北京信息科技大学学报（自然科学版）2023，Vol.38Issue(6)：70-79,87,11.DOI:10.16508/j.cnki.11-5866/n.2023.06.010
基于深度学习的实体关系抽取研究综述
Review of research on entity relation extraction based on deep learning
摘要
Abstract
As one of the primary tasks in natural language processing,entity relation extraction plays a crucial role in knowledge graphs,intelligent question answering and other fields.In order to provide a clear overview of the techniques used for entity relation extraction,the research methods of entity relation extraction were summarized by taking deep learning technology as a starting point.Initially,the definition and modeling methods of entity relation extraction tasks were briefly described.Next,various entity relation extraction methods were classified and analyzed from the perspective of deep learning technology,in order to sort out their characteristics.Then,the index parameters of major open-source dataset models were counted to clearly compare and summarize various methods performance.Finally,the different presentation forms of entity relation extraction tasks were described in order to analyze and prospect the future development direction.关键词
自然语言处理/知识图谱/实体关系抽取/深度学习Key words
natural language processing/knowledge graph/entity relation extraction/deep learning分类
信息技术与安全科学引用本文复制引用
任乐,张仰森,刘帅康..基于深度学习的实体关系抽取研究综述[J].北京信息科技大学学报（自然科学版）,2023,38(6):70-79,87,11.基金项目
国家自然科学基金项目(62176023) （62176023）
