety levels are determined.

[0661] Step 5:

[0662] The server prepares feedback for the user based on the analysis results. It uses risk assessment and emotional state data as input and constructs specific action suggestions and guidance for the user as output. Specific examples include behavioral instructions tailored to the security situation and relaxation advice.

[0663] Step 6:

[0664] The terminal receives feedback from the server and presents the results visually or audibly through the user interface. It receives feedback data based on input and generates screen displays or audio messages as output.

[0665] The specific processing unit 290 transmits the result of the specific processing to the headset terminal 314. In the headset terminal 314, the control unit 46A causes the speaker 240 and display 343 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0666] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0667] In the above embodiment, an example was given in which specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and specific processing may also be performed by the headset terminal 314.

[0668] [Fourth Embodiment]

[0669] Figure 7 shows an example of the configuration of the data processing system 410 according to the fourth embodiment.

[0670] As shown in Figure 7, the data processing system 410 includes a data processing device 12 and a robot 414. An example of the data processing device 12 is a server.

[0671] The data processing device 12 comprises a computer 22, a database 24, and a communication interface 26. The computer 22 is an example of a "computer" related to the technology of this disclosure. The computer 22 comprises a processor 28, RAM 30, and storage 32. The processor 28, RAM 30, and storage 32 are connected to a bus 34. The database 24 and the communication interface 26 are also connected to the bus 34. The communication interface 26 is connected to a network 54. An example of the network 54 is a WAN (Wide Area Network) and / or a LAN (Local Area Network).

[0672] The robot 414 includes a computer 36, a microphone 238, a speaker 240, a camera 42, a communication interface 44, and a controlled object 443. The computer 36 includes a processor 46, RAM 48, and storage 50. The processor 46, RAM 48, and storage 50 are connected to a bus 52. The microphone 238, speaker 240, camera 42, and controlled object 443 are also connected to the bus 52.

[0673] The microphone 238 receives voice signals from the user 20 and receives instructions from the user 20. The microphone 238 captures the voice signals from the user 20, converts the captured voice into audio data, and outputs it to the processor 46. The speaker 240 outputs audio according to the instructions from the processor 46.

[0674] Camera 42 is a small digital camera equipped with an optical system including a lens, aperture, and shutter, and an image sensor such as a CMOS (Complementary Metal-Oxide-Semiconductor) image sensor or a CCD (Charge Coupled Device) image sensor, and captures images of the area around the user 20 (for example, an imaging range defined by a field of view equivalent to the width of a typical healthy person's field of vision).

[0675] Communication interface 44 is connected to network 54. Communication interfaces 44 and 26 are responsible for the exchange of various information between processor 46 and processor 28 via network 54. The exchange of various information between processor 46 and processor 28 using communication interfaces 44 and 26 is performed in a secure manner.

[0676] The controlled object 443 includes a display device, LEDs in the eyes, and motors that drive the arms, hands, and feet. The posture and gestures of the robot 414 are controlled by controlling the motors of the arms, hands, and feet. Some of the robot 414's emotions can be expressed by controlling these motors. Furthermore, the robot 414's facial expressions can also be expressed by controlling the illumination state of the LEDs in its eyes.

[0677] Figure 8 shows an example of the main functions of the data processing device 12 and the robot 414. As shown in Figure 8, the data processing device 12 performs specific processing using the processor 28. The storage 32 stores the specific processing program 56.

[0678] The specific processing program 56 is an example of a "program" relating to the technology of this disclosure. The processor 28 reads the specific processing program 56 from the storage 32 and executes the read specific processing program 56 on the RAM 30. The specific processing is realized by the processor 28 operating as a specific processing unit 290 in accordance with the specific processing program 56 executed on the RAM 30.

[0679] The storage 32 stores the data generation model 58 and the emotion identification model 59. The data generation model 58 and the emotion identification model 59 are used by the identification processing unit 290.

[0680] In robot 414, the processor 46 performs the reception output processing. The storage 50 stores the reception output program 60. The processor 46 reads the reception output program 60 from the storage 50 and executes the read reception output program 60 on the RAM 48. The reception output processing is realized by the processor 46 operating as a control unit 46A according to the reception output program 60 executed on the RAM 48.

[0681] Next, the specific processing performed by the specific processing unit 290 of the data processing device 12 will be described. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0682] This invention is configured as a system for improving the efficiency of legal and security operations. Specific embodiments of this system are described below.

[0683] Data collection and management

[0684] The server collects legal and security-related data accumulated within the company over time. This data includes contracts, past consultation histories, and risk assessment reports. This data is centrally managed in a database.

[0685] Data preprocessing

[0686] The server removes unnecessary noise from the collected data and standardizes the text. This allows the AI ​​model to learn more efficiently. For example, it removes unnecessary document formatting information and extracts pure text data.

[0687] AI model training

[0688] The server uses a generative AI model that learns from organized data. This AI model learns organization-specific legal knowledge and industry-specific risk factors to achieve optimal analytical capabilities.

[0689] Automated contract review and risk detection

[0690] The server uses a trained AI model to automatically review newly submitted contracts. This review process can automatically detect potential risks hidden within the contract and tag them according to their type. For example, it can identify deficiencies in contract terms and potential legal risks.

[0691] Presentation of results and collection of feedback

[0692] The device presents the user with risks and review points detected by the AI. Each risk is accompanied by relevant information and recommended actions, allowing the user to make decisions based on this information. Simultaneously, the user provides feedback on the review results, which is sent to the server for model improvement.

[0693] This system aims to streamline contract review and risk management, supporting users in making quick and accurate decisions. Through its implementation methods, it improves the accuracy and speed of legal and security-related tasks.

[0694] The following describes the processing flow.

