A binary classification method based on class space model

Aiming at the shortages that text classification depends on the vector space model and document frequency (DF) feature extraction method in binary classification in the recent time, a binary classification method based on difference frequency space model is proposed. This method breaks through the restriction of vector space model, extracts the features with improved methods of DF of difference frequency, and the function of the binary classification is implemented. Experimental results show that the improved method is effective. The precision, the Recall ratio and F1 test value in the classification results are all improved, the precision of the classification is also increased. In addition the method in this paper can be also used in the binary classification of other domains.