Unsupervised Multi-View K-Means Clustering Algorithm

Since advanced technologies via social media, internet, virtual communities and networks internet of things (IoT), there are more multi-view data to be collected. Multi-view clustering is a substantial tool as natural design for data. K-means (KM) (single-view) had been extended handling data, called KM (MV-KM). In the literature, most MV-KM algorithms reported influenced by initializations also need given number clusters. this paper, we propose an unsupervised type algorithm so that it can automatically find optimal clusters without any initialization. We call (U-MV-KM). Moreover, three cluster validity indices, Dunn index (MV-Dunn), generalized (MV-G-Dunn) modified (MV-M-Dunn) indices algorithms. make experiments on some synthetic real sets comparisons with existing Based experimental results comparisons, proposed U-MV-KM actually shows good results. apply sets, demonstrate superiority usefulness