Research Article | Open Access
DETECTING PHISHING WEBSITES: EXPLORING VARIOUS ML CLASSIFIERS FOR ACCURACY AND ROBUSTNESS
Mrs A.DIVYA, LAKKARAJU SAI SRI HARSHA, KONDA VASUDEV, NOOKALA MOKSHITHA PREEYA, KALLU YASHWANTH REDDY
Pages: 825-830
Abstract
Machine learning techniques have emerged as a powerful arsenal in the ongoing battle against phishing websites. In this relentless pursuit of cyber resilience, a comprehensive project delves into the realm of ML to identify and classify these fraudulent sites. The study employs a diverse set of classifiers, including Gradient Boosting, Catboost , Support Vector Machine, Decision Tree, K-nearest neighbors (KNN), Logistic Regression, Naive Bayes, and Random Forest, offering a multifaceted approach to tackle the multifarious nature of phishing websites. The heart of this project lies in the rigorous Exploratory Data Analysis, which scrutinizes a vast dataset to extract critical insights. The findings underscore the importance of specific features such as "HTTPS," "Anchor URL," and "Website Traffic" as pivotal indicators for distinguishing phishing URLs from legitimate ones. These features, in conjunction with machine learning models, contribute significantly to the robustness and reliability of the system, ensuring that even the most sophisticated phishing websites can be detected and Moreover, this project pays meticulous attention to data quality, employing techniques to eliminate outliers and address missing values, which further enhances the models' accuracy. This quality assurance ensures that the ML models are well-equipped to combat the ever-evolving tactics employed by cyber criminals in their pursuit of sensitive data. The convergence of advanced ML algorithms and feature-rich datasets represents a formidable weapon in the ongoing battle against phishing websites, reinforcing cyber security and safeguarding individuals and organizations from the pernicious threats of the digital realm.
Keywords
Machine learning techniques have emerged as a powerful arsenal in the ongoing battle against phishing websites.