Systematic Literature Review of Artificial Intelligence in Malware Detection
Giancarlo Cassanova, Anderies, Alexander Agung Santoso Gunawan · 2022
Artificial Intelligence is a subfield of computer science. It has been applied in a lot of other fields including antivirus and it was proven that several AI models are capable of working on malware detection tasks. With a lot of antivirus software receiving negative reviews, we decided to learn how an AI model works when applied to solve problems related to malware detection so we can someday be able to make an antivirus by ourselves that is based on AI technology. We have conducted a systematic literature review on several papers which we found on various scholarly databases. We constructed three research questions and applied several selection criteria to find the papers we need. We found several models with varying purposes. Then, we took any data necessary to answer our research questions. It was found that a random forest method was able to achieve high accuracy when tested on samples almost half the size of the one it was trained on. Our study also found that there are techniques, such as k-fold and transfer learning that can be applied to reduce the data needed for training an AI model. Lastly, we also found that most of the models we reviewed were constructed using the deep learning method. With this, our future research will be directed toward the development of a better malware detection model by using the knowledge we have obtained from this research.