[0695] Step 1:

[0696] The server collects historical legal and security-related data from various sources within the company and stores it centrally in a database. This ensures that the necessary data is efficiently accessible.

[0697] Step 2:

[0698] The server performs preprocessing to remove noise from the collected data. Specifically, it standardizes the data by removing unnecessary headers, footers, and email signatures, and by unifying the text format.

[0699] Step 3:

[0700] The server retrains the generative AI model using pre-processed data. In this process, it learns past patterns and risk factors specific to the organization, improving the accuracy and effectiveness of the AI ​​model.

[0701] Step 4:

[0702] The server inputs the text data of the submitted new contract into an AI model for automatic analysis. Here, it detects deficiencies in the contract terms and legal risks, and tags the relevant sections with risk tags.

[0703] Step 5:

[0704] The server compiles the analysis results and uses them to create a detailed risk report. This report includes details of each detected risk and recommended actions for addressing them.

[0705] Step 6:

[0706] The terminal displays the generated risk report to the user. Based on this information, the user can make decisions regarding contract modifications and risk management.

[0707] Step 7:

[0708] Users review the presented report and provide feedback. This feedback is sent to the server and used to retrain the AI ​​model to further improve its accuracy.

[0709] (Example 1)

[0710] Next, we will describe Example 1. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0711] In modern organizations, legal and security-related operations rely on vast amounts of data. While efficiently managing and analyzing this data is crucial, manual processes are time-consuming, costly, and prone to overlooking risks and making misjudgments. A system is needed to address these challenges and improve the accuracy and speed of operations.

[0712] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 1 is realized by the following means.

[0713] In this invention, the server includes information processing means for collecting and centrally managing information, data preprocessing means for standardizing acquired data, and learning means for relearning knowledge from the preprocessed data using a generative model. This enables highly accurate and efficient data analysis and automated risk detection compared to conventional manual work.

[0714] "Information processing means" refers to technologies used to effectively collect and centrally manage data gathered within an organization.

[0715] "Data preprocessing means" refers to processes that perform noise reduction and standardization in order to improve the quality of acquired data.

[0716] A "generative model" is an artificial intelligence model that learns from large amounts of data and generates new information or responses.

[0717] "Learning methods" are techniques used to enable models to acquire new knowledge using collected data, thereby improving the accuracy and usefulness of their responses.

[0718] "Analysis means" refers to a process for automatically analyzing newly acquired document data to detect and classify potential risks.

[0719] "Result generation means" refers to technology for visually displaying analyzed data in an easy-to-understand format and outputting it in document format.

[0720] A "feedback mechanism" is a system that collects opinions and evaluations from users and uses them to improve the system or model.

[0721] A "user interface" is a visual interface that users use to interact with a system, obtain information, and perform operations.

[0722] This invention relates to a system aimed at improving the efficiency of legal and security operations. This system operates primarily through a server to effectively collect and manage information.

[0723] The server accesses an internal database to collect legal and security-related data accumulated to date. This data includes contract documents, past consultation history, and risk assessment records. At this stage, database software is used to efficiently centralize and manage the data.

[0724] Next, the server performs data preprocessing. Here, OCR (Optical Character Recognition) software is used to extract text from image files, and the text is de-noised and standardized. This process converts the data into a format suitable for training the generative AI model.

[0725] Subsequently, the server launches a generative AI model and trains it using the formatted data. This process utilizes machine learning frameworks such as TensorFlow or PyTorch. By learning organization-specific knowledge and industry-specific risk factors, the AI ​​model becomes capable of highly accurate data analysis.

[0726] For example, when reviewing a new contract using an AI model, risks are automatically detected and specific risks are tagged. As a result, deficiencies in contract terms and legal risks are identified in advance, enabling prompt countermeasures.

[0727] Users can use their devices to review risks and areas for improvement detected by the AI. The information is displayed visually, and relevant recommended actions are shown, allowing users to make informed decisions. Users can also provide feedback, which is sent to the server to further improve the accuracy of the AI ​​model.

[0728] As a concrete example, one might input the following prompt into the AI ​​model: "Use past contract data to learn risk patterns and detect risks under specific conditions."

[0729] This system improves the accuracy and speed of legal and security operations, and strengthens the organization's overall risk management capabilities.

[0730] The flow of the specific processing in Example 1 will be explained using Figure 11.

[0731] Step 1:

[0732] The server connects to an internal database to collect legal and security-related data. It uses contracts and consultation histories stored within the company as input. Because this data requires centralized management, it is saved in a specific folder via database software and then converted into a format usable for subsequent processing.

[0733] Step 2:

[0734] The server uses OCR software to extract text from scanned images and performs data preprocessing, including noise reduction and formatting standardization. Using scanned image data as input, it generates clean, standardized text data as output. This process makes the text suitable for AI models.

[0735] Step 3:

[0736] The server supplies the formatted data to the generating AI model and starts the learning process. Machine learning frameworks such as TensorFlow are used. Using clean text data as input, the output is an AI model that has learned organization-specific risk knowledge. The AI ​​model can understand various risk factors and perform highly accurate analysis.

[0737] Step 4:

[0738] The server receives newly submitted contract documents and automatically reviews them using a pre-trained AI model. It receives the new contract as input. The AI ​​analyzes potential risks from the contract document and tags any anomalies. It generates a list of tagged risks as output and sends it to the next step.

[0739] Step 5:

[0740] The terminal displays risk information analyzed by the server to the user. It receives risk information sent from the server as input and displays it on the user interface in a visually organized format. As output, the user can view detailed information about the risk and recommended actions, enabling quick decision-making.

