基于类型论的汉语分词系统TTCS  

Chinese Segmentation System TTCS Based on Type Theory

作　　者：高东平[1] 

机构地区：[1]中国医学科学院医学信息研究所,北京100020

出　　处：《重庆理工大学学报（社会科学）》2011年第8期61-66,共6页Journal of Chongqing University of Technology（Social Science）

基　　金：国家社科基金"面向自然语言信息处理的范畴类型逻辑研究"(09BZX046)资助

摘　　要：针对汉语分词中的未登录词识别和歧义消除这两个瓶颈问题和目前方法的不足,将类型论的思想引入到汉语分词领域中,提出了类型匹配模型,以类型论函数贴合为主要算法,融合了全切分、统计,前后缀识别信息等多种技术手段,将分词、未登录词识别、歧义消除,词典的动态扩充有机的融合到一个统一的理论模型当中。在此理论框架的基础上,最后实现了基于类型论的汉语分词系统TTCS。Chinese unknown word recognition and disambiguation are difficult problems of Chinese word segmentation.We introduce a method based on type theory for Chinese word segmentation in this paper.The model of type functional application is employed attempting to resolve Chinese ambiguity and unknown word recognition.A Chinese word segmentation system TTCS is proposed that it unites techniques including omni-segmentation,statistical method,prefix and suffix information expansion and so on.A major advantage of the TTCS is that it can deal with Chinese word segmentation,unknown word recognition,disambiguation and the dynamic expansion of the dictionary in a unified theoretical model.

关 键 词：类型匹配 汉语分词 歧义消除 未登录词 类型论 

分 类 号：B81[哲学宗教—逻辑学]