103 results about "Specialized knowledge" patented technology

This invention provides a web application system that allows instructors to effectively use generative AI in classes to match learning tasks, reducing the risk of students asking questions that are inappropriate for the task or trusting false information from the generative AI, and encouraging students' creative thinking through dialogue using generative AI, with the aim of supporting creative activities. SOLUTION: The web application system of this invention creates prompts by integrating questions from learners with prompts preset by the instructor, and provides automatically generated answers through API connection. The app is built using HTML code with a website creation tool, so it can be managed without specialized knowledge. Prompts are edited using three items: Theme, Conditions, and Additional Instructions, to control the answers of the generating AI. In addition, restrictions on questions outside the topic can be set to prevent misuse.

The present invention discloses a modular AI-based system for knowledge instantiation and evidence synthesis, unifying five modules: Knowledge Instantiation, ingesting anchor knowledge via retrieval- augmented generation and neuromorphic prompting; Case-Based Agent Generation, orchestrating domain constraints, performance criteria, and sub-task decomposition; Secondary Epistemogenesis, spawning sub-agents with inherited heuristics; Workflow Orchestration and Evidence Synthesis, merging outputs, tracking metadata, and storing final artifacts; and Reinforcement Learning from Human Refinement, capturing feedback and edits. The method for output generation initiates by defining knowledge requirements, indexing public and proprietary knowledge, validating output structure, drafting content, and refining outputs via sub-agent spawning and human oversight. The system ensures domain alignment, concurrency control, and persistent versioning. By combining specialized knowledge sources, multi-agent concurrency, and iterative human feedback loops, it dynamically adapts to evolving knowledge requirements. This invention emphasizes reliability, traceability, and AI-driven outputs in regulated industries, delivering trustworthy outcomes anchored in expert knowledge.

A computer-implemented system for automated document generation receives user input and processes it through artificial intelligence models to create professional documents. The system combines user-provided information with specialized knowledge databases containing domain-specific requirements and formatting standards to generate enhanced processing instructions. Trained machine learning models process these instructions to produce contextually appropriate content that addresses specific technical and legal requirements. The generated content is formatted using configurable templates to create completed documents that meet professional standards. The system provides adaptive user interfaces that adjust functionality based on user expertise levels, supports multiple operational modes including guided and automated generation, learns from example documents to improve output quality, and enables comprehensive workflow management through network-accessible services. Applications include patent applications, legal responses, technical documentation, and other specialized professional documents requiring domain expertise and regulatory compliance.

The invention relates to the technical field of artificial intelligence, and provides a medical knowledge question answering method, system and device based on a large language model, and the method comprises the steps: obtaining a question text inputted by a user; calling a quick question and answer language model to retrieve answers from a vector database according to questions proposed by a user, vector data in the vector database being real number vectors formed according to text fragments related to the digestive specialized knowledge, if the answer is retrieved, filling the question text and the retrieved answer into a prompt template to generate a standardized answer text, and outputting the standardized answer text to the user; and if no answer is retrieved, calling a deep reasoning language model to call an execution tool related to a reasoning task corresponding to the current question according to the question to perform online answer search, filling a question text and a search result into a prompt template to generate a second standardized answer text, and outputting the second standardized answer text to the user. Through cooperative work of the two intelligent models, the accuracy and the reply speed of medical knowledge questions and answers are considered.

To provide a system.SOLUTION: The system includes: means for providing a generative AI platform for a company that prepares an environment dedicated to each company and an environment that can be used by participating companies in common for small and medium-sized companies; means used for learning within a company only while data input by employees is not used for learning in the environment dedicated to the company; and means for preparing specialized knowledge in areas such as accounting and legal affairs in a common environment so that small and medium-sized companies utilize the specialized knowledge.SELECTED DRAWING: Figure 1

