Malware Detection Using Genetic Cascaded Support Vector Machine Classifier in Internet of Things
Shashi Kant Gupta, Birajashis Pattnaik, Vineet Agrawal, Raja Sarath Kumar Boddu, Archana Srivastava, Bramah Hazela
Abstract
The Internet of Things (IoT) is a network of computing devices that can transmit and obtain data across a network without human intervention. In the last couple of decades, software and communication technology have advanced tremendously, resulting in a considerable increase in IoT devices. The rapid expansion has raised security and privacy issues. Threats and malware attacks on IoT devices have increased dramatically recently. Hence, in this paper, we proposed a novel malware detection framework based on machine learning in IoT using a Genetic Cascaded Support Vector Machine (GC-SVM) classifier. We introduce the Chaotic Binary Coded Cuckoo Search Optimization Algorithm (CBC-CSOA) for optimizing the detection process. The performance of the proposed method is evaluated and compared with various conventional methodologies. The proposed method produced accurate outputs this approach may be used to forecast and identify malware in IoT-based systems, according to the study.