Due to the increasing complexity of software systems and the organizations, critical infrastructures and key software components are becoming more and more vulnerable to cyberattacks. Moreover, the widespread adoption of Internet technology as part of the business services and Internet of Things (IoT) devices are also adversely affecting the security aspects of many software-intensive systems.
In this project, we will focus on the malware (malicious software) detection and analysis problem of software-intensive systems and provide a holistic solution approach. The state-of-the-art in malware detection mainly relies on designing systems that are resilient to cyberattacks, and on proper machine learning approaches for detecting malware. In practice the design of cybersecure systems, and the analysis using machine learning have been carried out in isolation. In this project we aim to provide an integrated approach that adopts both an engineering design perspective and machine learning perspective. From the design perspective we will investigate and enhance the architecture design approaches for ensuring the prevention of malware in software-intensive systems. From the machine learning perspective, we will investigate the adoption of advanced machine learning algorithms such as deep neural networks and ensemble learning for supporting the detection of malware.