The Cooperation of Visible and Hidden Views in Multi-View Clustering

PROJECT TITLE :

Multi-View Clustering with the Cooperation of Visible and Hidden Views

ABSTRACT:

The use of multi-view data in real-world applications is becoming increasingly common, and as a result, numerous multi-view clustering algorithms have been proposed. Existing algorithms typically concentrate on how various visible views in the initial space work together, but they either ignore the influence that the hidden information has on these visible views or they only consider the hidden information that exists between the views. Since the algorithms do not make full use of the information that is available, they are inefficient. This is especially true with regard to the information concerning the otherness of the various views and the consistency of the information among them. Both the otherness information and the consistency information contained within multi-view data are extremely helpful when conducting efficient clustering analyses. In this research, an algorithm for multi-view clustering called MV-Co-VH, which stands for Multi-View clustering with Cooperation of Visible and Hidden views, is proposed. The MV-Co-VH algorithm begins by first projecting the multiple views from the various visible spaces to the common hidden space. This is accomplished through the utilization of non-negative matrix factorization in order to obtain the data for the common hidden view. After that, the procedure for clustering is modified to include collaborative learning, which is based on the visible views as well as the shared hidden view. Extensive testing on multi-view datasets from both the University of California, Irvine (UCI), as well as multi-view datasets from the real world produced results that demonstrate that the clustering performance of the proposed algorithm is on par with or even exceeds that of the algorithms that are currently in use.

Did you like this research project?

To get this research project Guidelines, Training and Code... Click Here