The invention relates to the technical field of semantic understanding, in particular to a special knowledge base fragmented slicing and semantic understanding answer matching method, which comprises the following steps of: obtaining document table images, identifying, segmenting and marking as a fragment set, generating a layered weight sequence by combining question comparison condition obligation and consequence aging reasoning, and obtaining a semantic understanding result; the method comprises the following steps of: identifying and segmenting a text to generate a fragment set, analyzing a connection relationship to form a trajectory link, segmenting candidate answers and performing semantic alignment to supplement missing, calculating a score accumulation result according to a weight sequence, adjusting a parameter sequence and priority, and generating and outputting an optimized answer, and comprises the following steps of: identifying and segmenting the text to generate the fragment set, comparing question contents with conditional obligations and combining consequence aging reasoning; a weighted logic chain is formed, a fragment connection relation dominantly constructs a link, candidate answers are segmented, tracks are aligned, a missing part dependency relation is supplemented, semantic precise alignment is performed on the whole, reasoning is coherent and complete, answer coverage is comprehensive, the integrating degree and precision are improved, and effectiveness and integrity in a complex scene are enhanced.

The invention discloses a preoperative discussion record automatic generation method and system based on a knowledge graph and natural language processing. The system automatically obtains multi-modal data from an electronic medical record, inspection and image system, and carries out standardization processing in combination with doctor input; medical entity recognition and relation extraction are realized through natural language processing, and a patient digital portrait is constructed; reasoning and calling a dynamic template engine based on the specialized knowledge graph, and generating a personalized preoperative discussion record; the generated content has interpretability, a doctor can modify and confirm, and continuous optimization is realized by system record feedback. The system is composed of a data input module, a data processing module, an intelligent generation module, an interactive display module and a learning optimization module. According to the method, intelligent, standardized and personalized generation of the preoperative discussion record is realized, and the writing efficiency and the medical safety are improved.

The present invention discloses a mine disaster tracing method based on a knowledge graph, applied in coal mine safety. First, a mine disaster-related knowledge graph is built, involving three main steps: constructing a conceptual model, extracting entities from relational databases and geological maps, and establishing entity relationships based on location and process logic. Next, characteristic indexes and transmission rules for entity objects are set, categorizing entities into four types: discrete reporting, continuous monitoring, geological structure, and geological continuity. When an early warning occurs, disaster tracing is performed using the entity transmission rules and a graph traversal algorithm, identifying direct and root causes. This invention offers systematic and timely disaster tracing by utilizing a specialized knowledge graph and graph algorithms.

The embodiment of the invention provides a multi-modal universal model self-evolution method and system based on incremental learning, and the method comprises the steps: receiving a current target task and multi-modal data corresponding to the current target task, and determining a corresponding trained target universal model; extracting data features of the multi-modal data to obtain multi-modal data features, and processing the multi-modal data features and the multi-modal data through a target special model to obtain a target result; historical multi-modal data features corresponding to the historical multi-modal data are obtained, and general-purpose model parameters of the trained target general-purpose model are updated; and monitoring the performance index of the trained target general-purpose model according to a preset monitoring period, and if the performance index corresponding to the current preset monitoring period does not meet the preset performance index, adjusting the trained target general-purpose model according to a preset adjustment mode. According to the method, the defect of insufficient collaboration between general knowledge and special knowledge is effectively overcome, and the collaboration effect and the updating effect of updating according to new knowledge are remarkably improved.

To efficiently create a contract considering a risk even without having expert knowledge using a system according to an embodiment.SOLUTION: A system according to an embodiment comprises: a reception unit; an analysis unit; a creation unit; and a provision unit. The reception unit receives input of contract content or a contract period. The analysis unit analyzes information input by the reception unit, and evaluates a risk. The creation unit creates a contract on the basis of the risk evaluated by the analysis unit. The provision unit provides the contract created by the creation unit to a user.SELECTED DRAWING: Figure 1

