IoT Devices Empowered for Data -Driven Intelligent Decision-Making with Machine Learning Algorithms

Currently, there is a growing prevalence of cyber-attacks due to the rising demand for IoT devices. There are numerous factors that contribute to the vulnerability of an IoT framework, making it susceptible to attacks by intruders. Implementing IoT-integrated data analytics and machine learning algorithms might greatly enhance productivity, performance, and outcomes in a variety of functional domains. As a result, we have effectively detected cyberattacks using the principles of machine learning. The purpose of this article is to enhance the effectiveness of these fundamental machine learning classifiers by the application of a Stacking Ensemble Method (SEM). By accurately identifying prediction mistakes and making necessary corrections, it can enhance the predictive probability of data instances and contribute to the development of more accurate models with lower error rates. Comparisons have been made between the proposed model and a variety of advanced machine learning techniques, including linear regression (LR), decision trees (DT), Multilayer perceptrons (MLP), and linear discriminant analysis (LDA). In terms of memory, f1-score, accuracy, and precision, the suggested strategy fared better than the others.