基于频繁特征项集的文档聚类研究

Documents clustering based on frequent term sets

提出了基于频繁特征项集的文档聚类方法。对预处理后的文档,通过Apriori算法找出文档频繁特征项集,依据其子集中频繁特征词语对相关文档进行聚类,该方法能够有效降低特征项的维数,并能够通过频繁特征词语集合对聚类后的类别进行适当的描述。

Novel method of frequent term sets-based text clustering was presented. For the pre-treated documents, such frequent sets could be efficiently discovered by using the Apriori method, which subset was used for documents clustering according to frequent terms＇ correlation. This method allows us to reduce drastically the dimensionality of the term sets and provides an understandable description of the discovered clusters by their frequent terms sets.