Classification of binary problems with SVM and a mixed artificial bee colony algorithm
Fatemeh Barani, Mina Mirhosseini · 2018
The feature selection is the one of important data preprocessing methods in classification problems. The more evolutionary algorithms have been used to reduce the features. In this paper, a new approach is proposed based on the support vector machine and the artificial bee colony algorithm, called MABC-SVM, to improve the accuracy classification using an effective subset of features. In order to simultaneously optimize the parameters of support vector machine and select an effective set of input features, a hybrid algorithm called Mixed-ABC has been proposed that combines the binary ABC algorithm with continuous ABC algorithm. The evaluation results on a number of data sets in the UCI show that in most cases the MABC-SVM presents the better results than other classification methods.