Abstract
For the kernel K-mean cluster method is run in an implicit feature space, the initial and iterative cluster centers cannot be defined explicitly. Against the deficiency of the initial cluster centers selected in the original space discretionarily in the existing methods, this paper proposes a new method for ensuring the clustering center that virtual clustering centers are defined in the feature space by the original classification as the initial cluster centers and the iteration clustering centers are ensured by the further virtual classification. The improved method is used for fault diagnosis of roller bearing that achieves a good cluster and diagnosis result, which demonstrates the effectiveness of the proposed method.
Highlights
The kernel K-mean cluster method is a combination of kernel method and K-mean method that the original items are embedded into a vector space called the feature space fist, and K-mean cluster is performed in the feature space [1,2]
For the kernel K-mean cluster method is run in an implicit feature space, the initial and iterative cluster centers cannot be defined explicitly
Kong [6] adopted the kernel K-mean clustering, in which the initial cluster centers are arbitrarily given in the original space, and the iteration clustering centers are obtained according to the previous classification results, which choose every sample of every category as a clustering center respectively, to calculate the intra-class distance between the other sample points and the clustering center, the iteration cluster centers are the point with the mini- mum sum of distances
Summary
The kernel K-mean cluster method is a combination of kernel method and K-mean method that the original items are embedded into a vector space called the feature space fist, and K-mean cluster is performed in the feature space [1,2]. For the kernel K-mean cluster method is run in an implicit feature space, the initial and iterative cluster centers cannot be defined explicitly. Zhang [5] proposed the kernel clustering algorithm, in which the initial cluster centers are given freely in the original space and the iteration clustering centers are changed through transforming the kernel matrix to realize the algorithm iterative. This process is complicated and difficult to understand. The proposed method is applied in the diagnosis of rolling bearing clustering to verify the effectiveness of the method
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have