[0741] Step 6:

[0742] Users review the information provided by the system through their terminals and send feedback to the server. They refer to the risk information and recommendations displayed on their terminals as input, and submit opinions and modifications. The feedback for improvement is sent to the server as output and used to improve the accuracy of the model.

[0743] (Application Example 1)

[0744] Next, we will explain Application Example 1. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0745] In modern society, streamlining legal and security operations is a critical challenge. However, traditional methods require significant time and effort to process vast amounts of document data and identify risks. Furthermore, the slow pace of risk detection and presentation to users leads to delays in decision-making. Moreover, the lack of mechanisms for continuous model improvement utilizing feedback limits the accuracy of existing systems.

[0746] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 1 is realized by the following means.

[0747] In this invention, the server includes information processing means for collecting and centrally managing past legal and security-related data accumulated within the company; data processing means for preprocessing the acquired data, removing noise data, and standardizing the text; and learning means for using a generation AI model to retrain based on the preprocessed data and generate responses using organization-specific knowledge. This enables automated review of contract documents and automated risk detection.

[0748] "Information processing means" refers to a device or system for collecting and centrally managing past legal and security-related data accumulated within a company.

[0749] "Data processing means" refers to a device or system for preprocessing acquired data, removing noise data, and standardizing text data.

[0750] A "generative AI model" refers to artificial intelligence technology that learns from data and generates responses to specific tasks.

[0751] A "learning tool" is a device or system that uses a generative AI model to retrain itself based on pre-processed data and generate responses using organization-specific knowledge.

[0752] A "risk detection means" is a device or system that analyzes newly acquired contract document data, automatically detects potential risks, and tags them.

[0753] "Result generation means" refers to a device or system for visualizing risk detection results and outputting them in report format.

[0754] A "feedback acquisition method" is a device or system that acquires feedback from users and uses that information to improve the accuracy of an AI model.

[0755] "Image processing means" refers to a device or system for analyzing image data from multiple input devices and performing optical character recognition.

[0756] "Display output means" refers to a device or system for visually displaying results based on analyzed text data.

[0757] The system for carrying out this invention consists of information processing means, data processing means, learning means using a generated AI model, risk detection means, result generation means, feedback acquisition means, image processing means, and display output means. Each of these means is described in detail below.

[0758] First, the server collects legal and security-related data accumulated within the company and uses an information processing system that centrally manages it in a single database. At this stage, more efficient data management becomes possible.

[0759] Next, the server preprocesses the acquired data, using data processing techniques to remove noise and standardize the text. This facilitates analysis by the generative AI model.

[0760] Generative AI models are used to learn organization-specific knowledge from pre-processed data, preparing for future contract reviews. OpenAI and Google AI Platform can be used at this stage.

[0761] For newly acquired contract documents, the image data is analyzed using image processing tools and converted into text data through optical character recognition. Image processing libraries such as OpenCV are often used for this purpose.

[0762] The risk detection system uses a generative AI model to analyze contract documents, identify potential risks, and tag them. The tagged risk information is then visualized and output in report format by the results generation system.

[0763] The device presents the results to the user, visually displaying detected risks and recommended actions. A user-friendly interface can be provided by using front-end frameworks such as Flutter or React.

[0764] Furthermore, user feedback is collected through feedback acquisition methods and used to continuously improve the accuracy of the AI ​​model.

[0765] A concrete example of its application is a process where, for instance, when a company creates a new internal policy contract, it uses "Secure Legal Review" technology to photograph the contract and then performs AI-powered risk detection. The AI ​​automatically detects potential problems within the contract and notifies the user along with recommended actions.

[0766] An example of a prompt for a generative AI model would be: "Detect potential risks within the contract and determine whether the clauses related to 'payment terms' are appropriate."

[0767] The flow of a specific process in Application Example 1 will be explained using Figure 12.

[0768] Step 1:

[0769] The server collects legal and security-related data accumulated within the company. This data includes contracts, consultation histories, and risk assessment reports. The entered data is centrally managed by information processing equipment and stored in a database.

[0770] Step 2:

[0771] The server preprocesses the acquired data. It uses data processing techniques to remove noise and standardize the text. Specifically, it removes unnecessary document formatting information and extracts pure text data. This results in data that is suitable for analysis by the generative AI model.

[0772] Step 3:

[0773] The server uses a generative AI model to learn organization-specific knowledge from pre-processed data. This AI model is built using frameworks such as TensorFlow and generates responses that can be used in future contract reviews throughout the learning process. The output of this step is the trained model.

[0774] Step 4:

[0775] The user inputs an image of a newly acquired contract document into the terminal. The terminal analyzes the image data using image processing equipment and performs OCR to convert it into text data. Using image processing libraries such as OpenCV, it extracts document information from the contract and generates the necessary text data.

[0776] Step 5:

[0777] The server analyzes the OCR-processed text data using risk detection methods. A generative AI model is used to automatically detect potential risks and tag them. As a result of the analysis, data classified according to the type of risk is output.

[0778] Step 6:

[0779] The server visualizes the risk detection results and outputs them in report format using a results generation mechanism. The output report includes the detected risk information and recommended actions based on it.

[0780] Step 7:

[0781] The device visually presents the user with detected risk information and recommended actions. Frontend frameworks such as Flutter and React are used to display the information clearly on the user interface.

[0782] Step 8:

[0783] Users provide feedback based on the results via their devices. This feedback is sent to a server via a feedback acquisition system and used to improve the accuracy of the AI ​​model.

[0784] Furthermore, an emotion engine that estimates the user's emotions may be incorporated. That is, the identification processing unit 290 may use the emotion identification model 59 to estimate the user's emotions and perform identification processing using the user's emotions.

[0785] This invention combines a system designed to improve efficiency in legal and security operations with an emotion engine that recognizes user emotions. This system provides efficient and user-friendly support through the following process.

