A hyper-parameter tuning approach for cost-sensitive support vector machine classifiers

Abstract In machine learning, hyperparameter tuning is strongly useful to improve model performance. our research, we concentrate attention on classifying imbalanced data by cost-sensitive support vector machines. We propose a multi-objective approach that optimizes model’s hyper-parameters. The devised for data. Three SVM performance measures are optimized. present the algorithm in basic version based genetic algorithms, and as an improved algorithms combined with decision trees. tested benchmark datasets either serial parallel version. reduces computational time needed finding optimized results empirically show suitable evaluation should be used assessing classification of models