SVM For Multiclass Recognition And Its Applications On Traffic Identification

In the early 90's, a new learning method had been proposed based on the statistical learning theory, called support vector machine (SVM). This thesis gives a detailed review of the SVM and proposes some improvements of the multiclass SVM. Then it proposes two new systems for traffic identification based on the binary SVM and multiclass SVM respectively.The main research contents of this thesis include two parts: the theory part and the application part.1. The theory part of this thesis also includes two parts:1) The first part gives a comprehensive survey of statistical learning theory and binary SVM. Then it gives a detailed discussion for some related knowledge like kernel methods and optimization problems.2) The second part makes an exhaustive review of the multiclass SVM and proposes two improvements based on the traditional methods. Then it demonstrates the rationality and the feasibility of the proposed methods.2. In the application part, the thesis combines SVM and the P2P traffic identification system innovatively and efficiently. It uses binary SVM and multiclass SVM to the practical P2P traffic identification problems respectively and gains faster speed and higher accuracy than the traditional methods. The experimental results show the successful combination of SVM and the traffic identification system.