MBA Journals Peer-reviewed management research from Consortium e-Learning Network

NOLEGEIN Journal of Management Information Systems

Phishing URL Detection in the Era of Machine Learning: A Comprehensive Analytical Review

  1. Piyush Rakhecha
  2. Ankit Kumar Tiwari
  3. Girraj Khandelwal

Abstract

The rapid expansion of internet usage and the exponential growth of digital data have significantly increased the global attack surface, leading to a rise in sophisticated cyber threats. Among these, phishing has emerged as one of the most widespread and persistent forms of cyberattacks. Phishing involves the use of social engineering and deceptive techniques to trick users into revealing sensitive information such as login credentials, financial data, and personal details, often resulting in severe security breaches. This paper reviews the application of machine learning (ML) techniques for phishing URL detection, which offer automated and scalable solutions to identify malicious activities. It explores various feature extraction approaches, including lexical, host-based, and content-based features, that help differentiate between legitimate and phishing URLs. The study also discusses key challenges in ML-based detection systems, such as imbalanced datasets, evolving attack patterns, feature selection complexity, and the need for real-time detection with low false positive rates. Furthermore, an experimental evaluation is conducted using the UCI Phishing Websites Dataset to compare the performance of multiple models, including Random Forest, XGBoost, LSTM, CNN, and a CNN-LSTM hybrid. Results show that Random Forest achieves the highest accuracy of 97.38%, demonstrating its effectiveness and reliability in phishing detection.
Support