Kernel Discriminant Analysis Algorithm for Document Categorization

Document categorization is one of the most crucial techniques to organize the documents in a supervised manner. To efficiently resolve document classification problems, a novel document classification algorithm based on kernel discriminant analysis (KDA) is proposed in this paper. The high-dimensional document sets are first mapped into lower-dimensional space with KDA, then the SVM is applied to classify the documents into semantically different classes. Experimental results demonstrate the effectiveness and efficiency of the proposed KDA algorithm.