《计算机应用》唯一官方网站 ›› 2026, Vol. 46 ›› Issue (8): 2394-2410.DOI: 10.11772/j.issn.1001-9081.2025070838

• 人工智能 • 上一篇    下一篇

基于知识图谱的问答方法综述

刘新亮, 徐雨时, 李杜白, 任延昭()   

  1. 北京工商大学 计算机与人工智能学院,北京 100048
  • 收稿日期:2025-07-25 修回日期:2025-11-04 接受日期:2025-11-04 发布日期:2025-12-22 出版日期:2026-08-10
  • 通讯作者: 任延昭
  • 作者简介:刘新亮(1972—),男,湖北黄冈人,教授,博士,CCF会员,主要研究方向:食品领域知识图谱、智慧农业、农产品质量安全追溯
    徐雨时(2000—),男,北京人,硕士研究生,主要研究方向:食品领域知识图谱、自然语言处理
    李杜白(2000—),女,北京人,硕士研究生,主要研究方向:自然语言处理
    任延昭(1986—),男,山东聊城人,讲师,博士,主要研究方向:数据处理及可视化。
  • 基金资助:
    北京市科技计划课题(Z221100007122003)

Survey of knowledge graph-based question answering methods

Xinliang LIU, Yushi XU, Dubai LI, Yanzhao REN()   

  1. School of Computer and Artificial Intelligence,Beijing Technology and Business University,Beijing 100048,China
  • Received:2025-07-25 Revised:2025-11-04 Accepted:2025-11-04 Online:2025-12-22 Published:2026-08-10
  • Contact: Yanzhao REN
  • About author:LIU Xinliang, born in 1972, Ph. D., professor. His research interests include knowledge graph in food field, smart agriculture, traceability for agricultural product quality and safety.
    XU Yushi, born in 2000, M. S. candidate. His research interests include knowledge graph in food field, natural language processing.
    LI Dubai, born in 2000, M. S. candidate. Her research interests include natural language processing.
  • Supported by:
    Beijing Science and Technology Program(Z221100007122003)

摘要:

随着人工智能技术的快速演进,知识图谱(KG)的研究与应用不断深化,推动基于KG的问答(KGQA)技术取得了显著进展。由于缺乏对现有问答方法的系统性划分框架,用户对各类KGQA模型的认知较为有限,因此,难以满足不同领域用户的参考需求。针对这一问题,文中对近 15 年KGQA领域的研究进展进行系统性综述。首先,归纳提炼出包括模板匹配、语义解析、深度学习、大语言模型(LLM)增强在内的4类核心问答策略;其次,对比4类策略下8种典型问答方法在复杂问题、多跳推理、低资源环境和多语言环境下的表现,并给出各个方法的适用场景;再次,归纳整理常用数据集和性能评测方法;最后,结合领域发展现状,对KGQA技术的未来研究方向提出针对性建议与展望,为后续相关研究与应用提供参考。

关键词: 知识图谱, 问答系统, 语义解析, 深度学习, 大语言模型

Abstract:

Significant advancements in Knowledge Graph Question Answering (KGQA) technology have been driven with the rapid evolution of artificial intelligence technologies and deepening research and application of Knowledge Graph (KG). Owing to lacking systematic classification framework for the existing question-answering methods, current users have limited awareness of various KGQA models. Therefore, it is difficult to meet the reference needs of users across different domains. To address this issue, a systematic review of research progress in KGQA domain over past 15 years was conducted. First, four core question-answering strategies were identified including template matching, semantic parsing, deep learning and Large Language Model (LLM) augmentation. Second, the performance of eight typical question-answering methods under four categories of strategies under complex question, multi-hop reasoning, low-resource environments, and multi-lingual environments was compared, and applicable scenarios for these methods were provided. Third, common datasets and performance evaluation methods were organized. Finally, combining the current state of domain development, targeted suggestions and prospects for future research directions in KGQA technology were proposed, aiming to provide references for subsequent related studies and applications.

Key words: Knowledge Graph (KG), Question Answering (QA) system, semantic parsing, deep learning, Large Language Model (LLM)

中图分类号: