Phishing URL Detection Using Random Forest Machine Learning

Phishing attacks continue to pose a significant threat to individuals and organizations by exploiting user trust to obtain sensitive information through fraudulent websites. As traditional blacklist-based detection methods often struggle to identify newly emerging phishing URLs, there is an increasing need for intelligent and adaptive detection techniques. This study presented a machine learning-based Phishing URL Detection system that utilizes the Random Forest classification algorithm to accurately distinguish between legitimate and phishing websites using URL-based features.   The proposed system was successfully implemented using Python and integrated with a Streamlit-based web application to provide a simple and interactive platform for real-time phishing URL prediction. Experimental evaluation demonstrated that the Random Forest classifier achieved an overall prediction accuracy of 96.97%, indicating its effectiveness in identifying phishing websites with high reliability. The integration of machine learning with a user-friendly interface enables users to verify suspicious URLs quickly and efficiently without requiring extensive technical knowledge.   The results obtained in this research confirm that URL-based feature analysis combined with ensemble learning techniques can provide an effective solution for phishing detection. Compared with conventional rule-based approaches, the proposed system offers improved adaptability and the ability to classify previously unseen phishing URLs based on learned patterns. Furthermore, the developed application demonstrates the practical applicability of machine learning in enhancing cyber-security and improving user awareness against phishing attacks.   Although the proposed system achieved promising performance, certain limitations remain. The current implementation focuses primarily on URL-based features and does not incorporate webpage content analysis, domain registration information, SSL certificate verification, or visual similarity analysis. Including these additional features may further improve detection accuracy and strengthen the system against more sophisticated phishing techniques. Future research can focus on evaluating additional machine learning and deep learning algorithms, such as Gradient Boosting, XGBoost, LightGBM, and Transformer-based models, to further enhance classification performance. The system can also be extended by integrating real-time threat intelligence services, browser extensions, and cloud-based deployment to support large-scale phishing detection in practical environments.   In conclusion, the proposed Phishing URL Detection Using Machine Learning system successfully achieved its objectives by providing a reliable, accurate, and user-friendly phishing detection solution. The achieved accuracy of 96.97% demonstrates the effectiveness of the Random Forest algorithm for phishing URL classification. The developed system contributes to improving online security by assisting users in identifying potentially malicious websites before they are accessed, thereby reducing the risks associated with phishing attacks and supporting safer internet usage.

[1]             R. M. Mohammad, F. Thabtah, and L. McCluskey, "Predicting phishing websites based on self-structuring neural network," Neural Computing and Applications, vol. 25, no. 2, pp. 443–458, Aug. 2014. [2]            F. Toolan and J. Carthy, "Feature Selection for Spam and Phishing Detection," Proceedings of the eCrime Researchers Summit (eCrime), Dallas, TX, USA, 2010, pp. 1–12. [3]            M. Aburrous, M. A. Hossain, K. Dahal, and F. Thabtah, "Intelligent Phishing Website Detection System Using Fuzzy Techniques," Proceedings of the International Conference on Information Networking and Automation, Kuala Lumpur, Malaysia, 2010, pp. 345–350.  [4]            C. Xiang, P. C. Yong, and L. S. Meng, "Detecting Phishing Websites Using Machine Learning Algorithms," International Journal of Computer Science and Information Security, vol. 16, no. 4, pp. 200–206, 2018.  [5]             F. Thabtah and D. Peebles, "A New Machine Learning Model Based on Induction of Rules for Phishing Websites Classification," Applied Soft Computing, vol. 68, pp. 329–342, Jul. 2018. [6]             S. Marchal, J. François, R. State, and T. Engel, "PhishStorm: Detecting Phishing with Streaming Analytics," IEEE Transactions on Network and Service Management, vol. 11, no. 4, pp. 458–471, Dec. 2014. [7]            T. Fawcett, "An Introduction to ROC Analysis," Pattern Recognition Letters, vol. 27, no. 8, pp. 861–874, 2006.  [8]            L. Breiman, "Random Forests," Machine Learning, vol. 45, no. 1, pp. 5–32, Oct. 2001.  [9]             F. Pedregosa et al., "Scikit-learn: Machine Learning in Python," Journal of Machine Learning Research, vol. 12, pp. 2825–2830, 2011. [10]            W. McKinney, Python for Data Analysis, 2nd ed. Sebastopol, CA, USA: O'Reilly Media, 2018.