Dissertation
Dissertation > Mathematical sciences and chemical > Mathematics > Control theory, information theory ( mathematical theory ) > Machine learning theory

Research based on support vector machine regression method

Author ZuoFei
Tutor HuangTingZhu
School University of Electronic Science and Technology
Course Computational Mathematics
Keywords Statistical Learning Theory Support Vector Machine Regression analysis Smooth function
CLC O234
Type Master's thesis
Year 2007
Downloads 192
Quotes 0
Download Dissertation

With the in-depth study of the theory of machine learning (LM) and data mining (DM) technology continues to improve, in the early nineties by Vapnik.V et al proposed a support vector machine (SVM) technology. It is built on top of the statistical learning theory (SLT), by means of optimization methods to solve new problems in the theory of machine learning and data mining. Traditional statistical and artificial neural network (ANN), support vector machine is mainly to study the laws and methods of statistical learning under conditions of small and medium-sized samples, and it used samples unlike statistics as to a specifically generated by the target sample. The practice shows that the support vector machine based on statistical learning theory is not only simple structure, but also has a strong theoretical and practical ability to promote. At present, the research on support vector machine has become a hot issue of the international artificial intelligence, database, and other fields. This article provides insight on the support vector machine technology. The major work of this paper are as follows: First, the simple in the first chapter introduces the basic concepts of the theory of machine learning and data mining and international research about them, and in general the introduction of a support vector machine technology. The next chapter describes the statistical regression problems, such as linear regression, log istic regression model. Again, in the third part of the paper, support vector machine technology in the first two chapters and regression analysis are combined, a systematic discussion of the data in support vector machine regression method, while the use of optimization tools to simplify the problem has been simple and reasonable structure model. Finally, in order to solve the data regression method of support vector machine is not smooth, especially in nondifferentiable on some of the inflection point in the text the fourth chapter of the system studied the smooth application of technology in support vector machine regression problem, given three in the theory and practice of smooth support vector machine regression method with good performance. At present, the research on support vector machine starting from the statistics, artificial intelligence, database, smooth support vector machine regression problem is relatively small. Support vector machine technology in both theory and practical application, it is of great significance.

Related Dissertations
More Dissertations