Abstract: In text categorization, one problem is usually confronted with feature spaces containing 10, 000 dimensions and more, even exceeding the number of available training samples. In order to enhance the operating speed and reduce the memory space occupied and filter out irrelevant or lower degree of features, feature selection algorithms must be used. In order to obtain more representative feature subset, it firstly presented document frequency method based on minimum word frequency, and then introduced rough sets and presented an algorithm of attribute reduction based on Beam scarch,finally, combined the attribute reduction algorithm with document frequency method based on minimum word frequency and proposed a comprehensive feature selection algorithm. The comprehensive algorithm firstly uses document frequency method based on minimum word frectuency to select feature, and then use the attribute reduction algorithm to eliminate redundancy, so can acquire the feature subset which arc more representative. Experimental results show that the comprehensive algorithm is effective.

Key words: Word frequency,Document frequency,Rough set,Beam search,Attribute reduction