[0786] Data collection and management

[0787] The server collects legal and security-related data from various internal sources and stores it in a database. This centralized management ensures data availability and integrity.

[0788] Data preprocessing

[0789] The server preprocesses the accumulated data, removing noise and standardizing the text data. This process makes the data suitable for efficient training by AI models.

[0790] AI model training

[0791] The server uses organized data to train a generative AI model. In particular, by acquiring organization-specific knowledge and industry-specific risk patterns, it becomes possible to respond in a way that is tailored to the organization's needs.

[0792] Automated contract review and risk detection

[0793] The server analyzes newly acquired contract data using an AI model to automatically detect potential risks. Risk points are tagged, and analysis is performed according to their importance. For example, it can identify ambiguity in payment terms and deficiencies in legal liability.

[0794] Emotional engine integration

[0795] The server analyzes user feedback using an emotion engine. This retrieves user emotion data at the time of feedback, which is then used to improve the system's response.

[0796] Presentation of results and feedback processing

[0797] The terminal presents the analysis results to the user, offering recommended actions along with detailed risk information. The presentation of results takes the user's emotions into consideration, adjusting the priority and format of information accordingly. For example, if the user expresses negative emotions, the server will make the explanation more helpful and detailed.

[0798] Improving model accuracy

[0799] User feedback and corresponding emotional data are stored on the server and used to improve the accuracy of the AI ​​model. This continuous improvement process enables the system to provide even greater value in the user experience.

[0800] This configuration improves the efficiency of contract review and risk management, while also enabling flexible responses that take user sentiment into consideration.

[0801] The following describes the processing flow.

[0802] Step 1:

[0803] The server collects historical legal and security-related data from the company's information systems and databases. The collected data is centralized and prepared as a foundation for subsequent processing.

[0804] Step 2:

[0805] The server preprocesses the collected data. Specifically, it removes unnecessary information and noise and converts the text data into a standard format. Improving the quality of the data at this stage enhances the accuracy of analysis by the AI ​​model.

[0806] Step 3:

[0807] The server uses the formatted text data to retrain the generative AI model. This process strengthens the system's knowledge base by learning industry-specific risk patterns and legal terminology.

[0808] Step 4:

[0809] The server analyzes newly submitted contract data through an AI model. While scanning the contract content, it detects potential risks and tags them according to their importance. For example, it identifies ambiguous clauses and potential legal issues.

[0810] Step 5:

[0811] The server uses a results generation mechanism to compile the risk detection results into a report. The report includes recommended actions for the user and details of the risks, providing a basis for future decision-making.

[0812] Step 6:

[0813] The terminal presents the user with a risk report. The user interface visually organizes the information and prompts the user to take necessary actions to aid their understanding.

[0814] Step 7:

[0815] The device and the emotion engine work together to analyze the user's emotions when they review risk reports. This data provides valuable feedback for improving the user experience.

[0816] Step 8:

[0817] Users provide feedback. The emotional information obtained here is sent to the server and used to adjust and improve the AI ​​model. This feedback loop continuously improves the system's accuracy and user satisfaction.

[0818] (Example 2)

[0819] Next, we will describe Example 2. In the following description, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0820] While there is a need for increased efficiency and accuracy in legal and security management operations, conventional systems struggle to collect and manage large amounts of data, making it difficult to quickly identify new risk factors. Furthermore, there is a lack of systems that reflect user feedback in real time, hindering the implementation of flexible responses based on user sentiment.

[0821] The identification process performed by the identification processing unit 290 of the data processing device 12 in Example 2 is realized by the following means.

[0822] In this invention, the server includes information gathering means for collecting and integrating a wide range of legal and security-related information from information sources; data processing means for removing interfering information by preprocessing the acquired information and arranging the information in a unified format; and a learning device that uses a generated machine learning model to perform further learning based on the processed information and generate responses using the organization's unique knowledge. This enables efficient and accurate risk identification and quick, flexible responses while considering user sentiment.

[0823] "Information sources" refer to means of providing diverse data related to legal and security management, including internal document management systems and communication systems.

[0824] An "information gathering method" is a system for centrally collecting and integrating data from various information sources.

[0825] "Data processing means" refers to methods for removing noise and unnecessary information from collected data and preparing the information in a format that can be analyzed.

[0826] A "machine learning model" is a framework that includes algorithms that use artificial intelligence technology to learn patterns and rules from data and generate responses and predictions based on that learning.

[0827] A "learning device" is a device that uses machine learning models to retrain based on processed data and generate responses using organization-specific knowledge.

[0828] A "risk detection device" is a device that analyzes newly acquired document data and automatically detects and assigns attributes to potential risks.

[0829] A "results generation device" is a device that visually displays the results of detected risks and outputs them in a report format.

[0830] A "response acquisition device" is a device that effectively acquires user feedback and uses that feedback to improve the accuracy of machine learning models.

[0831] "Emotional analysis tools" are means of analyzing user feedback and obtaining emotional information at that time.

[0832] This invention consists of a system that performs a series of processes including data collection, preprocessing, analysis, result presentation, and improvement based on feedback. Specifically, the server, terminal, and user elements each play their respective roles, enabling efficient support for legal and security operations.

[0833] First, the server automatically collects a wide range of legal and security-related information from internal sources. This utilizes hardware and software such as document management systems and mail servers over the network. The collected data is then processed using data processing tools to standardize the text, remove noise, and organize it into a unified format.

[0834] Next, the server uses the pre-processed data to train a generative AI model. This training process utilizes pattern recognition techniques to incorporate industry-specific hazard patterns and specific legal terminology into the model, enabling responses that leverage the organization's unique knowledge. The AI ​​model analyzes newly acquired agreement document information and automatically identifies potential risks using hazard detection devices.

