Faculty Advisor

Heineman, George T.

Project Center

MIT Lincoln Laboratory-Lexington, Massachusetts

Abstract

The rise of cybercrime has motivated the need for improved early detection and prediction mechanisms to prevent cyber-attacks from causing damage to unsuspecting victims. We developed and analyzed various machine learning algorithms to tackle one approach for early detection, URL classification. Unlike previous research, which focused on binary classification, our approach focuses on classifying URLs to their likely attack category. Through testing and evaluation, we found that ensemble methods perform the best with our optimal feature set, producing accuracies as high as 95%.

Publisher

Worcester Polytechnic Institute

Date Accepted

2019-10-12

Major

Computer Science

Project Type

Major Qualifying Project

Accessibility

Unrestricted

Advisor Department

Computer Science

Share

COinS