Suspicious E-Mail Detection using Keywords

Abstract

The practice of looking at substantial previous databases so as to produce new data and relationship among them. In this paper a Decision tree in categorizing the suspected mail recognition (emails about crimes). In view of the hypothesis of fraud a suspicious email will have suspicious words and action words. The words such as attack, hijack, RDX, bomb, etc represent the suspicious and action words. We connected this hypothesis to the mail training dataset and then connected ID3 algorithm to create the decision tree. The decision tree then is utilized to categorize the mail as suspicious or not. Specifically, we are keen on recognizing crimes from such information. E-mail communication has become a part of everyday life for millions of people and has changed the way we work. So, it is very important to develop such system to prevent and suspect the criminal activities over the internet.