摘要
信息抽取任务旨在从非结构化的文本中抽取出结构化的信息,帮助将海量信息进行自动分类、提取和重构,提高信息的利用率.目前,基于深度神经网络的信息抽取技术已经成为自然语言处理领域最重要的研究主题之一,它提供了分析非结构化文本的有效手段,是实现大数据资源化、知识化和普适化的核心技术,此外进一步为更高层面的应用和任务提供了支撑.文章对基于深度神经网络的信息抽取相关研究进行了综述,首先,简要概述了信息抽取的任务定义、目标和意义,然后,回顾了信息抽取任务的发展历程,接下来,从实体抽取、实体关系抽取、事件抽取和事件关系抽取4个方面梳理了近几年关键技术的研究进展.最后,文章对信息抽取领域的未来发展趋势进行了分析和展望.
Information extraction aims at extracting structured information from unstructured text,realizing automatically classify,extracting and reconstructing massive information,and enhancing the use of information.Recently,information extraction technology based on deep neural network is one of the most significant research topics in the field of natural language processing.It creates an effective way of analyzing unstructured text,and facilitates the realize the resource,knowledge and universality of big data.In addition,it further provides support for higher-level applications and tasks.In this paper,the related research on information extraction based on deep neural network has been reviewed.First,the task definition,goals,and meanings of information extraction has briefly been described,followed by an analysis of the development of the task.And then,the development of key technologies in recent years been summarized from four aspects:entity extraction,entity relation extraction,event extraction,and event relation extraction.Finally,the future development trends in the field of information extraction have been analyzed and looked forward to.
作者
代建华
彭若瑶
许路
蒋超
曾道建
李扬定
DAI Jianhua;PENG Ruoyao;XU Lu;JIANG Chao;ZENG Daojian;LI Yangding(Research Institute of Languages and Cultures/ Hunan Provincial Key Laboratory of Intelligent Computing and Language Information Processing, Hunan Normal University, Changsha 410081, China)
出处
《西南师范大学学报（自然科学版）》
CAS
2022年第4期1-11,共11页
Journal of Southwest China Normal University(Natural Science Edition)
基金
国家自然科学基金项目(61602059)
湖南省自然科学基金项目(2020JJ4624)
国家社会科学基金项目(20&ZD047)
湖南省教育厅科研基金项目(19A020)
湖南师范大学语言与文化研究院青年培育项目(2020QNP05).
关键词
信息抽取
深度神经网络
实体抽取
实体关系抽取
事件抽取
事件关系抽取
information extraction
deep neural network
entity extraction
entity relation extraction
event extraction
event relation extraction
作者简介
代建华,教授,主要从事人工智能、数据挖掘、机器学习、粗糙集理论、智能信息处理等方面的研究.
