Abstract

Phishing is a prevalent cyber threat that involves deceiving users into revealing sensitive information by masquerading as legitimate websites. To mitigate this, effective detection systems are essential. This study proposes a machine learning-based approach using Support Vector Machine (SVM) and Light Gradient Boosting Machine (Light GBM) algorithms for accurate phishing website detection. Various features such as URL-based, domain-based, and content-based attributes are extracted and analyzed. The model is trained and evaluated using a comprehensive dataset to compare the performance of both algorithms. Experimental results demonstrate that Light GBM outperforms SVM in terms of accuracy, precision, and recall. Additionally, the proposed system achieves high detection efficiency with minimal false positives, making it suitable for real-time applications. The use of feature engineering enhances the model's robustness, ensuring it adapts well to evolving phishing techniques. This research provides a scalable and effective solution for combating cyber threats, contributing to a safer online environment. Further advancements may include integrating additional data sources and optimizing model parameters to enhance detection accuracy.