[0835] The analysis results are presented to the user via a terminal. The terminal's interface, through a results generation device, visualizes risks and outputs them in report format. Information prioritization and presentation are adjusted according to the user's emotions, enabling more effective decision-making. For example, if the user expresses negative emotions, the server presents the information in a more user-friendly and easy-to-understand manner.

[0836] Finally, user feedback is received by the server's response acquisition device, and the user's emotional information is extracted by emotion analysis. This feedback information is used to improve the accuracy of the AI ​​model, and the model is continuously improved.

[0837] For example, by using a prompt such as, "Review the new contract, analyze potential risks, and create a prompt to consider improvements based on the feedback," more detailed and accurate analysis results can be obtained. In this way, the system continues to provide fast and reliable support in legal and security management.

[0838] The flow of the specific processing in Example 2 will be explained using Figure 13.

[0839] Step 1:

[0840] The server collects legal and security-related information from internal sources. Inputs include data from multiple databases, document management systems, and mail servers. The server automatically integrates this data to create a centralized dataset. The output is a unified database. Specifically, scheduled tasks are set to periodically collect data, ensuring that the most up-to-date information is accumulated.

[0841] Step 2:

[0842] The server preprocesses the collected data. The database integrated in step 1 is used as input. Denoising and standardization are performed to normalize text and numerical data with inconsistent formats. The output is a processed, clean dataset. Specifically, this involves text filtering using regular expressions and imputation of missing values.

[0843] Step 3:

[0844] The server trains a generative AI model using preprocessed data. The input is the clean data obtained in step 2. The processed data is fed to the AI ​​model, and it performs learning for pattern recognition and risk prediction. The output is an AI model that incorporates organization-specific knowledge. Specifically, it uses supervised learning methods to extract meaningful patterns from past data.

[0845] Step 4:

[0846] The server analyzes new contract data using an AI model to automatically detect potential risks. The input consists of new document data such as agreements and contracts. The AI ​​model analyzes these documents, identifies and tags risk factors. The output is an analysis result with visualized risks. Specifically, it classifies and prioritizes risks according to their importance.

[0847] Step 5:

[0848] The terminal presents the analysis results to the user. The risk analysis results obtained in step 4 are used as input. The terminal provides the user with visualized information via its display and presents recommended actions. The output provides information in a format that is easy for the user to understand. Specifically, text and graphs are displayed via the interface, and links to detailed information are provided as needed.

[0849] Step 6:

[0850] The user provides feedback based on the information presented. The input is the analysis information provided in step 5. Through the feedback, the user inputs evaluations and impressions into the server, which are used to improve the system. The output is the feedback data sent to the server. Specifically, information is entered using a feedback form.

[0851] Step 7:

[0852] The server receives feedback from users and uses it to improve the accuracy of the AI ​​model. The input is the feedback data obtained in step 6. Sentiment analysis is used to extract user emotion information and utilize it to improve the AI ​​model. The output is an updated AI model. Specifically, continuous training data is added and the model is retrained.

[0853] (Application Example 2)

[0854] Next, we will explain application example 2. In the following explanation, the data processing device 12 will be referred to as the "server" and the robot 414 as the "terminal".

[0855] In on-site security operations, real-time detection of risks and immediate response are required. However, conventional systems suffer from fragmented and inefficient information gathering and analysis, often relying on human judgment under stressful circumstances. As a result, the accuracy of risk management decreases, and the mental burden on users increases.

[0856] The specific processing performed by the specific processing unit 290 of the data processing device 12 in Application Example 2 is realized by the following means.

[0857] In this invention, the server includes data processing means for collecting and centrally managing historical information-related data accumulated within the company; information processing means for pre-processing the acquired data, removing unnecessary data, and standardizing the information; and learning means for retraining a generation AI model based on the pre-processed data to generate responses using organization-specific knowledge. This enables real-time automatic detection of risks and the presentation of appropriate countermeasures, and further improves work efficiency and reduces mental burden by providing support that takes into account the user's emotional data.

[0858] "Data processing means" refers to a device or method for collecting and centrally managing historical information-related data accumulated within a company.

[0859] "Information processing means" refers to an apparatus or method for pre-processing acquired data, removing unnecessary data, and organizing the information into a standardized format.

[0860] A "learning tool" is a device or method for generating responses using organization-specific knowledge by retraining a generative AI model based on pre-processed data.

[0861] A "risk detection means" is a device or method for analyzing newly acquired information document data, automatically detecting potential hazards, and labeling them.

[0862] A "risk assessment tool" is a device or method for analyzing video and audio data collected in real time to evaluate the safety of a site.

[0863] "Emotion analysis means" refers to a device or method for acquiring user feedback and user emotion data and using them to improve the accuracy of an AI model.

[0864] The system of this invention is designed to provide real-time risk detection and user support in security operations. In this system, a server primarily handles information processing, while terminals such as smart devices provide on-site data collection and interfaces.

[0865] The server collects and centrally manages historical information data accumulated within the company using data processing tools. Next, information processing tools preprocess the acquired data to remove unnecessary data and arrange it in a standardized format. Based on this standardized data, a learning tool retrains using a generation AI model, utilizing organization-specific knowledge in its responses. Furthermore, a risk detection tool analyzes new information document data, automatically detects potential risks, and labels them.

[0866] The terminal collects video and audio data in real time at the site using risk assessment tools and transmits it to the server. The server analyzes this data, assesses the safety of the site, and proposes immediate countermeasures. The sentiment analysis tool acquires user feedback and sentiment data and uses it to continuously improve the accuracy of the AI ​​model.

