基于领域本体半结构化文本知识自动获取方法设计和实现.pdf

28 12 Vol. 28 No. 12 2005 12 CHINESE JOURNA L OF COMPU TERS Dec. 2005 1) , 2) 1) ) 王海涛 曹存根 高 颖 1) ( 100080) 2) ( 1000 9) ) ( 100080) . , . , , . : OMKast; ePKPL, OMKastePKPL VM( ePKPL ) . , , . ; ; ; ; TP 18 Design and Implementation of a System for Ontology-Mediated Knowledge Acquisition from Sem-i Structured Text 1) , 2) 1) ) WAN G Hai-Tao CAO Cun-Gen GAO Ying 1) ( K ey L abor ato ry of I nte ll ige nt I nf o rmat io n Pr oce ss i ng , I ns ti tu te of Co mp u ti ng T echn ol og y , Chine s e A c ade my of S ci ence s, Bei i ng 100080) 2) ( Grad uat e S cho ol of t he Chi nes e A c ade my of S ci ence s, Bei i ng 1000 9) ) ( I ns ti t ut e of Sof tw ar e, Chi nes e A cad emy of Sc ienc es, B ei ing 100080) Abstract In the paper, a new ontology-mediated method for automatic know ledge acquisition from text is presented. And based on the method, a system has been implemented and tested in several applications. With domain-specific ontologies, know ledge acquisition from sem-i structure- d text is automated completely. Not only it makes human beings releasing from tedious manual w ork, but also the efficiency of acquisition is improved remarkably. A dditionally, it is a founda- tion for know ledge acquisition from