College of Engineering, University of Al-Nahrain discussed a master dissertation on the group learning technique for detecting malware for the Internet of Things by postgraduate student, Mrs. Ayat Tariq Salim.

The dissertation aimed at learning how the Internet of Things (IoT) is expanding rapidly in many industries, including wearable technology, sensor systems and local utilities, which led to the growth of Internet devices and thus increased cyber-attacks based on the Internet of Things.

The dissertation reviewed machine learning (ML) and deep learning (DL) approaches, which are the most suitable solutions for detecting threats posed by IoT nodes.

The dissertation demonstrated introducing a detection technique via using five deep learning algorithms LSTM, RNN, CNN, FNN and ensembles learning to detect malware in an IoT environment and using static analysis to scan executable files that extract Op Code features from raw bytes, as opposed to dynamic analysis which requires the executable file to be executed in a secure environment, as well as using of natural language processing (NLP) techniques to create the feature set for more accurate classification and the use of an information acquisition (IG) mechanism as a quick method for feature selection.