Knowledge base technical method and system based on RAG retrieval enhancement

By parsing the source documents into structured text fragments and constructing concept association graphs, combined with vector retrieval and graph retrieval, the shortcomings of existing knowledge base systems in intent understanding and data fusion are addressed, and high-precision and reliable knowledge base retrieval and answer generation are achieved.

CN120687593AActive Publication Date: 2025-09-23深圳市华磊迅拓科技有限公司
9 Cites 6 Cited by

Patent Information

Application Number
CN202510835801.8
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2025-06-20
Publication Date
2025-09-23
Estimated Expiration
2045-06-20

AI Technical Summary

Technical Problem

Existing knowledge base retrieval systems have difficulty in deeply understanding query intent, cannot effectively integrate external real-time data and user historical information, and cannot locally deploy cloud services that meet data privacy requirements, resulting in a lack of timeliness and consistency in retrieval results.

Method used

Parse the source document into structured text fragments, generate semantic vectors and store them in a vector database, build a concept association graph, combine vector retrieval with graph retrieval, build a set of candidate knowledge units, and dynamically verify the answers through the generation model.

Benefits of technology

It achieves deep semantic capture of query intent, improves retrieval accuracy and answer credibility, has dynamic time series perception capabilities, and ensures historical support and multi-source verification of generated answers.

✦ Generated by Eureka AI based on patent content.
Patent Text Reader

Abstract

The invention discloses a knowledge base technical method and system based on RAG retrieval enhancement, and relates to the technical field of knowledge bases. The method comprises the steps that a source document is analyzed into structured text fragments, and semantic vectors are generated; splitting the text segments into minimum knowledge units, extracting concepts and behavior trigger words to construct a concept association graph, and storing session abstracts in a fixed-length annular structure; the method comprises the following steps: receiving user query, generating a query vector, retrieving a most relevant text fragment, constructing and de-duplicating candidate knowledge units by combining hierarchical diffusion of a concept association graph and an annular abstract matching result, deeply splicing original text paragraphs according to a graph path, generating a dynamic prompt box, calling a generative model to generate a preliminary answer, and executing verification. And the verification state is marked on the final answer. By constructing a concept association map, the activeness weight and the time sequence fingerprint of a map edge are updated in real time, and accurate capture of deep semantics and logical relationships of query intentions is realized.
Need to check novelty before this filing date? Find Prior Art