AWARD

Methodological contribution

IDOIA OCHOA

Contribution: GENNIUS: an ultrafast drug-target interaction inference method based on graph neural networks.

Author(s): Uxía Veleiro, Jesús de la source, Guillermo Serrano, Marija Pizurica, Mikel Casals, Antonio Pineda-Lucena, Silve Vicent, Idoia Ochoa, Olivier Gevaert, Mikel Hernáez

Brief description of the qualities motivating the award: Major methodological contribution to a data science problem, offering an efficient solution with significant social and economic implications.

AWARD SHARED

AWARD

Applied contribution

IGNACIO RODRÍGUEZ CARREÑO

Contribution: AI perceives like a local: predicting citizen deprivation perception using satellite imagery.

Author(s): Angela Abascal, Sabine Vanhuysse, Taïs Grippa, Ignacio Rodriguez-Carreño, Stefanos Georganos, Jiong Wang, Monika Kuffer, Pablo Martinez-Diez, Mar Santamaria-Varas & Eleonore Wolff

Brief description of the qualities motivating the award: Deep understanding of technological advancements in the field, contributing to innovation in predicting social phenomena through urban image analysis.

AWARD

Applied contribution

RUBÉN ARMAÑANZAS ARNEDILLO

Contribution: Identification of concussion subtypes based on intrinsic brain activity. JAMA Netw Open

Author(s): Armañanzas R, Liang B, Kanakia S, Bazarian JJ, Prichep LS.

Brief description of the qualities motivating the award: Thoroug statistical analysis of EEG data aimed at classifying potential subtypes of concussions, a contribution that could aid future diagnosis in brain activity detection.

AWARD

Applied contribution

LUIS A. GIL ALAÑA

Contribution: Compendium of 4 contributions:

1. Persistence in UK Historical Data on Life Expectancy.
2. Long Memory Cointegration in the Analysis of Maximum, Minimum and Range Temperatures in Africa: Implications for Climate Change.
3. Measuring Persistence in the US Equity Gender Diversity Index.
4. Energy prices in Europe. Evidence of persistence across market.

Author(s): Guglielmo Maria Caporale, Juan Infante, Marta del Rio, Olaoluwa S. Yaya, Oluwaseun A. Adesina, Hammed A. Olayinka, Oluseyi E. Ogunsola, Miguel A. Martin-Valmayor, Luis A. Gil-Alana

Brief description of the qualities motivating the award: This is a compilation of four articles that address fractional integration and cointegration along with their empirical implementations. The first of these articles examines the historical evolution of life expectancy in the United Kingdom. The second article focuses on climate change in Africa and once again uses techniques of fractional integration and cointegration. The third article investigates gender diversity equity in the United States. Finally, the fourth article centers on the study of energy prices in Europe.

AWARD

Methodological contribution

ANGEL RUBIO

Contribution: Precision oncology: a review to assess interpretability in several explainable methods.

Author(s): Marian Gimeno, Katyna Sada del Real, Angel Rubio.

Brief description of the qualities motivating the award: In this paper, a novel algorithm called "Optimal Decision Trees" was introduced, whose goal is precisely to solve the PM problem. It is based on trees. In each bifurcation of the tree, the algorithm identifies the best marker (discrete or continuous) and the optimal drugs for the patients in each branch of the tree. Since the algorithm is very fast, it can be transformed into a random optimizing forest or an extreme gradient boost method. Another advantage is the simplicity of the method: the trees are self-explanatory and easy to understand.

AWARD

Applied contribution with an impact in the social sphere, innovation or knowledge transfer

MIGUEL VALENCIA USTARROZ

Contribution: An interactive framework for the detection of ictal and interictal activities: cross-species and stand-alone implementation.

Author(s): Guillermo M. Besné, Alejandro Horrillo-Maysonnial, María Jesús Nicolás, Ferran Capell-Pascual, Elena Urrestarazu, Julio Artieda, Miguel Valencia

Brief description of the qualities motivating the award: This work uses canned Matlab ML functions to implement 6 different ML methods for the detection of events from EEG data. The authors analyze the signals both in the time and frequency domain. It is an interdisciplinary work, which gives it an added value. They build customized ML models for the detection of ictal and interictal activities for the automatic annotation of epileptic traits based on electrophysiological recordings. It appears to require considerable input from the user, although the claim is made that the interactive app is simpler than available methods.

AWARD

Methodological contribution

FRANCISCO PLANES PEDREÑO

Contribution: BOSO: A novel feature selection algorithm for linear regression with high-dimensional data.

Author(s): Luis V. Valcárcel, Edurne San José-Enériz, Xabier Cendoya, Ángel Rubio, Xabier Agirre, Felipe Prósper, Francisco J. Planes

Brief description of the qualities motivating the award: The paper presents a new method for feature selection in high dimensional regression, and empirically demonstrates based on synthetic datasets how it works better than Lasso, forward selection, best subsets, and the relaxed lasso methods. It performs very well for data that can be modeled by a linear regression model. The broad applicability makes this an appealing work.