BUILDING KNOWLEDGE BASE FROM SEMI-STRUCTURED DATA

摘要：

Semantic pattern is kind of template with semantic information, and it can be used to structurize a sentence.In this paper, semantic pattern based question and its corresponding answers are used as the data resource of knowledge base.The exact answer is extracted from plain text answer by matching of question and answer.Two kinds of matching methods and the experiment results are simple presented.Extracted exact answer and semantic pattern based question are composed of a structured information unit which can be inserted into knowledge base with the assistance of ontology.The disease ontology is studied for illustration.Our approach to knowledge base construction is concise and effective.

关键词： Knowledge extraction Disease ontology Semantic pattern

作者: XIAO-LI LIU GUO-QING WU MIN YANG

作者单位: School of Computer, Wuhan University, China;Department of Computer Science, City University of Hong School of Computer, Wuhan University, China ZTE Corporation,Shenzhen,China

会议类型: 国际会议

会议名称: 2007 International Conference on Machine Learning and Cybernetics(IEEE第六届机器学习与控制论国际会议)

会议地点: 香港

会议语种:英文

页码: 839-843

在线出版日期: 2007-08-19（万方平台首次上网日期，不代表论文的发表时间）