PHIKIT A PHISHING KIT ATTACKS DATASET FOR PHISHING WEBSITES IDENTIFICATION FELIPE

Abstract

Phishing attacks remain one of the most prevalent cybersecurity threats, targeting users through fraudulent websites designed to steal sensitive information such as login credentials, financial details, and personal data. Detecting phishing websites accurately is challenging due to the increasing sophistication and variability of attack techniques. This study introduces PHIKIT, a curated dataset containing phishing kit attacks, aimed at improving the identification and classification of phishing websites. The dataset includes features extracted from URL structures, HTML content, domain registration details, and network behavior, enabling comprehensive analysis for machine learning and deep learning models. By leveraging PHIKIT, researchers and cybersecurity systems can train more robust detection models that distinguish phishing websites from legitimate ones with higher accuracy. Experimental results using various classifiers demonstrate that models trained on PHIKIT achieve improved detection rates, reduced false positives, and faster identification of phishing threats, providing a valuable resource for advancing anti-phishing solutions