Ensemble Classification and Extended Feature Selection for Credit Card Fraud Detection

Due to the rise of technology, the possibility of fraud in different areas such as banking has increased. Credit card fraud is a crucial problem in banking and its danger is ever increasing. This paper proposes an advanced data mining method, considering both the feature selection and the decision cost for accuracy enhancement of credit card fraud detection. After selecting the best and most effective features, using an extended wrapper method, an ensemble classification is performed. The extended feature selection approach includes a prior feature filtering and a wrapper approach using C4.5 decision tree. Ensemble classification is performed using cost sensitive decision trees in a decision forest framework. A locally gathered fraud detection dataset is used to estimate the proposed method. The method is assessed using accuracy, recall, and F-measure as the evaluation metrics and compared with the basic classification algorithms including ID3, J48, Naïve Bayes, Bayesian Network, and NB tree. The experiments carried out show that considering the F-measure as the evaluation metric, the proposed approach yields 1.8 to 2.4 percent performance improvement compared to the other classifiers.