Optimization Algorithms and the Clustering Problem

Optimization algorithms can play significant roles in enhancing and even changing the performance of clustering algorithms since the clustering problem is an optimization problem. This paper studies this point through an example of the k-means algorithm as it is the most widely utilized clustering method but it sometimes gets stuck at sub-optimal solutions. Meanwhile, optimization algorithms can avoid local optima besides their other merits. Therefore, many researchers have integrated several optimization algorithms with the clustering algorithms aiming at enhancing their performance and hence getting better clustering results. This work studies the utilization of genetic algorithms (as an example of optimization algorithms) besides the k-means algorithm for handling the clustering problem. It also demonstrates the clustering problem, its applications, etc.