一种基于知识图谱的智能法律咨询问答方法

By constructing a super-relational legal knowledge graph and adopting the MAYPL representation learning method, the problems of insufficient expression and inadequate reasoning in complex legal question-and-answer systems in existing legal consultation systems are solved, achieving more accurate legal knowledge modeling and question-and-answer processing, which is applicable to reasoning and interpretation of results for multi-condition legal facts.

CN122154951BActive Publication Date: 2026-07-17NANJING UNIV OF INFORMATION SCI & TECH
View PDF 1 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
NANJING UNIV OF INFORMATION SCI & TECH
Filing Date
2026-05-09
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing legal consultation systems struggle to fully express the multi-layered semantic dependencies of legal knowledge when dealing with complex legal question-and-answer scenarios. They are unable to effectively handle complex legal issues with multiple constraints and are prone to structural information loss during the legal reasoning process, leading to inaccurate and inconsistent question-and-answer results.

Method used

We employ a knowledge graph representation learning method based on MAYPL to construct a super-relational legal knowledge graph. Legal knowledge is represented as core legal relations and multiple limiting relation-entity pairs. Through co-occurrence relations, connection relations, and position types, we perform structural representation learning to generate vector representations of legal entities, relations, and facts, and then perform graph querying and candidate answer reasoning.

Benefits of technology

It enables a more precise expression of the limiting conditions and positional relationships in legal facts, improves the accuracy of legal knowledge modeling and question-and-answer reasoning capabilities, is applicable to complex legal consultation scenarios, has a clearer legal basis chain and interpretability, and can adapt to the expansion and updating of legal knowledge.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122154951B_ABST
    Figure CN122154951B_ABST
Patent Text Reader

Abstract

本发明公开了一种基于知识图谱的智能法律咨询问答方法，包括如下步骤：步骤1：构建超关系法律知识图谱；步骤2：基于MAYPL对步骤1构建的超关系法律知识图谱进行结构表示学习；步骤3：接收用户的咨询问题，从咨询问题中抽取关键法律实体、法律关系、咨询意图和限定条件，并映射为图谱查询结构；步骤4：根据图谱查询结构，在超关系法律知识图谱中召回候选答案；步骤5：根据最佳候选结果生成面向用户的法律咨询答案；步骤6：获取问答反馈结果，对超关系法律知识图谱进行更新。本发明实现面向复杂法律咨询场景的智能问答处理，能够更有效地表达法律事实中的限定条件和位置关系，提高法律知识建模精度、问答推理能力及结果可解释性。
Need to check novelty before this filing date? Find Prior Art

Citation Information

Patent Citations

  • Consultation method and system based on natural language processing and legal knowledge graph

    CN121117239A