Spectral Perturbation Meets Incomplete Multi-view Data
This addresses a realistic clustering scenario for multi-view data analysis, but it appears incremental as it builds on existing spectral clustering and matrix completion techniques.
The paper tackles incomplete multi-view clustering, where data instances are missing in some views, by linking spectral perturbation theory to the problem and proposing a method that transfers missing data issues to similarity matrices and uses matrix completion. Experimental results show the method is effective.
Beyond existing multi-view clustering, this paper studies a more realistic clustering scenario, referred to as incomplete multi-view clustering, where a number of data instances are missing in certain views. To tackle this problem, we explore spectral perturbation theory. In this work, we show a strong link between perturbation risk bounds and incomplete multi-view clustering. That is, as the similarity matrix fed into spectral clustering is a quantity bounded in magnitude O(1), we transfer the missing problem from data to similarity and tailor a matrix completion method for incomplete similarity matrix. Moreover, we show that the minimization of perturbation risk bounds among different views maximizes the final fusion result across all views. This provides a solid fusion criteria for multi-view data. We motivate and propose a Perturbation-oriented Incomplete multi-view Clustering (PIC) method. Experimental results demonstrate the effectiveness of the proposed method.