arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1408.2869cs.LGstat.ML
收藏

基于聚类的支持向量机RBF核

Cluster based RBF Kernel for Support Vector Machines

Wojciech Marian Czarnecki, Jacek Tabor

摘要:

在经典的高斯SVM分类中,我们使用特征空间投影将点转换为具有固定协方差矩阵的正态分布(标准RBF中的恒等式和Mahalanobis RBF中整个数据集的协方差)。在本文中,我们添加额外的信息,高斯支持向量机考虑局部几何相关的特征空间投影。我们强调,我们的方法实际上是一个新的高斯型内核的建设算法。我们表明,更好的(相比标准的RBF和Mahalanobis RBF)分类结果得到在简单的情况下,空间是初步划分为两个集的k-means和点表示为正态分布的协方差计算根据数据集分区。我们将构造的方法C称为$_k$ RBF,其中$k$代表k均值中使用的聚类数量。我们在UCI存储库的9个数据集上实证表明,C $_2$ RBF增加了网格搜索的稳定性(以找到好参数的概率来衡量)。

英文摘要:

In the classical Gaussian SVM classification we use the feature space projection transforming points to normal distributions with fixed covariance matrices (identity in the standard RBF and the covariance of the whole dataset in Mahalanobis RBF). In this paper we add additional information to Gaussian SVM by considering local geometry-dependent feature space projection. We emphasize that our approach is in fact an algorithm for a construction of the new Gaussian-type kernel. We show that better (compared to standard RBF and Mahalanobis RBF) classification results are obtained in the simple case when the space is preliminary divided by k-means into two sets and points are represented as normal distributions with a covariances calculated according to the dataset partitioning. We call the constructed method C$_k$RBF, where $k$ stands for the amount of clusters used in k-means. We show empirically on nine datasets from UCI repository that C$_2$RBF increases the stability of the grid search (measured as the probability of finding good parameters).

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