Smart 6G IoT Configuration Enhancement through Hybrid Machine Learning Classification Approaches

The emergence of sixth-generation (6G) wireless communication is set to transform the Internet of Things (IoT) by enabling ultra-reliable, high-capacity, and intelligent connectivity. However, optimizing the configuration of large-scale, heterogeneous IoT networks remains a significant challenge due to multi-parameter dependencies, dynamic traffic patterns, and diverse quality-of-service (QoS) requirements. Traditional optimization methods struggle to capture these complex interlinkages, highlighting the need for machine learning (ML)-driven adaptive management. This work proposes an intelligent ML-based management framework designed to classify 6G IoT configurations into optimized and non-optimized categories. The framework employs multiple supervised learning models—including Decision Tree, Naive Bayes, Support Vector Machine (SVM), Random Forest, and Artificial Neural Network (ANN)—within a structured pipeline of data preprocessing, feature normalization, class balance handling, and performance evaluation. Metrics such as Accuracy, Precision, Recall, and F1-Score are used to assess model effectiveness. Experimental findings reveal that while individual algorithms offer unique benefits—Decision Trees for interpretability, Naive Bayes for computational simplicity, and SVM for generalization—ensemble methods like Random Forest and deep models like ANN consistently outperform. Their strength lies in capturing non-linear and high-dimensional feature interactions, enabling more precise identification of optimal configurations in dynamic 6G IoT scenarios. Overall, the proposed framework enhances classification accuracy, adaptability, and resource efficiency, offering a robust pathway toward intelligent configuration optimization. By integrating multi-algorithm ML strategies, this study contributes to the advancement of selfconfiguring, scalable, and sustainable 6G IoT infrastructures suitable for real-world applications.