Intelligent Phishing Detection: A Machine Learning and Deep Learning Approach

Phishing attacks represent a major risk to internet security by tricking users into disclosing confidential data. As phishing methods become more sophisticated, traditional detection systems have proven less reliable. This study explores advanced phishing detection techniques leveraging machine learning and deep learning models. The study conducts a detailed comparison of various algorithms, including Logistic Regression, Decision Trees, Random Forests, Support Vector Machines, Naive Bayes, and Artificial Neural Networks, analyzing them as individual models and in combination with boosting approaches such as AdaBoost, Gradient Boosting, and XGBoost. The analysis is performed using an extensive dataset of 1 1, 4 3 0 unique URLs through a systematic workflow involving feature selection, preprocessing, model training, and performance assessment. This investigation supports the development of phishing detection frameworks by presenting important analysis of traditional and modern algorithmic approaches, stressing their influence on detection performance improvement.