Evaluating Machine Learning Algorithms for Intrusion Detection: A Step Toward Securing Real-Time Big Data

Nowadays, Big Data (BD) plays an important role in driving business decisions and strategic planning by extracting valuable insights from massive amounts of data generated every day. However, as BD systems grow in size and complexity, they become more vulnerable to cyberattacks such as data breaches and Distributed Denial of Service (DDoS) attacks. This paper presents a comparison study to evaluate three Machine Learning algorithms: Random Forest, XGBoost, and SGD Classifier, using two datasets (CIC-IDS-2017 and CICIoV2024) to determine which is best suited for future use in real-time BD security.