Dimensionality reduction algorithm based on p-Laplacian in spectral clustering

In recent years,spectral analysis approaches have received much attention in machine learning and data mining areas,due to their rich theoretical foundations.This paper addressed the problem of Laplacian in spectral clustering,which couldn't get ideally graph cut criterion,by introducing the p-Laplacian operator,this paper proposed a new dimensionality reduction algorithm based on p-Laplacian.The experimental results denote that,the approach can get a approximation of the optimal graph cut,and can accurately get embedding mapped of original data in low dimensionality space.