A probabilistic approach for multi-objective clustering using game theory

Multi-Objective clustering as the most important and fundamental unsupervised learning has been in the gravity of focus of quite a lot numbers of researchers over several decades. In this paper, we suggest a multi-objective clustering technique based on the notion of game theory. The presented method is designed to optimize two intrinsically conflicting objectives, named, compaction and equi-partitioning. The key contributions of the proposed approach is that the proposed method performs better off by utilizing the advantages of mixed strategies as well as those of pure ones, considering the existence of mixed Nash Equilibrium in every game. The approach known as Mixed Game Theoretic Kmeans offers the optimal solution in a very promising manner by optimizing both objectives simultaneously. The experimental results suggest that the proposed approach significantly outperforms other rival methods across real world and synthetic data sets.