To provide a method of determining interaction information, a device, an electronic apparatus, a storage medium and a computer program which can create questioning dimensions for query information by combining large model previous specialized knowledge and subsequent supplements in user historical query information by means of the method of the chain of thoughts.SOLUTION: A method is to determine a plurality of questioning dimensions according to query information of a subject and historical query information, where each questioning dimension includes a dimension name and a plurality of options. The method is also to determine a target questioning dimension from the plurality of questioning dimensions according to evaluation values of the plurality of questioning dimensions and whether semantic information of the plurality of questioning dimensions are consistent with semantic information of a query result associated with the query information; and determine the interaction information according to the dimension name and the plurality of options in the target questioning dimension.SELECTED DRAWING: Figure 2

To provide a membrane module which even persons poor in specialized knowledge and experience can easily remove and clean a filtration membrane as necessary and which is less likely to be broken by repeated attachment and detachment of the filtration membrane; and a membrane filtration device which has inexpensive production cost and an excellent maintenance property.SOLUTION: A membrane module 100 provided with a filtration membrane includes: a membrane fixing part 2 which fixes an end part 11 of a filtration membrane 1; a housing 3 which houses the filtration membrane 1; and a cap 4 which detachably fixes the housing 3. The housing 3 is detachably attached to the cap 4 in a screw type or a bayonet type. The membrane fixing part 2 and the filtration membrane 1 are detachable concomitantly with detachment of the housing 3. Detachment of the filtration membrane 1 and the membrane fixing part 2 from the housing 3 can make the filtration membrane 1 as a whole in an exposed state.SELECTED DRAWING: Figure 1

To provide an information processing method, and the like, that enables updating an IFC file without requiring expert knowledge.SOLUTION: A computer is configured to acquire update information indicating contents of update on a 3D model for a civil engineering structure generated using building information modeling (BIM) according to a program. The computer is configured to input the acquired update information to a language generation model designed to generate, on receipt of the update information, industry foundation classes (IFC) information related to IFC corresponding to the BIM, to generate IFC information corresponding to the update information.SELECTED DRAWING: Figure 1

The invention discloses a retrieval enhancement generation method and device based on an obstetrics and gynecology specialized knowledge base. The server side method comprises the following steps: in response to a retrieval request sent by a client side, preprocessing a current retrieval demand description text carried by the retrieval request and a cached historical retrieval demand description text to obtain a plurality of retrieval keywords; semantic retrieval is conducted according to each retrieval keyword and a preset RAG vector knowledge base, document content meeting the retrieval request is obtained, and the preset RAG vector knowledge base is generated on the basis of a dynamically-updated gynecologic and obstetric specialist knowledge base; inputting the plurality of retrieval keywords and the document content into a generation model which is finely adjusted in advance to generate an answer text corresponding to the retrieval request; and outputting an answer text corresponding to the retrieval request, and sending the answer text to the client. Therefore, by the adoption of the method and device, the model can understand the semantic and logical relation of the context, the model can generate information conforming to the fact, and therefore the accuracy of system retrieval is improved.

To provide a system including: a robot with a generative AI mounted thereon; means for making a response with natural words to a question or a request from a user; and means for handling complicated problems or questions based on specialized knowledge.SOLUTION: In a data processing system, a data processing apparatus 12 having a specific processing unit is connected to a smart device 14 via a network. The data processing apparatus executes specific processing. The system includes a robot with a generative AI mounted thereon, the robot being configured to receive a question or a request from a user and make a response with natural words using the generative AI. Specifically, the robot uses a voice recognition system to convert the question or request from the user into a text, and inputs the text to the generative AI. The generative AI generates an appropriate response based on the input text, converts the response into voice using a voice synthesis system, and outputs the voice from the robot.SELECTED DRAWING: Figure 2

The present invention relates to the field of semantic understanding technology, and more specifically, to a method for fine-grained parsing of user comments based on a large language model. The method includes preprocessing user comments; matching the comments against specialized knowledge in a specific field to obtain target knowledge; training the large language model based on target tasks and dynamic tuning prompts, the target tasks including aspect word extraction and sentiment polarity classification; and fusing the comments with the target knowledge and inputting them into the large language model to obtain parsed results for the comments. The present invention utilizes a large language model and an optimized prompting strategy to perform fine-grained parsing of user comments.

