Abstract
Malware (malicious software) are available as software or program that is deliberately developed to cause disturbance in the computation systems such as computers, servers, or networks. Typically, malware aims to drip private data/information, gain unlawful access to system resources (hardware, software, and information/data), deny authorized users from accessing system resources, or even destroy or corrupt system resources. While the level of impact for malware might range from limited to severe, it is essential to detect malware in the system at earlier stages to enable the proper defense to be activated in response to malware. In this paper, we propose a machine learning-based model for identifying malware from goodware by analyzing the API call sequences over the operating system (Windows OS) using support vector machines (SVM). The experimental results show that our model can analyze API call sequences to malware provide identification with an accuracy rate of 98.7% in 13.5 mu s only. Besides, the comparison with other state-of-the-art models exhibits the advantage of our model in terms of detectability at high inferencing rates.