Cost-Sensitive Spam Detection Using Parameters Optimization and Feature Selection Sang Min Lee (Department of Computer Engineering, Korea Aerospace University, Seoul, Korea minuri33@kau.ac.kr) Dong Seong Kim (Department of Electrical and Computer Engineering, Duke University, Durham, USA dongseong.kim@duke.edu) Jong Sou Park (Department of Computer Engineering, Korea Aerospace University, Seoul, Korea jspark@kau.ac.kr) Abstract: E-mail spam is no more garbage but risk since it recently includes virus attachments and spyware agents which make the recipients’ system ruined, therefore, there is an emerging need for spam detection. Many spam detection techniques based on machine learning techniques have been proposed. As the amount of spam has been increased tremendously using bulk mailing tools, spam detection techniques should counteract with it. To cope with this, parameters optimization and feature selection have been used to reduce processing overheads while guaranteeing high detection rates. However, previous approaches have not taken into account feature variable importance and optimal number of features. Moreover, to the best of our knowledge, there is no approach which uses both parameters optimization and feature selection together for spam detection. In this paper, we propose a spam detection model enabling both parameters optimization and optimal feature selection; we optimize two parameters of detection models using Random Forests (RF) so as to maximize the detection rates. We provide the variable importance of each feature so that it is easy to eliminate the irrelevant features. Furthermore, we decide an optimal number of selected features using two methods; (i) only one parameters optimization during overall feature selection and (ii) parameters optimization in every feature elimination phase. Finally, we evaluate our spam detection model with cost-sensitive measures to avoid misclassification of legitimate messages, since the cost of classifying a legitimate message as a spam far outweighs the cost of classifying a spam as a legitimate message. We perform experiments on Spambase dataset and show the feasibility of our approaches. Keywords: Feature Selection, Intrusion Detection, Parameters Optimization, Random Forests, Spam Detection, Spambase Categories: I.2.6, I.5.1, K.6.5, L.4.0 1 Introduction An electronic mail (e-mail) is an efficient and increasingly popular communication method. Concern about the proliferation of unsolicited bulk e-mail, commonly referred to as “spam”, has been steadily increasing [Cranor and LaMacchia 98]. When people receive in a small amount of spam, it rarely poses a significant problem. Journal of Universal Computer Science, vol. 17, no. 6 (2011), 944-960 submitted: 15/5/10, accepted: 30/11/10, appeared: 28/3/11 © J.UCS