NLP-DRIVEN METHODS FOR ENHANCED PHISHING WEBSITE DETECTION
V. Christy, S. Anbu selvan · 2025
Phishing websites pose a major cybersecurity risk by imitating legitimate platforms to steal sensitive information [1].Traditional detection methods, such as blacklists and heuristic approaches, struggle against evolving phishing tactics, requiring advanced solutions [2].Natural Language Processing (NLP) enhances detection by analyzing textual and structural elements of web content, identifying deceptive keywords and manipulative language[3].This study proposes an NLP-based phishing detection framework integrating tokenization, sentiment analysis, and machine learning classifiers to improve accuracy [4].Recent deep learning advancements, including CNNs and RNNs, further strengthen phishing detection by analyzing both text and visual elements [5].However, challenges remain, such as phishing websites constantly changing content and structure to evade detection [6].Multilingual NLP models and social media phishing detection techniques continue to evolve, improving adaptability [7].This research contributes by developing an adaptive and scalable phishing detection system to enhance cybersecurity defenses[8].--