Provided is a mild cognitive impairment (MCI) screening method with which it is possible to easily screen for MCI with high accuracy in a short time, without using a large, complicated, expensive device, and without requiring specialized knowledge. Also provided are a computer program and screening device that are associated with said method.　A mild cognitive impairment screening method according to the present invention includes: a measurement step S1 for measuring a finger tapping motion that is an opening / closing motion of two fingers; a calculation step S3 for calculating a prescribed feature quantity on the basis of measurement data obtained in the measurement step within a prescribed time period during finger tapping of only the right hand and / or during finger tapping of only the left hand and / or during finger tapping of both hands simultaneously and / or during finger tapping of both hands alternately; and a determination step S5 for performing a determination regarding mild cognitive impairment on the basis of the feature quantity calculated in the calculation step S3. For the determination step S5, a discriminative model is generated (step S4) by machine learning including artificial intelligence, and the generated discriminative model is used to perform the determination regarding mild cognitive impairment.

An object of a system according to an embodiment is to efficiently and accurately extract data without expert knowledge.SOLUTION: A system includes a data extraction condition input part, an SQL generation part, a mart definition cooperation part, and a customization part. The data extraction condition input unit allows a user to input a data extraction condition in a chat form. The SQL generation unit generates an optimal SQL query based on the extraction condition input by the data extraction condition input unit. The mart definition association unit associates the SQL query generated by the SQL generation unit with the mart definition to reflect the latest definition. The customization unit performs customization for each company.SELECTED DRAWING: Figure 1

The invention discloses a knowledge base dynamic construction and intelligent retrieval method based on a large model, and relates to the technical field of computer retrieval, a DeepSeek large model is taken as a teacher model, a lightweight student model is trained through a specialized knowledge distillation strategy, and local deployment is realized in combination with a containerization technology; the specialized knowledge distillation strategy comprises dynamic modal perception distillation, knowledge graph structure guide distillation and online incremental distillation; according to the method, the overall process efficiency is remarkably improved compared with a traditional scheme through multi-modal data automatic processing and lightweight model local deployment; cross-modal semantic alignment and joint reasoning are realized, and the accuracy of retrieval and answer generation is improved; a real-time updating mechanism and version management enable knowledge to always meet actual requirements, and the logic contradiction occurrence rate is remarkably reduced; the fact reliability of the generated content is greatly improved by a fact anchoring and double-stage verification mechanism; local deployment and containerization isolation design avoid leakage of sensitive information, and meet the requirements of data security compliance.

In order to address the problem that specifying information and giving advice on a kinesiological state of the body of a user require specialized knowledge and experience and the effort and cost required are relatively high, an information processing apparatus 100 includes: a sole information acquiring unit 141 that acquires sole information on a sole of a user; a determination information acquiring unit 149 that acquires determination information on a kinesiological state of the body of the user, using the sole information and learning information that is prepared in advance; an output information acquiring unit 151 that acquires output information based on the determination information; and an output unit 161 that outputs the output information. Accordingly, it is possible to easily specify information on a kinesiological state of the body of a user.

To provide a system including means for selecting the optimum combination on the basis of what is desired and means for a plurality of tasks, the system being applicable for all people including people without specialized knowledge of AI and allowing all people to realize their potential.SOLUTION: The system includes: means for selecting the optimum combination from a plurality of generation AIs on the basis of what is desired, which is input; and means for a plurality of tasks including generation of sentences, generation of images, generation of music, brain storming of a project idea.SELECTED DRAWING: Figure 1

The invention relates to a large-model-based future quantitative back-testing method and system, and the method comprises the steps: constructing a future quantitative knowledge base for the field of future based on the construction logic of a FinDKG dynamic knowledge graph; performing enhanced training on the open source large language model through a GRPO enhanced training algorithm by taking improvement of the generation capability of the futures quantification code of the large language model as a target; and under a context retrieval enhancement generation framework combined with ACE and RAG, according to the obtained user natural language and the futures quantification knowledge base, generating a futures quantification code through the trained open source large language. According to the method and the device, a complete and executable quantitative backtest code, a specialized knowledge base of futures, algorithm fine tuning and an ACE + RAG framework can be automatically generated by relying on a large language model and enabling a user with a non-professional programming background to describe strategy requirements through a natural language, so that the intelligence and the accuracy of model code generation are greatly improved.

