Malware Analysis and Classification Using Grid Search Optimization

Authors:

Abstract
Internet of Things (IoT) technology is experiencing rapid growth, but security and privacy remain a major concern. This is because IoT devices are vulnerable to being targets of malware attacks. On the other hand, detecting malware in IoT networks is challenging. Previous research has introduced several methods for IoT malware detection using several Machine Learning algorithms. However, only a few studies discussed parameter optimization in machine learning models. Thus, this research aims to develop a malware detection model using hyperparameter optimization with grid search in several Machine Learning algorithms. Several Machine Learning algorithms are utilized, such as Decision Tree (DT), Random Forest (RF), Support Vector Machine (SVM), and k Nearest Neighbors (k-NN). This research aims to improve IoT network security by developing a model to mitigate and detect the presence of malware attacks. The experiment using the IoT23 dataset shows a good result with the RF model. RF obtained the best result, achieving accuracy of 99.09%, precision of 98.54%, recall of 99.05%, and an F1 score of 98.79%.