[0867] For example, suppose a security staff member wearing smart glasses is patrolling a facility when a sudden unusual sound is detected. The server analyzes the anomaly using risk assessment tools and determines that it "may be an animal or a small moving object," then issues instructions to the on-site staff. At the same time, it analyzes the staff member's stress level from their voice and gestures using emotion analysis tools and suggests, "Take a three-second deep breath" to encourage calm behavior.

[0868] An example of a prompt is, "Explain how this security system detects on-site hazards and supports the mental health of personnel." This prompt allows the AI ​​model to learn how to respond in specific scenarios.

[0869] The flow of a specific process in Application Example 2 will be explained using Figure 14.

[0870] Step 1:

[0871] The terminal acquires video and audio in real time as on-site monitoring data and transmits it to the server. It receives raw data from devices such as cameras and microphones as input and converts it into a processable data format as output.

[0872] Step 2:

[0873] The server processes the received video and audio data as input to a risk assessment system. This process utilizes a generative AI model to analyze specific movements and sounds. The output identifies potential hazards, and labels are applied as needed. Specific actions include detecting suspicious behavior and unusual sounds.

[0874] Step 3:

[0875] The server uses information processing tools to apply standardized data from existing databases to perform matching with new data and anomaly detection. It receives pre-processed data as input and generates information about potential risks as output.

[0876] Step 4:

[0877] The server uses emotion analysis tools to analyze user input data, such as voice tone and facial expression data. Based on this input, a generative AI model performs analysis, and the user's emotional state data is generated as output. Specifically, stress levels and anxiety levels are determined.

[0878] Step 5:

[0879] The server prepares feedback for the user based on the analysis results. It uses risk assessment and emotional state data as input and constructs specific action suggestions and guidance for the user as output. Specific examples include behavioral instructions tailored to the security situation and relaxation advice.

[0880] Step 6:

[0881] The terminal receives feedback from the server and presents the results visually or audibly through the user interface. It receives feedback data based on input and generates screen displays or audio messages as output.

[0882] The specific processing unit 290 transmits the result of the specific processing to the robot 414. In the robot 414, the control unit 46A causes the speaker 240 and the controlled object 443 to output the result of the specific processing. The microphone 238 acquires audio indicating user input for the result of the specific processing. The control unit 46A transmits the audio data indicating user input acquired by the microphone 238 to the data processing unit 12. In the data processing unit 12, the specific processing unit 290 acquires the audio data.

[0883] Data generation model 58 is a type of so-called generative AI (Artificial Intelligence). One example of data generation model 58 is ChatGPT (Internet search<URL: https: / / openai.com / blog / chatgpt> ), Gemini (Internet search) <url: https: gemini.google.com ?hl="ja">Examples of generative AI include the following. The data generation model 58 is obtained by performing deep learning on a neural network. The data generation model 58 is input with prompts containing instructions, and with inference data such as audio data representing speech, text data representing text, and image data representing images. The data generation model 58 infers from the input inference data according to the instructions indicated by the prompts, and outputs the inference results in data formats such as audio data and text data. Here, inference refers to, for example, analysis, classification, prediction, and / or summarization.

[0884] In the above embodiment, an example was given in which the specific processing is performed by the data processing device 12, but the technology of this disclosure is not limited thereto, and the specific processing may also be performed by the robot 414.

[0885] Furthermore, the emotion identification model 59, acting as an emotion engine, may determine the user's emotion according to a specific mapping. Specifically, the emotion identification model 59 may determine the user's emotion according to a specific mapping, which is an emotion map (see Figure 9). Similarly, the emotion identification model 59 may also determine the robot's emotion, and the identification processing unit 290 may perform identification processing using the robot's emotion.

[0886] Figure 9 shows an emotion map 400 in which multiple emotions are mapped. In the emotion map 400, emotions are arranged in concentric circles radiating from the center. The closer to the center of the concentric circles, the more primitive the emotions are located. Further out of the concentric circles, emotions representing states and actions arising from mental states are located. Emotion is a concept that includes feelings and mental states. On the left side of the concentric circles, emotions that are generally generated from reactions occurring in the brain are located. On the right side of the concentric circles, emotions that are generally induced by situational judgment are located. Above and below the concentric circles, emotions that are generally generated from reactions occurring in the brain and induced by situational judgment are located. In addition, the emotion of "pleasure" is located on the upper side of the concentric circles, and the emotion of "displeasure" is located on the lower side. Thus, in the emotion map 400, multiple emotions are mapped based on the structure in which emotions arise, and emotions that are likely to occur simultaneously are mapped close together.

[0887] These emotions are distributed at the 3 o'clock position on the Emotion Map 400, and usually fluctuate between feelings of security and anxiety. In the right half of the Emotion Map 400, situational awareness takes precedence over internal feelings, resulting in a calm impression.

[0888] The inside of the Emotion Map 400 represents inner thoughts, while the outside represents actions. Therefore, the further you go from the outside of the Emotion Map 400, the more visible (expressed in actions) your emotions become.

[0889] Here, human emotions are based on various balances, such as posture and blood sugar levels. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. Similarly, in robots, cars, motorcycles, etc., emotions can be created based on various balances, such as posture and battery level. When these balances deviate from the ideal, it results in discomfort, and when they approach the ideal, it results in pleasure. The emotion map can be generated, for example, based on Dr. Mitsuyoshi's emotion map (Research on a system for analyzing brain physiological signals of speech emotion recognition and emotion, Tokushima University, doctoral dissertation: https: / / ci.nii.ac.jp / naid / 500000375379). The left half of the emotion map contains emotions belonging to a region called "response," where sensation is dominant. The right half of the emotion map contains emotions belonging to a region called "situation," where situational awareness is dominant.

