MIS智能接口中汉语分词系统的设计与应用

The Design and Application of Chinese Word Segmentation System in a MIS Intelligent Interface

提供汉语检索接口是MIS应用的一大趋势,其主要困难在于如何让计算机理解汉语检索用语,为此本文构建了MIS智能检索接口中的汉语分词系统,并提出了分词策略。对汉语切分中的歧义问题进行了深入的探讨,应用互信息和t-信息差完成了消歧算法的设计。实验表明,该系统具有较高的切分正确率与效率。

It is a trend that MIS provide a Chinese indexing interface,its main difficulty is how to let the computer comprehend Chinese In this paper we set up a word segmentation system of Chinese intelligent indexing interface and propose its word segmentation strategy. After disussing the ambiguity problems of Chinses word segmentation, we give a diambiguity alogrithem based on mutual information and difference t-test. The experimental results show that it has a high segmentation accuracy rate and efficiency.