Manuel Jesús Jiménez-Navarro
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Manuel Jesús Jiménez-Navarro has a total of 39 co-authors. Here are the top co-authors by number of shared publications:
- María Martínez-Ballesteros
Universidad de Sevilla
16 shared publications - Francisco Martínez-Álvarez
Pablo de Olavide University
10 shared publications - Gualberto Asencio-Cortés
Pablo de Olavide University
6 shared publications - Isabel Sofia Brito
Escola Superior de Tecnologia e Gestão, Instituto Politécnico de Beja
4 shared publications - María Lourdes Linares-Barrera
University of Seville
4 shared publications - Aitor Rodríguez-Dueñas
Universidad de Sevilla
3 shared publications
Latest Publications
- 6 November 2026
- Published by Springer Nature
- p. 78-89
Explainable Foundation Models for Epileptic Seizure Detection in Large-Scale Imbalanced EEG Data
- 6 November 2026
- Published by Springer Nature
- p. 78-89
Explainable Foundation Models for Epileptic Seizure Detection in Large-Scale Imbalanced EEG Data
- Open Access
- 31 August 2026
- Published by Elsevier in SSRN Electronic Journal
Environmental and Spatiotemporal Data Fusion for Particle Number Size Distribution Prediction
Atmospheric particle number size distributions (PNSDs), defined as the particle number concentration relative to particle diameter, are critical descriptors of aerosols since particle size dictates both atmospheric light interaction and deposition within the human respiratory tract. However, forecasting PNSDs is inherently difficult due to the wide range of factors involved, the strong interdependencies across sizes, and the complex physico-chemical processes governing their transformation. This study introduces a multi-level framework for PNSD forecasting by integrating heterogeneous data sources, including harmonized hourly measurements from 36 global monitoring stations and multi-scale land-use descriptors. We systematically compare two target representations: direct multi-output regression of 47-bin histograms and the parametric prediction of nine variables defined by a trimodal lognormal distribution. Different machine learning models were optimized using a temporally consistent validation scheme designed to emulate operational forecasting conditions. These models were then integrated into a stacking-based ensemble to combine and improve the forecasting provided by the individual models. Results demonstrate that anthropogenic land use and tourism-related activity are primary drivers of particle dynamics, while agricultural and natural terrain features critically influence secondary formation and deposition processes. Furthermore, the inclusion of cyclical temporal variables captures the essential impacts of seasonality and diurnal cycles on emissions and boundary layer height. Also, while the parametric representation produces more physically stable distributions, the bin-wise approach achieves lower squared-error metrics. Ultimately, the ensemble framework successfully leverages these multifaceted spatial and temporal predictors, significantly enhancing the robustness and generalization of global PNSD forecasting.
- 31 August 2026
- Published by Elsevier in SSRN Electronic Journal
Open AccessEnvironmental and Spatiotemporal Data Fusion for Particle Number Size Distribution Prediction
Atmospheric particle number size distributions (PNSDs), defined as the particle number concentration relative to particle diameter, are critical descriptors of aerosols since particle size dictates both atmospheric light interaction and deposition within the human respiratory tract. However, forecasting PNSDs is inherently difficult due to the wide range of factors involved, the strong interdependencies across sizes, and the complex physico-chemical processes governing their transformation. This study introduces a multi-level framework for PNSD forecasting by integrating heterogeneous data sources, including harmonized hourly measurements from 36 global monitoring stations and multi-scale land-use descriptors. We systematically compare two target representations: direct multi-output regression of 47-bin histograms and the parametric prediction of nine variables defined by a trimodal lognormal distribution. Different machine learning models were optimized using a temporally consistent validation scheme designed to emulate operational forecasting conditions. These models were then integrated into a stacking-based ensemble to combine and improve the forecasting provided by the individual models. Results demonstrate that anthropogenic land use and tourism-related activity are primary drivers of particle dynamics, while agricultural and natural terrain features critically influence secondary formation and deposition processes. Furthermore, the inclusion of cyclical temporal variables captures the essential impacts of seasonality and diurnal cycles on emissions and boundary layer height. Also, while the parametric representation produces more physically stable distributions, the bin-wise approach achieves lower squared-error metrics. Ultimately, the ensemble framework successfully leverages these multifaceted spatial and temporal predictors, significantly enhancing the robustness and generalization of global PNSD forecasting.
- 16 June 2026
- Published by Springer Nature
- p. 183-195
An Explainable Framework for Football Match Outcome Analysis Using Event-Based Data
- 16 June 2026
- Published by Springer Nature
- p. 183-195
An Explainable Framework for Football Match Outcome Analysis Using Event-Based Data
- 16 June 2026
- Published by Springer Nature
- p. 298-309
Epileptic Seizure Detection in Imbalanced Large-Scale EEG Data: A Machine Learning Approach with Explainability Analysis
- 16 June 2026
- Published by Springer Nature
- p. 298-309
Epileptic Seizure Detection in Imbalanced Large-Scale EEG Data: A Machine Learning Approach with Explainability Analysis
- 16 June 2026
- Published by Springer Nature
- p. 323-334
Explainable Analysis of Fake Instagram Account Detection Using Graph and Behavioral Features
- 16 June 2026
- Published by Springer Nature
- p. 323-334
Explainable Analysis of Fake Instagram Account Detection Using Graph and Behavioral Features
- 2 June 2026
- Published by Springer Nature
- p. 217-226
Exploring Zero-Shot Learning for Time Series Forecasting Using Large Language Models
- 2 June 2026
- Published by Springer Nature
- p. 217-226
Exploring Zero-Shot Learning for Time Series Forecasting Using Large Language Models
- 2 June 2026
- Published by Springer Nature
- p. 227-236
Explainable Graph Neural Networks for Omics-Based Cancer Classification
- 2 June 2026
- Published by Springer Nature
- p. 227-236
Explainable Graph Neural Networks for Omics-Based Cancer Classification
- Open Access
- 17 February 2026
- Published by Elsevier in Sustainable Computing: Informatics and Systems
- Vol. 50
Optimizing power and performance in time series forecasting: A feature selection approach for green computing
- 17 February 2026
- Published by Elsevier in Sustainable Computing: Informatics and Systems
- Vol. 50
Open AccessOptimizing power and performance in time series forecasting: A feature selection approach for green computing
- Open Access
- 15 January 2026
- Published by Elsevier in Environmental Modelling & Software
- Vol. 198
A novel interpretable ozone forecasting approach based on deep learning with masked residual connections
- 15 January 2026
- Published by Elsevier in Environmental Modelling & Software
- Vol. 198
Open AccessA novel interpretable ozone forecasting approach based on deep learning with masked residual connections
- 1 February 2025
- Published by Elsevier in Neurocomputing
- Vol. 626
A real-time machine learning-based methodology for short-term frequency nadir prediction in low-inertia power systems
- 1 February 2025
- Published by Elsevier in Neurocomputing
- Vol. 626
A real-time machine learning-based methodology for short-term frequency nadir prediction in low-inertia power systems