PhishSim: Aiding Phishing Website Detection With a Feature-Free Tool

In this presentation, the presenter will explore PhishSim, a cutting-edge tool designed to enhance phishing detection by utilizing a feature-free method. Traditional phishing detection methods often rely on specific website features, which can become outdated as phishing techniques evolve. PhishSim addresses this challenge by employing the Normalized Compression Distance (NCD), a parameter-free similarity measure that eliminates the need for feature extraction. The tool examines the HTML content of webpages, comparing them to known phishing sites to accurately classify new threats. Additionally, they will discuss the implementation of an incremental learning algorithm, allowing PhishSim to continuously adapt to new phishing strategies without the need to re-extract features. This presentation will highlight PhishSim’s effectiveness, demonstrated by its high accuracy and low false positive rate, making it a robust solution for real-world deployment in information security. Attendees will gain insights into how this innovative approach can be integrated into existing security frameworks to better protect against evolving phishing threats.