A Comparative Analysis of Evolutionary Algorithms for Data Classification Using KEEL Tool

Evolutionary algorithms are inspired by the biological model of evolution and natural selection and are used to solve computationally hard problems, also known as NP-hard problems. The main motive to use these algorithms is their robust and adaptive nature to provide best search techniques for complex problems. This paper presents a comparative analysis of classification of algorithm's family instead of algorithms comparison using KEEL tool. This work compares SSMA-C, DROP3PSO-C, FURIA-C, GFS-MaxLogitBoost-Cand CPSO-C algorithms. Further, these selected evolutionary algorithms are compared against two statistical classifiers using the Wilcoxon signed rank test and Friedman test on following datasets—bupa, ecoli, glass, haberman, iris, monks, vehicle, and wine—to calculate classification efficiencies of these algorithms. Experimental results reveal some differences among these algorithms. Visualization module in the model has been used to give overall results as a summary while statistical test using Clas-Wilcoxin-ST compared the algorithms in a pair-wise fashion to conclude experimental findings.