CyberFence : Intelligent Defence Against Phishing link

Phishing continues to be one of the most prevalent and damaging forms of cybercrime, exploiting social engineering techniques to deceive users into divulging sensitive information such as login credentials, banking details, and personal data. Traditional defenses such as spam filters, antivirus software, and employee awareness campaigns have proven insufficient against the growing sophistication of phishing attacks. This research proposes a machine learning–based phishing domain detection system that can identify malicious URLs before users interact with them, thereby minimizing the risks of compromise. The system analyzes domain-level and lexical features—such as URL length, numerical patterns, IP address usage, and entropy—to classify URLs as benign or phishing. By leveraging modern machine learning algorithms and a balanced dataset of phishing and legitimate domains, the model achieves efficient real-time classification that can be integrated into web browsers, email gateways, and social platforms. The implementation incorporates Python, TensorFlow/Keras, and Scikit-learn for model training, with FastAPI for backend services and a React-based frontend for usability. This approach not only reduces user dependency as the last line of defense but also enhances proactive cybersecurity measures, offering a scalable, adaptable, and cost-effective solution to counter phishing in the digital era.