A Fast Nature Inspired Algorithm for Data Clustering

Abstract

Data clustering is a collection of data objects similar to one another within the same cluster and dissimilar to the objects in other clusters. The Shuffled frog leaping algorithm is a nature-inspired algorithm that mimics the natural biological evolution process of frogs. This algorithm also consists of elements like local search and exchanging information globally. This algorithm faces the problem of converging in local optima due to the limitations of the local search method used to explore search space. In this paper, an enhanced shuffled frog leaping algorithm is introduced for clustering to offer improved cluster solutions with good convergence and address the problem of local optima by exploring and exploiting nearly the full search space. The proposed algorithm uses a simulated annealing search method instead of a simple local search to improve the search behavior for selecting fitter solutions required in each iterations. The performance of the proposed algorithm is tested in terms of purity, entropy, completeness score (CS), homogeneity score (HS), and V-measure (VM) over six real-time time benchmark datasets. Fitness functions used to optimize are total within-cluster variance (TMCW) and the Silhouette coefficient (SC). The performance of the proposed algorithm is observed to be good when compared to the performance of existing algorithms.