A new feature selection method based on support vector machines for text categorisation

As a machine intelligence paradigm, the support vector machines (SVMs) have tremendous potential for helping people to classify text document into a fixed number of predefined categories. The purpose of this paper is to discuss a new method of feature selection combined with principal component analysis and class profile-based feature as an input vector for SVMs classifier, and to demonstrate the effectiveness of this process. This paper also demonstrates that an applied method with SVMs improves categorisation performance and reduces the amount of time required to configure a learning machine.