查询结果:   杜泽宇,杨燕,贺樑.基于中文知识图谱的电商领域问答系统[J].计算机应用与软件,2017,34(5):153 - 159.
中文标题
基于中文知识图谱的电商领域问答系统
发表栏目
人工智能与识别
摘要点击数
759
英文标题
QUESTION ANSWERING SYSTEM OF ELECTRIC BUSINESS FIELD BASED ON CHINESE KNOWLEDGE MAP
作 者
杜泽宇 杨燕 贺樑 Du Zeyu Yang Yan He Liang
作者单位
华东师范大学信息科学技术学院 上海 200062     
英文单位
School of Information Science Technology, East China Normal University,Shanghai 200062,China     
关键词
自动问答 知识图谱 语义网 本体
Keywords
Question answering Knowledge map Semantic Web Ontolog
基金项目
国家高技术研究发展计划项目(2015BAH12F01-04);上海市科委重点项目(14511107000)
作者资料
杜泽宇,硕士生,主研领域:自然语言处理。杨燕,博士生。贺樑,教授。 。
文章摘要
随着知识图谱的迅速发展,面向知识图谱的中文领域问答系统已成为目前最新最热的研究方向之一,对于提高专业领域服务智能化程度具有较高的意义和价值。针对中文口语语义表达多样化、不符合语法规范以及电商领域特殊性问题,提出一套流式的中文知识图谱自动问答系统CEQA,能够较好地完成电商领域商品咨询以及统计推理等复杂问题,特别是有效地提升了中英文混合商品名称识别、语义链接以及复杂问句的依存分析等方面的性能。实验结果表明,该系统在电商领域问答应用中具有较高的准确率和实用价值。
Abstract
With the rapid development of knowledge map, the Chinese domain question answering system for knowledge map has become one of the newest and hotest research directions at present, and it is of great significance and value to improve the intelligence level of professional field. In this paper, a set of streaming Chinese knowledge map automatic question answering system (CEQA) is proposed for the diversification of Chinese spoken language semantic expression, grammatical specification and the particularity of electricity business domain. It can accomplish the complex problem of commodity consultation and statistical reasoning in the field of electric business, especially the improvement of the interdependence between Chinese and English mixed commodity name recognition, semantic link and complex question. The experimental results show that the system has high accuracy and practical value in the application of question and answer.
下载PDF全文