Adaptation of the Rounding Search-Based Algorithm for the k-Clustering Minimum Completion Problem

This study proposes an algorithm based upon the rounding strategy for the k-clustering minimum completion problem. An instance of the problem is defined in a complete bipartite graph of S and C vertices. The goal of the problem is to decompose the initial graph into k-clusters, where each cluster is a complete bipartite subgraph. Since the problem is NP hard, any exact solver, like Cplex, is often not sufficient to achieve solutions with relatively hight quality. Thus, we propose a first alternative solution procedure for tackling large-scale instances. The designed method can be viewed as a special variant of the rounding search-based algorithm and it can be applied for solving several complex optimization problems. The proposed algorithm is evaluated on a set of benchmark instances related to the k-clustering minimum completion problem, where its achieved results are compared to the best results available in the literature.