Project Team
Students
Jiya Patel
Computer Science
Harrisburg
Faculty Mentors
Dr. Bimal Ghimire
Harrisburg
School of Science, Engineering, and Technology
Project
Project Video
Project Abstract
URLs (Uniform Resource Locator) often become the mode of cyber crime. From leaking data to online stalking, there are several ways in which the attackers can hurt the victims. As new ways of crimes emerge everyday, so must new ways of preventing them. In this paper, we discuss the different parts of URL, the existing methods of malicious URL detection from the traditional Blacklisting to the modern Machine Learning (ML) and even Deep Learning (DL) methods. We go through several ML and DL models followed by the feature selection process and the datasets used to train and test the models. Then we learn the advantages and challenges of the models to compare them. The analysis of different ML models helps us decide how accurate they are. Varied applications of DL methods open new possibilities for URL classification. The survey shows that there are hardly any models with ML and DL techniques combined, and whitelisting also has its benefits but it needs to be tested to see how scalable it is. These ideas have potential and can help with deciding the future actions needed. URLs need to be classified before the user clicks on them and this is the first step to be taken.
Evaluate this Project
Use this form link to provide feedback to the presenters, and add your project evaluation for award(s) consideration.