Fig. 2
From: A voting-based machine learning approach for classifying biological and clinical datasets
The flowchart of the suggested Trader optimization algorithm for selecting a near-optimal subset of features/genes. The algorithm generates some random candidate solutions (CS) and evaluates them using the value of accuracy obtained from the support vector machine (SVM) classifier. Next, the algorithm divides the CSs into several groups and modifies them with three operators. At the final step, the best acquired CS is introduced as a near-optimal subset of features/genes, which can enhance the prediction ability of the SVM classifier