Improved Malware Detection for IoT Devices Using Random Forest Algorithm Comparing with Decision Tree Algorithm

The primary goal of the current work is to carry out Malware Detection for IoT devices by comparing the performance of different classifiers. Malware is software that causes harm to our systems or network. Random Forest Algorithm (RFA) and Decision Tree Algorithm (DTA) are two types of algorithms that can be considered. The methods were built and evaluated on a 19612 record dataset. With 10 example sizes, emphasis was performed on each gathering to accomplish better precision. The error rate power was utilized as 80% to perform G-power testing. The experiment’s findings revealed that the Random Forest Algorithm had a mean accuracy of 99.0320 and the Decision tree had a mean accuracy of 98.5140 for malware detection. Using independent sample t-tests, the statistically significant variance in accurateness between the two models was obtained as p = 0.030. This research aims to apply a novel technique to present Machine Learning Classifiers for malware detection. When comparing the Random Forest Algorithm to the Decision Tree Algorithm, the findings signify that the RFA outperforms the DTA.