面向GIS基于专有名词优先的中文分词方法.pdfVIP

第 30卷第 7期 计算机应用 Vo.l 30 No. 7 2010年 7月 Journal of ComputerApp lications July 2010 : 1001- 9081( 2010) 07- 1941- 03 GIS 1 1 2 罗 浩 , 魏祖宽, 金在弘 ( 1. 610054; 2. , 370701) ( lhao77@ gma i.l com) :提出了一种面向地理信息系统领域的基于专有名词优 的中文分词方法: 利用专业词典通用词典和同 义词词典相结合的词典机制, 优 切分专有名词, 对粗分结果利用 Trigram 模型进行消歧而获取最终结果实验证 明, 该分词算法对专业文献的分词处理具有较好速度和准确性 : 中文分词; 专业词典; T rigram模型; 同义词词典 : T 391 : A Ch in ese word segm en tation for G IS based on priority special nam e 1 1 2 LUO H ao, WEI Zukuan , K IM Jaehong ( 1. School of Comp uter Sc ience and E ng ineering, University of E lectron icS cience and Technology of China, Chengdu S ichuan 610054, Ch ina; 2. Dep artment of Comp uter and Inf orm ation Engineering, YoungD ong Un iv ersity, YongdongGun Chung bu 370701, K orea) A bstract: A Ch inese word segm entation algorithm for Geograph ic Inform ation System (G IS) based on priority special name w as designed: use dictionarym echan ism w hich com bines synonym s d ictionary, general dictionary and special d ictionary, cu t the sentences by special nam e firstly, and get the segm entation result of d isamb iguating w ith Trigram mode lastly. The expermi ental resu lts show that the segm entation algorithm has good speed and accuracy in segm entation processing of professional literature. K ey w ords: Chinesew ord segm entation; special d ictionary; T rigram m ode; synonym s dictionary , 0 引言 , , , , : (Geographic Inform ation System, GIS),