关键词: 特征选择, 文本分类, 词频, 文档频, 粗糙集, 属性约简

Abstract: Feature selection is one of the core research topics in text categorization.Word frequency and document frequency are analyzed simply.Category concentration based on word frequency and document frequency is presented.Set covering is introduced into rough sets and an attribute reduction algorithm based on minimal set covering is provided.A new feature selection method combined the provided attribute reduction algorithm with the category concentration is proposed.The new method uses the category concentration to select feature and filter out some terms to reduce the sparsity of feature spaces，and then employs the proposed attribute reduction algorithm to eliminate redundancy，so that the more representative feature subset is acquired.The experimental results show that the new method is promising.

Key words: feature selection, text categorization, word frequency, document frequency, rough sets, attribute reduction