Using a Nearest Neighbor Rule for the Clustering Method Based on One-Class Support Vector Machines

In this paper, a nearest neighbor rule is applied to the clustering method based on one-class support vector machines. Although the traditional clustering method inspired the k-means clustering employs the kernel-based one-class support vector machines in improving the clustering performance, it forms the coarse decision boundaries. So this paper uses a nearest neighbor rule to establishing the better decision boundaries. Experimental results show that the novel clustering algorithm can increase the clustering accuracies according to a nearest neighbor rule.