Study on the Web Classification Based on Proximal Support Vector Machine

Classification of the documents is a very important task. Based on the proximal support vector machines (PSVM) classification method could solve classification problem with small training set,without too much loss of classification accuracy. This paper describes a new PSVM training algorithm based on descending dimension methods,which has faster training speed and smaller memory requirements advantages. Finally we apply the method to solve text classification problem. Experiments results show that the new classification algorithm has better classification performance.