Classification of IoT Malware based on Convolutional Neural Network

Abstract

In this paper, we propose an algorithm for the malware classification problem in the IoT domain. Application executions are represented by sequences of consecutive API calls. The time series of data are analyzed and filtered based on information gains, which reduce sequence lengths while keeping important information. We use convolutional neural networks to classify various types of malwares. The experimental results on real world IoT malware samples show that this approach has a faster and more accurate classification rate than the recurrent neural networks and some other machine learning classification algorithms.