将传统的数据挖掘领域中的聚类技术引入网络信息搜索中(Web聚类技术),以解决搜索引擎结果集不便于用户浏览问题为目的。分析维吾尔语的词法特点,选择适合维吾尔文文本聚类的特征,构造可扩展后缀树,选择基类,合并基类,把维吾尔文网页按相似度分成不同的类并形成类标签。 The traditional data mining field of clustering technology into the network information search (Web clustering technology), in order to solve the search engine result set is not easy for users to browse for the purpose of the problem. This paper analyzes the lexical characteristics of Uyghur language, selects features suitable for Uyghur text clustering, constructs an extensible suffix tree, selects base classes, merges base classes, divides Uyghur web pages into different classes according to similarities and forms class labels.