• Zainab
• Dr. Kumaraswamy S

Phishing is one of the most effective ways to obtain passwords and perpetrate online fraud. In this paper, we present a real-time URL classification method that employs many signals to identify if a URL is phishing or secure. The classification process includes reputation checks from reliable threat intelligence sources (URLVoid, McAfee, Sucuri), confirming the SSL certificate and WHOIS information for the domain, computing topological features for each domain (domain age, IP blacklist), and finally classifying the URL using an RF classifier trained on a variety of labeled phishing datasets. To assess the system's efficacy, we collected a range of data from Phishtank and OpenPhish in addition to a few secure URLs. to evaluate the accuracy, recall, ROC-AUC, and F1 score of the system. We also looked at how obfuscation techniques impact the changing nature of phishing and how it relates to the overall resilience of our system. Based on our findings, we believe that integrating data from several sources and feature sets, rather than relying just on one source or feature set, greatly improves the ability to detect phishing. Consequently, this research develops a real-time phishing detection system that is proven, useful, and effective. Additionally, it provides a framework for identifying which features are essential for identifying possible threats and improving the capabilities of a phishing detection system, as well as supporting the need to combine several signals.

Phishing Detection, URL Classification, Threat Intelligence Aggregation, Machine Learning, Real-Time Security, LightGBM, Random Forest.

"A DATA DRIVEN MULTI-SIGNAL FRAMEWORK FOR THE DETECTION OF PHISHING URLS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.13, Issue 1, page no.c515-c523, January-2026, Available :http://www.jetir.org/papers/JETIR2601262.pdf

"A DATA DRIVEN MULTI-SIGNAL FRAMEWORK FOR THE DETECTION OF PHISHING URLS", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.13, Issue 1, page no. ppc515-c523, January-2026, Available at : http://www.jetir.org/papers/JETIR2601262.pdf