RELATION EXTRACTION FROM WIKIPEDIA ARTICLES BY ENTITIES CLUSTERING

摘要：

　　Wikipedia is an encyclopedia based on wiki technology.It is multilingual high quality knowledge base.In this work a episode based extraction method are proposed to extract relations from Wikipedia articles.The entities are clustered and labeled.The relation extraction is benefited by the information redundancy provided by the clusters.A strict Wikipedia entities clustering algorithm based on the category system and first sentence of the article is approached.This work required less manual assist.And the relations are abundant.The results are comparable with other works 1,2.

关键词： Relation extraction Entities clustering Episode

作者: Song Liu Fuji Ren

作者单位: School of Computer Science,Beijing University of Posts and Telecommunications,Beijing 100876,China;F Faculty of Engineering,The University of Tokushima,Tokushima 770-8506,Japan

会议类型: 国际会议

会议名称: 2012 2nd IEEE International Conference on Cloud Computing and Intelligence Systems (2012年第2届IEEE云计算与智能系统国际会议(IEEE CCIS2012))

会议地点: 杭州

会议语种:英文

页码: 1973-1977

在线出版日期: 2012-10-30（万方平台首次上网日期，不代表论文的发表时间）