摘要： 应用一种新的分词算法对汉语文本进行分词,该算法具备不使用词典和语法知识、不使用汉语词法规则、无监督等特点。采用模糊聚类算法对汉语文本进行聚类,该模糊聚类算法优于普通的聚类算法,聚类结果能充分体现汉语文本的多样性和大量性的特点。

关键词: 文本聚类, 分词, 模糊c-原型算法, 非欧几里德关系数据, 竞争凝聚算法

Abstract: This paper uses a new algorithm of segmentating Chinese document, which doesn't rely on a lexicon, grammar, word-formation heuristics or pre-segmented data. Then it applies FCA (Fuzzy Clustering Algorithm) to Chinese document clustering. FCA is superior to other ordinary clustering algorithm and the result can embody the wide diversity and large amount of Chinese document.

中图分类号: 

• TP391.1