7 results about "Implicit knowledge" patented technology

This invention relates to artificial intelligence and big data processing technologies, aiming to provide a method for constructing multimodal intelligent knowledge cards and self-organizing links by topic. The method includes: collecting data from multiple sources, preprocessing it to extract multimodal information, and constructing several standard structured data objects as knowledge cards; encapsulating the data in a unified format to build a knowledge card database; constructing multidimensional hybrid embedding vectors based on the attributes of the data in the knowledge card database using different knowledge cards; establishing a multimodal semantic adaptive topology network under spatiotemporal constraints to achieve self-organizing links by topic; storing the knowledge cards as nodes in a backend knowledge graph and establishing connections with the frontend display / interactive interface. This invention can be used for intelligent processing of data extraction, structured encapsulation, and dynamic association. Through vector space alignment, it can discover implicit knowledge and potential connections, achieving dynamic evolution and adaptability.

This invention relates to the field of intelligent customer service, specifically disclosing a method and system for constructing a customer service knowledge base based on a knowledge graph. The method includes: identifying implicit needs by analyzing the spatiotemporal correlation between user actions and customer service dialogues, generating enhanced knowledge units that integrate explicit and implicit knowledge; constructing a knowledge graph, achieving adaptive optimization of the knowledge graph through real-time monitoring of relationship strength changes, forming a dynamic knowledge graph; capturing the user's current contextual information, matching it with the dynamic knowledge graph to generate contextualized knowledge subgraphs, and using multi-path reasoning to generate personalized response content; the system includes modules for implementing the above steps. This invention effectively solves the shortcomings of traditional customer service knowledge bases in implicit knowledge mining, dynamic updating, and deep contextual awareness, achieving improvements in knowledge acquisition from partial to comprehensive, knowledge base from static to dynamic, and responses from general to personalized, significantly enhancing the cognitive depth and service quality of intelligent customer service systems.

This application provides a method, apparatus, device, and computer storage medium for constructing a tag tree. The tag tree construction method includes: obtaining a tag tree construction request, the construction request including an object to be analyzed; parsing the construction request to determine key request information, the key request information including at least one of the following: the modal type of the object to be analyzed, and the field type corresponding to the construction request; determining associated tag knowledge corresponding to the construction request based on the key request information; and processing the construction request and associated tag knowledge using a tag tree generation model to generate a target tag tree corresponding to the construction request. In this embodiment, the associated tag knowledge is determined based on the key request information. Since the key tag knowledge can fully understand and operate on the professional terms and implicit knowledge corresponding to the object to be analyzed, the quality and efficiency of tag tree construction are effectively guaranteed when generating the target tag tree based on the associated tag knowledge.

This invention provides a visual system for interactive SQL generation and modification that integrates implicit knowledge, belonging to the field of interactive database query technology. By constructing a visual system including an implicit knowledge base building module and an interactive generation and modification module, implicit knowledge items are automatically extracted and categorized from historical SQL scripts to build an implicit knowledge base, enhancing the code generation capabilities of LLM (Low-Level Management). Simultaneously, a multi-view interactive interface is provided, mapping the generated SQL code back to implicit knowledge items in natural language form, supporting user review and correction at the knowledge level rather than just the code level. This invention enhances the accuracy of SQL generation by reusing historical implicit knowledge and achieves knowledge-level understanding and correction through multi-view interaction. It solves the problems of low accuracy, poor interpretability, and difficulty in correction of existing NL-to-SQL tools when handling implicit knowledge, significantly improving the efficiency and accuracy of SQL writing and making it more practical.

This invention relates to the field of educational informatization technology, and particularly to curriculum standard text processing and knowledge graph construction technology. In existing technologies, converting curriculum standards into knowledge graphs mainly relies on manual compilation, which suffers from high costs, poor consistency, and long cycles. Furthermore, existing templates and general tools have shortcomings such as low recall rates, inability to capture implicit knowledge, and lack of educationally specific semantics. This invention provides an intelligent knowledge graph construction system based on curriculum standard text, characterized by employing the LangChain architecture and Neo4j graph database. Through an end-to-end system driven by a hierarchical graph neural network enhanced with curriculum semantics, it combines modules for data acquisition, parsing and preprocessing, knowledge entity and relation extraction, data integration and deduplication, knowledge graph storage, visualization, and retrieval. This achieves automatic conversion of curriculum standard text into a knowledge graph without manual intervention, covering all educational stages and knowledge nodes, with a node recall rate of no less than 95%. It uses Chinese semantic tags to characterize multi-dimensional hierarchical relationships and supports active learning iteration, lightweight deployment, and minute-level incremental updates. Compared with existing technologies, this invention reduces the construction time of a single-discipline knowledge graph from several weeks to less than one hour, achieves a node coverage rate of over 90%, and reduces the workload of manual verification by 90%. It effectively solves many problems of traditional methods, can be interconnected with other educational platforms, and meets the needs of low-cost, scalable, and high-precision educational informatization. It has significant technical effects and industrial application value.

The application provides a strip product process parameter step-by-step recommendation method and system based on representation learning and a storage medium, and relates to the technical field of information. The method comprises the following steps: obtaining strip product production whole-process data and constructing a structured representation through a knowledge graph; embedding modeling entities and relationships in the knowledge graph based on a RotatE representation learning algorithm of complex space relationship rotation to generate vector representation of complex relationships; generating multi-process process parameter recommendation based on a step-by-step recommendation strategy of time sequence dependence, and realizing cross-process parameter constraint through dynamic updating of the knowledge graph. The application solves the complex relationship modeling problem based on the RotatE representation learning algorithm of complex space relationship rotation; improves the implicit knowledge capturing capability through the construction of a modular semantic aggregation knowledge representation model; in addition, a step-by-step recommendation strategy is proposed to process the time sequence dependence through a "link prediction-graph completion-link prediction" iterative process.

The application provides a test scene safety question and answer system and a construction method. The system comprises a document analysis and preprocessing module, a semantic perception text segmentation module, a test field optimized vector embedding module, a vector storage and retrieval module, a large language model reasoning module and a natural language interaction interface module. The method comprises the following steps: S1, document analysis and preprocessing; S2, intelligent block of the preprocessed text; S3, conversion of the text block into a vector and saving into a local vector library; S4, deployment of an open source large language model and configuration of model loading parameters and reasoning environment; S5, construction of a retrieval question and answer chain; and S6, user interaction and feedback collection. The method breaks the keyword retrieval limitation through an embedding model, improves the retrieval efficiency, makes the test implicit knowledge explicit and reusable through intelligent block, reduces the knowledge transmission cost, realizes the whole link intranet localization deployment, prevents sensitive information leakage and meets the safety and compliance requirements.