Comparison of Different Binary Classification Algorithms for Malware Detection

The main task for companies in the ongoing digitalization is to ensure the security of company and customer data. Forecasts show an increase in cyber incidents, while also increasing the number of users who are becoming more aware of such risks. The main element of protection to prevent the consequences of such incidents is the development of advanced algorithms for accurate detection of malicious software and the education of users. A complex solution can be found in the combination of training, technical means, and cyber insurance. In this regard, the current article deals with malware detection from a technical point of view. For this purpose, some binary classification algorithms for malware detection are used and compared with respect to four metrics. The obtained results for all algorithms are encouraging, but one of them shows better performance toward nine malware types. Additional experiments are to be done to prove its applicability in different volume data.