关键词: 知识图谱, 问答方法, 语义解析, 深度学习, 大语言模型

Abstract: Significant advancements in Knowledge Graph Question Answering (KGQA) technology has been driven with the rapid evolution of artificial intelligence technologies and deepening research and application of Knowledge Graphs (KG). Owing to lacking a systematic classification framework for existing question-answering methods, current users have limited awareness of various KGQA models. Therefore, it is difficult to meet the needs of users across different domains to understand KGQA. To address this issue, this paper provided a systematic review of research progress in KGQA field over past 15 years: Firstly, four core question-answering strategies were identified including template matching, deep learning and large language model augmentation. Secondly, eight typical question-answering methods under four categories of strategies were compared in complex questions, multi-hop reasoning, low-resource environments, and multilingual settings. Meanwhile, applicable scenarios had been provided for these classic question-answering methods. Then, common datasets and performance evaluation methods were organized. Finally, building upon the current state of domain development, this study offers targeted suggestions and prospects for future research directions in KGQA technology, aiming to provide valuable references for subsequent related studies and applications.

Key words: Knowledge Graph (KG), question answering method, semantic parsing, Deep Learning (DL), Large Language Model (LLM)