D4.4 Datasets Used to Train AI/ML Models
Year: 2025
Type: article
Abstract: Deliverables D4.1, D4.2 and D4.3 have proposed and evaluated different solutions based on Artificial Intelligence (AI) and Machine Learning (ML) with the objective to improve the energy efficiency of the Radio Access Network (RAN) in Beyond 5G (B5G) cellular networks. This deliverable D4.4 provides the main relevant details of the datasets that have been used to train and test these AI/ML methodologies. First, this document describes how the data is managed in the context of the BeGREEN project, covering aspects of data creation, processing, storage, sharing and security. In addition, a description of the Findability, Accessibility, Interoperability and Reusability (FAIR) principles is provided. These principles aim to provide general guidelines for scientific data management and stewardship. This document also includes a description of a variety of datasets generated/used by the project. The characteristics of the datasets are presented in different tables which cover different aspects such as a brief description of the dataset, the main metrics that are collected and stored in the dataset, the way how the FAIR principles are considered, aspects of data security, archiving and preservation, and ethics and legal aspects.
Authors: Sanchez-Gonzalez, Juan, Perez-Romero, Jordi, Sallent Roig, José Oriol, Umbert Juliana, Anna, Salvat Lozano, Josep Xavier, Ayala-Romero, Jose A., Catalan-Cid, Miguel, Reiss de Fez, David, Armstrong, Joss, Castellanos, German, Sanchez-Gonzalez, Juan
Language: English
Subfield: Computer Networks and Communications
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
OpenAlex ID: W7082638860
FWCI:
Cites:
Cited by:
Related to: