Malware detection technique in IoT with Data mining methods

Abstract

Malware plays a major role as a threat to the security of computer systems. As the Internet of things and its systems of connectivity increase all around the world, it has led to an astronomical increase of malware that target these IoT devices. From DDoS attacks to crytomining malware, companies and industries nowadays encounter problems through malware attack that were not existent a few years ago or have evolved to the new environment of IoT, taking advantage of its vulnerabilities such as the inadequate security monitoring and protection systems. This thesis research surveys the types of attack that are common to IoT technology, current detection techniques, learning techniques and machine learning algorithms that are popularly used for malware detection. This paper then further continues to use a dataset of extracted network traffic features from benign and malicious trace data. With the aid of tools such as Rapid Miner and the use of algorithms such as Artificial Neural Network, statistical analysis of data is evaluated with clear evidence of anomaly detection and a proposed model for network anomaly detection with a low false positive rate and high detection accuracy is presented.