Feature subset selection based on ant colony optimization and support vector machine

One of the significant research problems in pattern recognition is the feature subset selection. It is applied to select a subset of features. from a much larger set, through the elimination of variables that produce noise or strictly correlated with other already selected features, such that the selected subset is sufficient to perform the classification task. A hybrid method using ant colony optimization and support vector machine is proposed. The ant colony optimization searches the feature space guided by the result of the SVM. The tests on datasets show the effectiveness of the method.