AI/ML Performance Evaluation (multiple projects possible)

Reema Abaoud


Supervised by Philipp Reinecke; Moderated by Matthew J W Morgan

Android malware growth has been increasing dramatically along with increasing the diversity and complicity of their developing techniques. One of the main technique that is used to detect malware is machine learning methods. This is why there is a crucial demand to evaluate machine learning algorithms performance in detecting malware. Hence,the purpose of this work is to evaluate and examine the machine learning algorithms performance in classifying malware and benign files. Three different classification methods were evaluated in this research Random Forest, Support Vector Machine, and K-Nearest Neighbors. In addition to that several default and adjusted hyperparameters values of the three classification algorithms were evaluated. Finally, a relation between accuracy and run time was examined. This work presents recommended methods for machine learning based malware classification and detection, as well as the guidelines for its implementation. Moreover, the study performed can be useful as a base for further research in the field of evaluating the performance of machine learning algorithms and their hyperparameters in detecting malware.

Initial Plan (03/02/2020) [Zip Archive]

Final Report (29/05/2020) [Zip Archive]

Publication Form