一种基于知识图谱的文档格式知识表示方法及装置

CN122114115BActive Publication Date: 2026-07-17WUHAN UNIV

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
WUHAN UNIV
Filing Date
2026-04-22
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing technologies cannot effectively combine semantic generation and format control when generating official documents, resulting in chaotic document layout, insufficient template flexibility, fragmented format rules, and difficulty in adapting to policy changes.

Method used

A knowledge graph-based approach is adopted to transform format specifications into structured graph representations, construct a format knowledge graph containing structural, semantic, and functional layers, and introduce an adaptive feedback mechanism to identify format entities and their relationships through a deep learning model, thereby achieving dynamic optimization of format knowledge.

Benefits of technology

It achieves precise calling and adaptive capabilities of format rules in the official document generation system, improves the format compliance and adaptability of generated documents, and can dynamically optimize based on feedback from downstream generation results.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122114115B_ABST
    Figure CN122114115B_ABST
Patent Text Reader

Abstract

本发明公开了一种基于知识图谱的文档格式知识表示方法及装置,旨在解决现有智能文档生成系统中,格式规则难以被机器利用、语义生成与格式控制相分离的问题,基于知识图谱的文档格式知识表示方法包括:构建多源格式语料库并进行结构化预处理;利用深度学习模型抽取格式实体及其关系;基于抽取结果构建包含结构层、语义层和功能层的格式知识图谱;并引入基于下游任务反馈的自适应调节机制,对图谱进行动态优化。本发明将非结构化的格式规范转化为结构化图表示,使智能体能够精准调用格式规则,并具备自学习进化能力,显著提升了公文生成的格式合规性与适应性。
Need to check novelty before this filing date? Find Prior Art