Design a new Multiobjective Algorithm for Data lustering

Abstract

Clustering is an unsupervised learning technique. It is a collection of objects that are grouped together on the basis of distance measure. As the number of population increases the data is also increasing, so we need to organize this data based on their similarities. The problem in clustering is single-objective because due to vast data, results are not accurate and performance are not that much good. In this project, clustering is seen as multi-objective rather than single-objective. In multi-objective clustering, more than one objective is optimized simultaneously and aim of multiobjective is to improve the performance of data clustering. Vibrating Particle System (VPS) algorithm is used for optimization in multiobjective clustering. The results of the multiobjective clustering algorithm are more accurate than that of the single-objective algorithm. Two objectives that are optimized is compactness and connectedness. The first objective is intra-cluster variance, we have to compute the distance of the object to the nearest cluster center and we can also call that overall deviation of a partitioning. The second objective is connectedness of the cluster, neighboring data objects have to identify whether they belong to the same cluster or not. Using these two objectives we will try to achieve a more accurate result, better performance, and efficiency.