Clustering Multiple Views of Data through Many-objective Evolutionary Approach
Krishna Priyanka Ponnaganti, Muvva Venkateswara Rao, Lekhana Pinninti, Anudeep Peddi, Sai Vignesh Chereddy, Lakshmikanth Paleti · 2023
Multiple data sources must be taken into account in several application areas. Each of those data views often offers a unique viewpoint on a certain group of things. Several data perspectives with varying degrees of dependability may also be produced in practice as a result of some complicated real-world challenges. The clustering algorithms in general can incorporate the information of given data points. Existing algorithms like Co-clustering and Multi-view K Means failed in considering the varying degrees of dependability of different data points in a single dataset and are limited to considering only two data views at a time. Here, to address these types of challenges, we propose the Multi-view Multiple Clustering (MVMC) approach. MVMC is designed to allow the inclusion of multiple data sources in the clustering process. Furthermore, it is capable of automatically adjusting the weights assigned to the different data views to obtain the best clustering results.