Network anomaly detection using machine learning

The constant increase of network attacks in the digital world creates a significant threat to system security and availability. Anomaly detection plays a crucial role in identifying previously unknown network attacks and potential malicious activities. This thesis focuses on leveraging machine learning techniques for effective network anomaly detection to enhance cybersecurity measures. The study explores various machine learning models to develop a robust and efficient anomaly detection system. At the end of the research, a novel framework based on autoencoders is proposed to further enhance the detection capabilities.