Text Classification with Support Vector Machine and Back Propagation Neural Network

摘要

We compared a support vector machine (SVM) with a back propagation neural network (BPNN) for the task of text classification of XiangShan science conference (XSSC) web documents. We made a comparison on the performances of the multi-class classification of these two learning methods. The result of an experiment demonstrated that SVM substantially outperformed the one by BPNN in prediction accuracy and recall. Furthermore, the result of classification was improved with the combined method which was devised in this paper.