Performance evaluation of Machine Learning and Deep Learning models for 5G resource allocation

The deployment of 5G networks introduces challenges in resource allocation and maintaining Quality of Service (QoS). This study aims to develop and benchmark machine learning (ML) and deep learning (DL) models for predicting high-resource demands using real-world KPIs such as signal strength, latency, and bandwidth. By applying rigorous data pre-processing, we compare models including Logistic Regression, Random Forest, XGBoost, and GRU with Attention. A hybrid XGBoost-GRU-Attention model achieves 99.50% accuracy, demonstrating a superior ability to model temporal and feature interactions. These findings underscore the potential of AI-driven techniques for intelligent and real-time 5G optimization.