Machine learning is transforming how organisations detect cyber threats — but most security professionals lack hands-on experience building and deploying ML models. This course closes that gap, taking you from core ML concepts to practical, applied threat detection on real cybersecurity datasets.
You'll start with the foundations: model training, learning types, and measuring model accuracy. You'll also learn how attackers exploit ML systems through inference, poisoning, and adversarial input — giving you a security-first perspective from the start. From there, you'll move into hands-on application. You'll load, preprocess, train, and test classification and regression models to identify malware, detect fraud, and analyse network traffic. You'll apply artificial neural networks to classify malware binaries and behavioural patterns. In the final section, you'll build network anomaly detection models using K-Nearest Neighbors (KNN) and One-Class SVM to identify outlier traffic and distinguish normal behaviour from potential attacks. Designed for security analysts, SOC teams, IT engineers, and data scientists entering cybersecurity. Basic cybersecurity knowledge is recommended. Job skills taught: Machine Learning for Cybersecurity · Threat Detection · Malware Analysis · Network Anomaly Detection · ML Model Training and Evaluation · Classification and Regression Modelling · Fraud Detection · Artificial Neural Networks · Network Traffic Analysis Features Coursera Coach, Dialogues and Role Plays - a smarter way to learn with interactive, real-time conversations that help you test your knowledge, challenge assumptions, and deepen your understanding as you progress through the course.