Improved Jaya Optimization Algorithm for Feature Selection on Cancer Diagnosis Data using Evolutionary Binary Coded Approach
P. D. Sheth, S. T. Patil · Solid State Technology · 2020
Cancer Diagnostic Decision Support System may mislead the classification algorithms bymultiplicity of features. Hence, feature selection becomes an essential step in data mining of cancerdiagnosis data. Feature selection is a multi-objective optimization problem that systematically selectsa subset of most informative features for model building. Recently, randomization basedEvolutionary Algorithms (EAs) are becoming popular for feature selection than traditional filter andwrapper methods. EAs explore the entire search space using some heuristic techniques in less time.Jaya Optimization Algorithm (JOA) is a competitive Swarm Intelligence (SI) based EA, yet is simpleand easy to implement for solving optimization problems. The present work proposes BinJOA-SAlgorithm, an improved binary variant of JOA for solving feature selection problem on cancerdiagnosis data using Sigmoidal function. BinJOA-S algorithm employs the scalarization method tosolve feature selection as a multi-objective problem. The experimental results show that thealgorithm can obtain competitive performance when evaluated for parameters such as best, averageand worst fitness, average classification accuracy, best classification accuracy, the average number offeatures selected and, CPU computational time. It is observed that the proposed algorithm producesthe effective performance of the cancer diagnosis system