Sajad Homayoun

Industriel postdoc

Sajad Homayoun

Department of Applied Mathematics and Computer Science

Richard Petersens Plads

Building 322 Room 219

2800 Kgs. Lyngby




Machine learning/AI Computer networks Cyber Security

I have a Ph.D. degree in Computer Networks and recently joined the Cybersecurity Section at DTU (Denmark Technical University) as a Postdoc. I am interested in cybersecurity researches, such as malware detection through innovative Machine Learning (ML) techniques. I have some experience in Static, Dynamic, and Hybrid analysis of Ransomware malware samples. During my Ph.D. project, I created a new Testbed for malware analysis called "Systematic Python Auto Testbed." It has the capability of executing malware samples automatically & remotely in a real and controlled environment to help security researchers in the field of malware analysis to analyze samples' behavior when they are executing by the Operating System (OS). I also created a unique dataset of ransomware "System Calls" collected by my developed Testbed which is now publicly available on Github for all researchers around the world. I also worked on detecting Botnets by taking advantage of novel customized features extracted by deep learning techniques, specifically Deep AutoEncoders from inbound and outbound network traffics. So far, I've been able to publish in some authoritative journals, namely IEEE Transactions, Elsevier and Wiley, in the field of cybersecurity. I have gained a lot of teamwork/research experience with international professors in Canada and the United States of America. I am working on SecDNS (Secure Domain Name System) project for detecting malicious DNS activities. We apply Machine Learning and Deep Learning techniques to create a robust detection engine with the capability of identifying adversarial samples.