Khan, SohialsohialAlmarhbi, EhdaaSimbawah, Ghazaloudah, Walaa2025-03-112025-03-112024-12-05http://hdl.handle.net/20.500.14131/2048Abstract: This project presents an innovative anomaly-based Intrusion Detection System (IDS) for detecting Advanced Persistent Threats (APTs) using artificial intelligence. It integrates three models—Support Vector Machines (SVM), Multi- Layer Perceptrons (MLP), and Long Short-Term Memory (LSTM)—through a voting mechanism to improve accuracy and reduce false positives. The system uses a robust dataset, DAPT 2020, and applies data preprocessing techniques such as feature extraction and normalization to enhance model performance. Training and optimization of these models were conducted to ensure high precision, recall, and F1-scores. The IDS is deployed in a scalable cloud environment for real-time network traffic monitoring, paired with a user-friendly interface to facilitate usability. This approach supports proactive detection of sophisticated cyber threats, aligning with Vision 2030’s digital transformation goals and SDG 9 by enhancing infrastructure security and fostering innovation.enIntrusion Detection System (IDS), Advanced Persistent Threats (APTs), Machine Learning, Deep Learning, SVM, MLP, LSTM, Cybersecurity, Anomaly Detection, Voting Mechanism, Real-Time Monitoring.Capstone