DETECTION OF PHISHING URL BASED ON TEXT FEATURE EXTRACTION

Abstract

Phishing poses a significant threat to users' personal information by tricking them into divulging sensitive data through deceptive websites. This study aims to develop a robust system to safeguard users against such attacks. Our approach focuses on analyzing the content and URLs of web pages to identify potential phishing patterns. The process begins with parsing web pages to extract plain text terms and URLs. These elements are then assessed using advanced techniques like TF-IDF and a custom URL weighting system to determine their relevance in detecting phishing indicators. Furthermore, we employ a search engine lookup to correlate key terms, aiding in the identification of URLs likely to target victims. Additionally, WHOIS lookup is utilized to compare website registration details, ensuring accurate classification of websites as phishing or legitimate.By integrating these methodologies, our system aims to proactively mitigate phishing threats, thereby enhancing user security in the online environment.