Development of New Clustering Algorithm Based on Firefly Optimization

Abstract

Clustering is an unsupervised classification, is a group of clustering method, the clusters in the same group are very similar and the clusters in the other group are differentThis clustering can be done with many clustering algorithms, it is important to find the best cluster centers among the data. In this study, the fire optimization algorithm, which is a global search capability and used to solve many difficult problems, is used to find optimum cluster centers. The proposed clustering algorithm was tested on 12 data sets from UCI data warehouse. The test results of the proposed clustering algorithm are compared with the clustering algorithms of SFLA, ABC, PSO, Bayes Net, Mlp ANN, RBF, KStar, Bagging, Multi Boost, NB Tree, Ridor and VFI clustering algorithms. It has performed better than many clustering algorithms in many datasets.