文档介绍：摘要
支撑矢量机(SVM,Support Vector Machine)是基于统计学习理论的一种模式识别方法。使用结构风险最小化原则替代经验风险最小化原则,避免了一些长期困扰其他模式识别方法的问题,使它对于小样本学习有着较好的处理能力。利用核函数,把非线性空间的问题转换到线性空间上来解决,降低了算法的复杂度。由于具有得天独厚的优点(完备的理论基础和较好的学习性能),使它成为当前模式识别领域研究的热点。
首先对SVM的理论基础—统计学习理论和相关概念进行了介绍。然后对二类SVM实现算法进行深入研究,并对他们的性能进行分析,对其优缺点进行了总结。下来对多类分类及实现算法进行了研究与分析,并对他们的算法特点进行对比。
针对大规模训练集,提出了一种增量学习方法。这种算法通过分析SV分布的特点,采用小规模的矩阵运算来代替大规模的矩阵运算。实验结果表明,该算法有效的提高了训练速度。
关键词支撑矢量机统计学习理论模式识别
Abstract
Subject : Pattern Recognition Algorithm Based on Support Vector Machine
Specialty : Applied Mathematics
Name :
Instructor :
ABSTRACT
Support vector machine is a pattern recognition algorithm based on statistical learning theory. Being substituted structural risk minimization for empirical risk minimization, support vector machine solves some problems puzzling pattern recognition field within a long period. Support vector machine has a good ability processing less simples. A nonlinear problem can be transformed to a linear problem with using kernel functions. The transformation plexity of the algorithm. With some highlights such as perfect theories, support vector machine is a hot point in pattern recognition field nowdays.
Theoretical basis for SVM-related statistical learning theory and concepts are introduced at the beginning. Some analysis of 2-category SVM algorithms performance and a summary of their advantages and disadvantages be done in chapter 2. Realization of multi-category classification algorithms be researched and analyzed in the next chapter. parisons of their features algorithm be done in same chapter.
For massive training sets, a incremental learning methods be proposed in chapter 4. Such algorithms through analysis Support Vectors distribution characteristics, use small-scale matrix operations to replace large-scale matrix operations. Experimental results show that the algorithm effectively improve training speed.
Keywords : Suport Vector Machine; Statistical Learning Theory; Pattern Recognition
Thesis : Applied Reserch