Existing solutions are unable to detect unknown APT attacks and are limited by the crumbling of the heterogeneous threat intelligence and slow detection of threats and forensic analysis. Thus, to overcome these limitations, this research proposal aims to design and develop an efficient, fast and scalable
APT detection and Forensic analysis tool using the stream processing approach and advanced deep federated learning architecture that is easy to deploy and user friendly.