[0890] The emotion map defines two emotions that promote learning. One is the emotion around the middle of the negative "repentance" and "reflection" on the situation side. In other words, it is when the robot experiences negative emotions such as "I never want to feel this way again" or "I don't want to be scolded again." The other is the emotion around the positive "desire" on the reaction side. In other words, it is when the robot has positive feelings such as "I want more" or "I want to know more."

[0891] The emotion identification model 59 inputs user input into a pre-trained neural network, obtains emotion values ​​representing each emotion shown in the emotion map 400, and determines the user's emotion. This neural network is pre-trained based on multiple training data sets, which are combinations of user input and emotion values ​​representing each emotion shown in the emotion map 400. Furthermore, this neural network is trained so that emotions located close together have similar values, as shown in the emotion map 900 in Figure 10. Figure 10 shows an example where multiple emotions such as "reassured," "calm," and "confident" have similar emotion values.

[0892] The above description primarily focuses on the functions of the data processing device 12 in relation to this disclosure. However, the system related to this disclosure is not necessarily implemented on a server. The system related to this disclosure may be implemented as a general information processing system. This disclosure may be implemented, for example, as a software program that runs on a personal computer or as an application that runs on a smartphone. The method related to this disclosure may be provided to users in SaaS (Software as a Service) format.

[0893] In the above embodiment, an example was given in which a specific process is performed by a single computer 22. However, the technology of this disclosure is not limited thereto, and a distributed processing of the specific process may be performed by multiple computers, including computer 22. For example, a data generation model 58 may be provided in an external device of the data processing device 12, and the external device may generate data according to the input data.

[0894] In the above embodiment, an example was given in which the specific processing program 56 is stored in the storage 32, but the technology of this disclosure is not limited thereto. For example, the specific processing program 56 may be stored in a portable, computer-readable, non-temporary storage medium such as a USB (Universal Serial Bus) memory. The specific processing program 56 stored in the non-temporary storage medium is installed in the computer 22 of the data processing device 12. The processor 28 executes specific processing according to the specific processing program 56.

[0895] Alternatively, the specific processing program 56 may be stored in a storage device such as a server connected to the data processing device 12 via the network 54, and the specific processing program 56 may be downloaded and installed on the computer 22 in response to a request from the data processing device 12.

[0896] Furthermore, it is not necessary to store the entirety of the specific processing program 56 in a storage device such as a server connected to the data processing device 12 via the network 54, or to store the entirety of the specific processing program 56 in the storage 32; it is acceptable to store only a portion of the specific processing program 56.

[0897] The following types of processors can be used as hardware resources to perform specific processing. Examples of processors include a CPU, a general-purpose processor that functions as a hardware resource to perform specific processing by executing software, i.e., a program. Other examples of processors include dedicated electrical circuits, such as FPGAs (Field-Programmable Gate Arrays), PLDs (Programmable Logic Devices), or ASICs (Application Specific Integrated Circuits), which have circuit configurations specifically designed to perform specific processing. All of these processors have built-in or connected memory, and all of them perform specific processing by using memory.

[0898] The hardware resource that performs a specific process may consist of one of these various processors, or it may consist of a combination of two or more processors of the same or different types (for example, a combination of multiple FPGAs, or a combination of a CPU and an FPGA). Alternatively, the hardware resource that performs a specific process may consist of a single processor.

[0899] Examples of configurations using a single processor include, firstly, a configuration in which one or more CPUs and software are combined to form a single processor, and this processor functions as a hardware resource that performs a specific process. Secondly, there is a configuration using a processor that realizes the functions of the entire system, including multiple hardware resources that perform a specific process, on a single IC chip, as exemplified by SoCs (System-on-a-chip). In this way, a specific process is realized using one or more of the above types of processors as hardware resources.

[0900] Furthermore, the hardware structure of these various processors can more specifically utilize electrical circuits that combine circuit elements such as semiconductor devices. Also, the specific processing described above is merely an example. Therefore, it goes without saying that unnecessary steps can be deleted, new steps added, or the processing order rearranged, as long as it does not deviate from the main purpose.

[0901] The descriptions and illustrations presented above are detailed explanations of the technical aspects of this disclosure and are merely examples of the technical aspects. For example, the above descriptions of the structure, function, operation, and effect are examples of the structure, function, operation, and effect of the technical aspects of this disclosure. Therefore, it goes without saying that you may delete unnecessary parts, add new elements, or replace elements in the descriptions and illustrations presented above, as long as you do not deviate from the essence of the technical aspects of this disclosure. Furthermore, in order to avoid confusion and facilitate understanding of the technical aspects of this disclosure, explanations of common technical knowledge and the like that do not require special explanation to enable the implementation of the technical aspects of this disclosure have been omitted from the descriptions and illustrations presented above.

[0902] All documents, patent applications, and technical standards described herein are incorporated by reference to the same extent as if each individual document, patent application, and technical standard were specifically and individually noted as being incorporated by reference.

[0903] The following is further disclosed regarding the embodiments described above.

[0904] (Claim 1)

[0905] An information processing system for collecting and centrally managing past legal and security-related data accumulated within the company,

[0906] A data processing method that preprocesses acquired data, removes noise data, and standardizes the text,

[0907] A learning method that uses a generative AI model to retrain based on pre-processed data and generate responses using organization-specific knowledge,

[0908] A risk detection method that analyzes newly acquired contract document data, automatically detects potential risks, and tags them,

[0909] A result generation method that visualizes risk detection results and outputs them in report format,

[0910] A means of obtaining user feedback and using it to improve the accuracy of the AI ​​model,

[0911] A system that includes this.

