摘要
知识库问答是当今自然语言处理的热门研究方向,它允许用户输入自然语言问句,问答系统分析问句、查询知识库并智能返回给用户答案,无须用户进一步查询搜索。开放域问答更加拓宽了用户查询的知识领域范围。如何准确处理用户输入的自然语言问句并在知识库中进行推理是知识库问答的难题之一。文章主要研究并讨论了知识库问答的命名实体识别和关系抽取,这些任务主要应用了深度学习技术。
Knowledge base question-answer is a popular research direction in natural language processing.It allows users to input natural language questions.Question answering system can analyze questions,query knowledge base and return answers to users in⁃telligently without further query and search.Open domain question-answer widens the scope of user query knowledge.How to deal with the natural language questions input by users accurately and reasoning in the knowledge base is one of the difficult problems in the knowledge base question answering.The paper mainly studies and discusses named entity recognition and relation extraction of knowledge base question-answer.These tasks mainly apply deep learning technology.
作者
李东奇
李明鑫
张潇
LI Dong-qi;LI Ming-xin;ZHANG Xiao(China University of Mining&Technology,Beijing 100083,China)
出处
《电脑知识与技术》
2020年第36期179-181,共3页
Computer Knowledge and Technology
基金
中国矿业大学(北京)大学生创新训练项目资助。
关键词
知识库问答
命名实体识别
关系抽取
深度学习
自然语言处理
knowledge base question answer
named entity recognition
relation extraction
deep learning
natural language process⁃ing
