Chinese Domain Named Entity Recognition Based on Ontology

摘要：

Named Entity Recognition (NER) is one of kernel tasks in many Natural Language Processing (NLP) applications, which has recently become the hot spot of researches. Named entity is classified into General Named Entity (GNE) and Domain Named Entity (DNE) in this paper. We put forward a method of Chinese Domain Named Entity Recognition (DNER) based on ontology, which combines Conditional Random Field (CRF) with the rule templates of POS acquired through domain ontology already constructed. Results of experiments indicate that F-measure has reached 92.36％. Experimental data also show that domain ontology can have great effects on recognizing the boundaries of most DNEs and DNEs in special forms.

关键词： DNE DNER Domain Ontology POS rule-template CRF

作者: SHI Shu-min FENG Chong HUANG He-yan LIU Dong-sheng

作者单位: School of Computer Science and Technology,Nanjing University of Science & Technology,Nanjing,210094; Research Center of Computer & Language Information Engineering,CAS,Beijing,100097 Computer & Information Engineering College,Inner Mongolia Normal University,Huhhot,010022

会议类型: 国际会议

会议名称: 2008高等智能国际会议(2008 International Conference on Advanced Intelligence)

会议地点: 北京

会议语种:英文

在线出版日期: 2008-10-18（万方平台首次上网日期，不代表论文的发表时间）