Decision Model Analysis for Spam

Abstract

One of the security challenges of e-Government is to offer a fluid dialogue to citizens that guarantees the availability, confidentiality and integrity of the information interchanged. Spam jeopardizes the sur- vival of electronic mail as a communication mean. Many approaches to solve the spam problem has been proposed. This paper shows the ne- cessity of studying the real value of spam filters. Contrary to common belief, false positive rate and false negative rate do not completely reveal to what extent a junk filter is worth using. Very important parameters like the hostility of the environment (summarized by the probability of receiving spam) or the error costs associated with the filter play a decisive role.