To provide a program generation device which allows even users having neither expert knowledge nor an advanced skill level to generate operation programs of a robot.SOLUTION: A program generation device comprises: an acquisition unit configured to acquire a plurality of movement points corresponding to a position of a destination of a predetermined portion of a robot and a job corresponding to a plurality of movement methods of the predetermined portion; a display control unit configured to display movement point codes representing the plurality of movement points arranged in movement order, and also display icons representing the movement methods in association with the displayed movement point codes; and a generation unit configured to generate an operation program which operates the robot according to contents displayed by the display control unit.SELECTED DRAWING: Figure 5

To reduce the effort and cost of tax consultation through procedure proxy or answer service provided free of charge by an AI system for easily using consultation related to tax business even for a user without having expertise.SOLUTION: A system for converting consultation related to tax business into AI includes means for receiving tax consultation; means for providing suitable answer on the basis of the consultation content; and means for automating the proxy business of actual procedure or document preparation.SELECTED DRAWING: Figure 1

The system according to the embodiment aims to enable experts to promptly and appropriately consult on specialized knowledge. [Solution] A system according to an embodiment includes a reception unit, an analysis unit, and a provision unit. The reception unit receives consultation content from an expert. The analysis unit analyzes the consultation content received by the reception unit. The provision unit provides related expertise based on the information analyzed by the analysis unit.

To provide an AI training system for efficiently replacing human tasks.SOLUTION: An AI training system for taking charge of human tasks instead, includes means for training an AI having specialized knowledge of industry, and means for training a generative AI in customer support, language translation, media production, design, market analysis, accounting, programming, or the like.SELECTED DRAWING: Figure 1

We provide the system. [Solution] A means for a person with specialized knowledge to input that knowledge and speaking style as information, A method for training an AI model using natural language processing technology based on that input information, A means of storing and making available trained AI models on the platform, A means for users to search for AI models through the platform and begin interacting with them, A means by which an AI model generates a response to that AC input and sends it back to the user, A means of providing responses as an interactive learning experience and enhancing immersion through visual and auditory means, A system that includes this.

To easily set up network devices without specialized knowledge by using the system according to the embodiment.SOLUTION: A system according to the embodiment has a reception unit, a generation unit, and a provision unit. The reception unit accepts input from the user. The generation unit analyzes the information received by the receiving unit and generates the settings of the network device. The provision unit provides the settings generated by the generation unit to the user.SELECTED DRAWING: Figure 1

Personalizing the services offered enables the delivery of services that are tailored to the user's needs. [Solution] When a user is engaging in a conversation with a dialogue model for general conversation (S235-S246, NO in S247, NO in S249), the system switches to the task processing model to engage in conversation depending on the content of the conversation (NO in S247, S248), enabling the use of the task processing model's specialized knowledge in a particular field. Furthermore, through conversations with individual users by the dialogue model, the dialogue model is personalized to suit the individual user (YES in S245, S246).

The system according to the embodiment aims to generate database queries efficiently and accurately and extract data without requiring specialized knowledge. [Solution] A system according to an embodiment includes a reception unit, a generation unit, an execution unit, a linking unit, and a customization unit. The reception unit receives data extraction conditions from a user. The generation unit analyzes the conditions received by the reception unit and generates an efficient SQL query. The execution unit executes the SQL query generated by the generation unit and extracts data. The linking unit updates the data extracted by the execution unit based on the latest data definition. The customization unit supports customization for each company.