1 The Maersk Mc-Kinney Moller Institute, Faculty of Engineering, SDU2 The Maersk Mc-Kinney Moller Institute University of Southern Denmark, Campusvej 55, 5230 Odense M, Denmark3 The Maersk Mc-Kinney Moller Institute, Faculty of Engineering, SDU
In this paper, we present a fraudulent email detection model using advanced feature choice. We extracted various kinds of features and compared the performance of each category of features with the others in terms of the fraudulent email detection rate. The different types of features are incorporated step by step. The detection of fraudulent email has been considered as a classification problem and it is evaluated using various state-of-the art algorithms and on CCM  which is authors' previous cluster based classification model. The experiments have been performed on diverse feature sets and the different classification methods. The comparison of the results is also presented and the evaluations shows that for the fraudulent email detection tasks, the feature set is more important regardless of classification method. The results of the study suggest that the task of fraudulent emails detection requires the better choice of feature set; while the choice of classification method is of less importance.
Egyptian Informatics Journal, 2014, Vol 15, Issue 3, p. 169-174