Support Vector Machine And Its Applications In Multiple Attribute Decision Making

Support Vector machine(SVM), a new machine learning technique based on the statistical learning theory, was proposed by Vladimir N.Vapnik and the study Group during the middle of the 1990s, SVM has these theoretical properties as well as the considerable performance of learning, recently, it becomes a new research hotspot and attracts more attentions after the research of Artificial Neural Networks.Today, Statistical Learning Theory is the transition period from theory to practice, the algorithms of SVM must be improved for the need of the practical application. This paper starts with an introduction to the foundational theory and an analysis of the properties of SVM. This paper studies SVM and its applications in Multiple Attribute Decision Making from the respective of integration of theory, algorithm and application. The main work of this paper can be described as follows:Firstly, the background knowledge of the paper is introduced and the theory of support vector machines is studied, On this basis, In view of the outlier and noise points in the samples, the processing of Fuzzy Support Vector Machine, for membership of design, under the hyper plane classification principle of Support Vector Machine, the membership of the design method based on the plane of class is proposed, the results of experiments in the artificial data and the real data prove the effectiveness and feasibility of the algorithm. Then, the method of Multiple Attribute Decision Making based on the support vector machines is studied, for the interval number Multiple Attribute Decision Making question, we uses the method of interval number quantifying to transform it as Multiple Attribute Decision Making, the positive and negative ideal points and the center points structure study sample. Multiple Attribute Decision Making based on support vector machines method is used to solve, the concrete experiment indicates that the more effective utility function can be established and the intelligent decision-making is realized; finally, the summary of the full text is carried on, the directions of further research are pointed out.