This project focuses on developing machine learning solutions for various use cases within 5G New Radio (NR) networks, specifically under the Open Radio Access Network (O-RAN) framework.

The project covers five primary use cases:

1. Context-Based Dynamic Handover Management for V2X: Enhancing vehicle-to-anything communication by optimizing handover processes to avoid anomalies like short stays and ping-pong effects using regression models.
2. QoE Optimization: Predicting the optimal Mean Opinion Score (MOS) to enhance user experience using classification algorithms such as SVM, K-NN, Random Forest, Gradient Tree Boosting, MLP, SGD, and Decision Tree.
3. QoS Optimization: Improving Quality of Service by managing resource allocation and reducing latency, utilizing Support Vector Regressor, Decision Tree, Linear Regressor, and KNN algorithms.
4. Local Indoor Positioning in RAN: Implementing a robust indoor positioning system using Random Forest, KNN, Gradient Boost, Logistic Regression, and SVM classifiers to improve real-time position tracking.
5. Energy Saving: Predicting cell states (idle or downloading) to optimize energy usage in LTE networks using Random Forest Classifier and k-means clustering.

Technologies used include Python, scikit-learn, TensorFlow, PyTorch, and data visualization tools like Matplotlib and Seaborn. Google Colab was utilized for interactive development and documentation. This project aligns with the industry's vision for more open, flexible, and innovative wireless networks, aiming to enhance service delivery, user experiences, and sustainability in the telecommunications sector.