【作者】 周浪； 史树敏； 冯冲； 黄河燕；

【Author】 ZHOU Lang~(1,2),SHI Shu-Min~2,FENG Chong~2,HUANG He-Yan~(2,3) 1.School of Computer Science and Technology,University,Nanjing University of Science and Technology,Nanjing 210094; 2.Research Center of Computer & Language Information Engineering,CAS,Beijing 100097 3.School of Computer Science & Technology,Beijing Institute of Technology,Beijing 100081

【机构】 南京理工大学计算机科学与技术学院； 中国科学院计算机语言信息工程研究中心； 北京理工大学计算机学院；

【摘要】 大部分的术语抽取工作都将重点放在词组型术语的识别上,忽略了单词型术语。虽然在整个术语系统中,单词型术语的数量要比词组型术语少得多,但它却是构成词组型术语的重要元素。由于单词型术语具有语法边界清晰的特点,引入模糊C-均值聚类算法,将术语识别工作转化为两类聚类任务,从而实现无监督自动标注的目的,并获得了令人满意的结果。更多还原

【Abstract】 Since multi-word terms occur much more frequently than the single-word terms,most term extraction systems focus on the former.But many foundational terms in specific domain are in the form of single word,and they’re the important components in the multi-word terms.In the single-word extraction process,fuzzy C-means clustering algorithm is introduced to transform the extraction task into clustering task.Without manual assistant and additional resource,this approach could get a satisfying result.更多还原