Enhanced Slice Prediction in 5G Network using Ensemble-Based Classification

Abstract

The evolution of 5G networks has introduced the need for intelligent network slicing to support diverse service requirements such as enhanced Mobile Broadband (eMBB), massive Machine-Type Communications (mMTC), Ultra-Reliable Low-Latency Communications (URLLC), and vehicular applications (V2X). In this study, preprocessing steps were applied, including data cleaning, reclassification of slice attributes, integration of V2X parameters from literature, and feature encoding to define four slice categories. Following feature selection and a 70/30 Train-Test Split, machine learning models were developed for slice classification. The Random Forest model achieved an accuracy of 0.987, while Gradient Boosting recorded 0.986, demonstrating strong predictive capability and generalization. These findings highlight the effectiveness of ensemble learning techniques for precise 5G slice identification and support the advancement of intelligent network management frameworks.