AI powered Vishing and Smishing detection using machine learning and deep learning

The increasing threat of Vishing and Smishing attacks necessitates the development of robust detection systems to safeguard users from fraudulent activities. This study explores an AI-powered approach for detecting Vishing (voice-based phishing) and Smishing (SMS-based phishing) using machine learning and deep learning algorithms. The proposed system utilizes a combination of traditional machine learning models and advanced deep learning architectures to classify SMS text and voice data as either SPAM or HAM. The machine learning algorithms implemented include Decision Tree and Random Forest, while the deep learning models employ RoBERTa integrated with BiLSTM, MobileNet, and ResNet for enhanced accuracy and efficiency. The system's frontend is developed using HTML, CSS, and JavaScript, ensuring a user-friendly interface, while the backend is powered by Python with the Flask framework. The aim of this work is to create a scalable and efficient solution for detecting phishing attacks in real-time, offering potential applications in cybersecurity for mobile communication systems and enhancing user protection from fraudulent activities.

Keywords: Vishing Detection, Smishing Detection, Machine Learning, Deep Learning, Decision Tree, Random Forest, RoBERTa, BiLSTM, MobileNet, ResNet, Flask, SMS Classification, Voice-based Phishing, Cybersecurity, Phishing Detection.