基于ICA与聚类分析的支持向量机分类研究  

Research of SVM’s classification based on ICA and cluster

作　　者：彭红毅[1] 蒋春福[2] 杜明 

机构地区：[1]华南农业大学理学院统计系,广州510642 [2]深圳大学数学与计算科学学院,广东深圳518060 [3]南雄市地税局,广东韶关512400

出　　处：《计算机工程与应用》2008年第8期169-171,共3页Computer Engineering and Applications

基　　金：广东省科技公关计划( the Key Technologies R&D Program of Guangdong Province, China under Grant No.B10101033, No.A10202001);广东省哲学社会科学十一五规划项目( No.06M03); 华南农业大学校长基金( No.4900- K06166)

摘　　要：在ICA与聚类分析的基础上提出了一种改进的支持向量机分类模型——ICSVM模型。ICSVM模型中利用一种指标筛选算法与独立成分分析的方法将各数据指标转化为互相独立成分的数据指标。接着运用K-means方法对独立成分样本数据集进行聚类分析,再由获得的各子类中心数据构造初始的超平面,筛选出靠近初始超平面的支持类与亚支持类,并展开支持类与亚支持类中的样本数据点重新构造超平面,以便对数据进行分类。实验表明,对于样本比较多的数据集,与标准的SVM算法相比,ICSVM算法能够节约训练时间,同时能够提高分类的正确率。Based on ICA and cluster analysis,this paper proposes ICSVM model.ICSVM model makes use of a selecting indices’ algorithm and ICA to transform the correlative indices into independent indices firstly.Then an initial classes is got by K-means cluster,and an initial super plane is made through the center of all subclass.By that the support classes and sub-support classes neighboring the initial super plane can be selected.Then expand the sample data of the support classes and sub-support classes and build up a new super plane by using them.Thus data can be classified by the new super plane.Compared with standard SVM,ICSVM has both better correct rate of classification and better training speed of ICSVM.

关 键 词：支持向量机 独立成分分析 聚类分析 相关关系 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程]