[0912] (Claim 2)

[0913] The system according to claim 1, comprising a user interface for presenting results and a display means for visually displaying detected risks and areas for improvement.

[0914] (Claim 3)

[0915] The system according to claim 1, wherein the learning means trains a model on industry-specific risk patterns and legal terminology, enabling a response tailored to specific needs.

[0916] "Example 1"

[0917] (Claim 1)

[0918] Information processing means for collecting and centrally managing information,

[0919] A data preprocessing means for standardizing acquired data,

[0920] A learning method that uses a generative model to retrain knowledge from preprocessed data,

[0921] An analytical means that analyzes newly acquired document data, automatically detects and classifies potential risks,

[0922] A result generation means that visualizes the analysis results and outputs them in document format,

[0923] A feedback mechanism to obtain opinions from users and use them to improve the accuracy of the model,

[0924] A system that includes this.

[0925] (Claim 2)

[0926] The system according to claim 1, comprising a user interface for displaying results and a display means for visually displaying detected risks and areas for improvement.

[0927] (Claim 3)

[0928] The system according to claim 1, wherein the learning means trains a model on risk patterns and terminology in a specific domain, enabling a response that corresponds to a specific request.

[0929] "Application Example 1"

[0930] (Claim 1)

[0931] An information processing system for collecting and centrally managing past legal and security-related data accumulated within the company,

[0932] A data processing method that preprocesses acquired data, removes noise data, and standardizes the text,

[0933] A learning method that uses a generative AI model to retrain based on pre-processed data and generate responses using organization-specific knowledge,

[0934] A risk detection method that analyzes newly acquired contract document data, automatically detects potential risks, and tags them,

[0935] A result generation method that visualizes risk detection results and outputs them in report format,

[0936] A means of obtaining user feedback and using it to improve the accuracy of the AI ​​model,

[0937] Image processing means that analyzes image data from multiple input devices and performs optical character recognition,

[0938] A display output means for visually displaying the analyzed text data,

[0939] A system that includes this.

[0940] (Claim 2)

[0941] The system according to claim 1, comprising a user interface for presenting results and display means for visually displaying detected risks and recommended actions.

[0942] (Claim 3)

[0943] The system according to claim 1, wherein the learning means trains a model on industry-specific risk patterns and legal terminology, enabling it to respond to specific requirements.

[0944] "Example 2 of combining an emotion engine"

[0945] (Claim 1)

[0946] Information gathering means for collecting and integrating a wide range of legal and security-related information from various sources,

[0947] A data processing means that removes interfering information by preprocessing the acquired information and arranges the information into a unified format,

[0948] A learning device that uses a generated machine learning model to perform further learning based on the processed information and generates a response using the organization's unique knowledge,

[0949] A hazard detection device that analyzes newly acquired agreement document information and automatically detects and attributes potential hazards,

[0950] A result generation device that visualizes the results of hazard detection and outputs them in a report format,

[0951] A response acquisition device that obtains user responses and uses them to improve the accuracy of machine learning models,

[0952] A system including an emotion analysis means that analyzes responses provided by the user using emotion analysis means and acquires emotion information.

[0953] (Claim 2)

[0954] The system according to claim 1, comprising a display device that provides a user interface for displaying results, visually displays identified risks and areas for improvement, and adjusts the priority and presentation of information according to the user's emotions.

[0955] (Claim 3)

[0956] The system according to claim 1, wherein the learning device trains a model on industry-specific risk patterns and legal terminology, enables responses to specific requests, and utilizes acquired emotional information to improve the response.

[0957] "Application example 2 when combining with an emotional engine"

[0958] (Claim 1)

[0959] A data processing method for collecting and centrally managing historical information-related data accumulated within the company,

[0960] An information processing means that preprocesses acquired data, removes unnecessary data, and standardizes the information,

[0961] A learning method that uses a generative AI model to retrain based on pre-processed data and generate responses using organization-specific knowledge,

[0962] A risk detection means that analyzes newly acquired information document data, automatically detects potential hazards, and labels them,

[0963] A risk assessment method that analyzes video and audio data collected in real time to evaluate on-site safety,

[0964] A sentiment analysis method that acquires user feedback and user sentiment data and uses it to improve the accuracy of the AI ​​model,

[0965] A system that includes this.

[0966] (Claim 2)

[0967] The system according to claim 1, comprising a display means that provides an interface for presenting results, visually displays detected risks and areas for improvement, and makes suggestions based on the user's emotions.

[0968] (Claim 3)

[0969] The system according to claim 1, wherein the learning means trains a model on risk patterns and specialized terminology specific to a particular field, enabling it to respond to specific requests. [Explanation of symbols]

[0970] 10, 210, 310, 410 Data Processing Systems 12 Data Processing Devices 14 Smart Devices 214 Smart Glasses 314 Headset-type terminal 414 Robots< / url:> < / url:> < / url:> < / url:>

Claims

1. An information processing system for collecting and centrally managing past legal and security-related data accumulated within the company, A data processing method that preprocesses acquired data, removes noise data, and standardizes the text, A learning method that uses a generative AI model to retrain based on pre-processed data and generate responses using organization-specific knowledge, A risk detection method that analyzes newly acquired contract document data, automatically detects potential risks, and tags them, A result generation method that visualizes risk detection results and outputs them in report format, A means of obtaining user feedback and using it to improve the accuracy of the AI ​​model, Image processing means that analyzes image data from multiple input devices and performs optical character recognition, A display output means for visually displaying the analyzed text data, A system that includes this.

2. The system according to claim 1, comprising a user interface for presenting results and a display means for visually displaying detected risks and recommended actions.

3. The system according to claim 1, wherein the learning means trains a model on industry-specific risk patterns and legal terminology, enabling it to respond to specific requirements.

Images