• Suraj Alam
• Sheetal Shimpikar
• Aman Jankar
• Prathamesh Golhar
• Hemant Koli

The AI-powered phishing detection system leverages Natural Language Processing (NLP) and Machine Learning (ML) to identify and mitigate phishing threats in emails and websites. The system takes raw email content or website URLs as input and processes them using advanced techniques such as tokenization, stopword removal, and TF-IDF vectorization. A Random Forest Classifier and LSTM-based deep learning model are employed to classify inputs as phishing or legitimate. The input data is transformed into numerical features, which are fed into the trained model for real-time prediction. The output is a binary decision (phishing or non-phishing),delivered through a Flask-based REST API. The system is implemented as a browser extension and email plugin, enabling seamless integration into user workflows. The model is trained on a diverse dataset of phishing and legitimate samples, achieving high accuracy and robustness. This project highlights the application of AI in cybersecurity, offering an effective solution to combat phishing attacks and enhance user safety.

Artificial Intelligence, Machine Learning, Phishing Detection, Cybersecurity, Natural Language Processing, Email Classification, Feature Extraction, Data Preprocessing, Supervised Learning, Anomaly Detection, Neural Networks, Model Training, URL Analysis, Spam Filtering, Threat Intelligence, Pattern Recognition, Real-time Detection

"AI-Driven Phishing Detection:A Machine Learning Approach Using NLP.", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 10, page no.d400-d406, October-2025, Available :http://www.jetir.org/papers/JETIR2510357.pdf

"AI-Driven Phishing Detection:A Machine Learning Approach Using NLP.", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 10, page no. ppd400-d406, October-2025, Available at : http://www.jetir.org/papers/JETIR2510357.pdf