Finding Playing Styles of Badminton Players Using Firefly Algorithm Based Clustering Algorithms

Cluster analysis can be defined as applying clustering algorithms with the goalof finding any hidden patterns or groupings in a data set. Different clusteringmethods may provide different solutions for the same data set. Traditionalclustering algorithms are popular, but handling big data sets is beyond theabilities of such methods. We propose three big data clustering methods basedon the firefly algorithm (FA). Three different fitness functions were definedon FA using inter-cluster distance, intra-cluster distance, silhouette value, andthe Calinski-Harabasz index. The algorithms find the most appropriate clustercenters for a given data set. The algorithms were tested with nine popularsynthetic data sets and one medical data set and are later applied on two bad-minton data sets with the intention of identifying the different playing stylesof players based on their physical characteristics. The results specify that thefirefly algorithm could generate better clustering results with high accuracy.The algorithms cluster the players to find the most suitable playing strategyfor a given player where expert knowledge is needed in labeling the clusters.Comparisons with a PSO-based clustering algorithm (APSO) and traditional al-gorithms point out that the proposed firefly variants work in a similar fashion asthe APSO method, and they surpass the performance of traditional algorithms.