Clustering algorithm based on kernel methods and its application
-
摘要: 在分析核方法的核心概念基础上,提出了一种基于核方法的聚类算法.通常,传统聚类算法只在数据特征差异较大时才有效,当数据特征差异较小时,很难取得较好的聚类效果.引入核函数,将原始数据由数据空间映射到特征空间,在特征空间中进行聚类.核函数的非线性映射使得原始数据的特征更完整地显现出来,从而能够更客观准确地聚类.与传统聚类方法相比,该方法聚类结果更客观有效.以16组实际数据为例,将该方法应用于数据分类研究中,聚类结果表明了该方法的可行性和有效性,从而为数据分类提供了一种新的可行方法.Abstract: Based on the analysis of the core concepts of the kernel methods, a clustering algorithm based on kernel methods was put forward. In general, traditional clustering algorithms are suitable to implement clustering only if the feature differences of data are large. If the feature differences are small and even cross in the original space, it is difficult for traditional algorithms to cluster correctly. By using kernel functions, the data in the original space was mapped into a high-dimensional feature space, in which more features of the data were exposed so that clustering could be performed efficiently. Compared with the traditional clustering methods, this clustering method had superiorities in dealing with the nonlinear data, which made its clustering result more objective and valid. This method was applied to the classification of 16 groups of data, and results show the feasibility and effectiveness of the kernel clustering algorithm.
-
Key words:
- clustering algorithm /
- kernel methods /
- feature space /
- kernel function /
- classification
-
[1] Muller K R, Mika S, Ratsch G, et al. An introduction to kernel-based learning algorithms [J]. IEEE Trans on Neural Networks, 2001,12(2):181-201 [2] Mika S, Ratsch G, Weston J, et al. Fisher discriminant analysis with kernels Neural Networks Signal Process Proc IEEE. Piscataway, NJ:IEEE, 1999:41-48 [3] Klinke S, Cook D. Binning of kernel-based projection pursuit indices in XGobi[J]. Computational Statistics & Data Analysis, 1997,25(3):363-369 [4] 肖健华.基于支持对象的野点检测方法[J].计算机工程,2003,29(11):43-45 Xiao Jianhua.Approach of outlier detection based on support objects [J]. Computer Engineering,2003,29(11):43-45(in Chinese) [5] 魏宏业,王建华,何葳.销售量预测的支持向量机建模及参数选择研究[J]. 系统仿真学报,2005,17(1):33-36 Wei Hongye,Wang Jianhua,He Wei. Study on support vector machines model for sales volume prediction and parameters selection[J]. Journal of System Simulation, 2005,17(1):33-36(in Chinese) [6] 肖健华,吴今培,杨叔子.基于SVM的综合评价方法研究[J].计算机工程,2002,28(8):28-30 Xiao Jianhua, Wu Jinpei,Yang Shuzi.Approach of evaluateon system based on support vector machine[J].Computer Engineering, 2002, 28(8):28-30(in Chinese) [7] 李焕荣,林健.基于一类分类方法的多类分类及其应用[J].华南理工大学学报(自然科学版),2004,32(8):82-88 Li Huanrong, Lin Jian. Multiclass classification based on the one-class classification and its application [J]. Journal of South China University of Technology (Natural Science), 2004, 32(8):82-88(in Chinese) -
点击查看大图
计量
- 文章访问数: 3443
- HTML全文浏览量: 224
- PDF下载量: 1459
- 被引次数: 0
下载:
