Weighted multi-view co-clustering (WMVCC) for sparse data

Multi-view clustering has gained importance in recent times due to the large-scale generation of data, often from multiple sources. refers a set objects which are expressed by features, known as views, such movies being list actors or textual summary its plot. Co-clustering, on other hand, simultaneous grouping data samples and features under assumption that exhibit pattern only subset features. This paper combines multi-view with co-clustering proposes new Weighted Multi-View Co-Clustering (WMVCC) algorithm. The motivation behind approach is use diversity provided sources information while exploiting power co-clustering. proposed method expands objective function unified across all views. algorithm follows k-means strategy iteratively optimizes updating cluster labels, view weights. A local search also employed optimize result using weighted multi-step paths graph. Experiments conducted several benchmark datasets. results show converges quickly, performance significantly outperforms state-of-the-art algorithms sparse