A novel locality sensitive k-means clustering algorithm based on subtractive clustering
Lei Gu · 2016
A locality sensitive k-means clustering approach has been presented recently. This approach can increase clustering accuracies. However, it is affected by the initial centers and often attain the unstable clustering results. In this paper, a novel locality sensitive k-means clustering algorithm based on subtractive clustering is proposed. The initial centers are produced by subtractive clustering rather than the random way. Some experiments are done on several datasets to investigate the effectiveness of this new method. Experimental results show that our proposed method can improve the clustering performance compared to the traditional k-means, the subtractive clustering and the previous locality sensitive k-means.