SVM在数据挖掘中的应用

Application of Support Vector Machine in Data Mining

随着数据库系统使用的普及，数据库的规模也越来越大，如何从海量数据库中挖掘出有用的信息以供企事业单位使用，已经越来越引起人们的兴趣。支持向量机(SVM)是在统计学习理论的基础上发展起来的一种新的机器学习方法，它基于结构风险最小化原则，能有效地解决过学习问题，具有良好的推广性能和较好的分类精确性。该文首先介绍统计学习理论和支持向量机的概念，然后进一步论述了SVM在数据挖掘中的应用。

With the popular use in all kind of fields, the scale of DB system is increasingly large. The problem, how to discover the useful information from such huge data, is catching more and more peoples eyes. SVM is a kind of machine learning method, developing on statistics learning theory. Based on structure risk minimum principal, SVM can efficiently solve the learning problem, with the strengths of good generalization and correct classification. This paper firstly introduces some concepts of SLT and SVM, then discusses SVMs application on data mining.