Innovative Research Award
Gracia Sanchez Carpena — University of Murcia, Spain
| Gracia Sanchez Carpena |
| Affiliation |
University of Murcia |
| Country |
Spain |
| Scopus ID |
7202034595 |
| Documents |
27 |
| Citations |
587 |
| h-index |
13 |
| Subject Area |
Machine Learning |
| Event |
Technology Scientists Awards |
Gracia Sanchez Carpena is a researcher affiliated with the University of Murcia whose scholarly work is situated within machine learning, evolutionary optimization, feature selection, and ensemble learning. Recent publications address computational strategies for high-dimensional data and integrated optimization of predictive models, feature subsets, and aggregation mechanisms. [1] [2]
Abstract
Gracia Sanchez Carpena’s research profile reflects sustained engagement with machine learning and evolutionary computation, particularly feature selection and ensemble optimization. Her recent scholarly contributions address high-dimensional data through permutation-based multi-objective feature selection and the joint evolutionary optimization of heterogeneous ensembles, learner-specific feature subsets, and aggregation weights. These studies investigate methods that balance predictive performance, computational considerations, and model complexity. The reported work spans classification, regression, and ensemble learning settings, demonstrating an applied research orientation toward scalable optimization strategies. Her publication record and citation indicators provide additional evidence of scholarly activity in machine learning. [1] [2]
Keywords
Machine Learning; Feature Selection; Multi-Objective Evolutionary Algorithms; High-Dimensional Data; Ensemble Learning; Evolutionary Optimization; Regression; Predictive Modeling. [1] [2]
Introduction
High-dimensional machine learning requires effective strategies for identifying informative variables while controlling computational cost and model complexity. Recent research associated with Gracia Sanchez Carpena examines multi-objective evolutionary approaches that address these challenges by combining predictive performance with feature reduction. One study evaluates feature subsets through permutation-based degradation of model performance, extending conventional feature-importance concepts toward subset-level analysis. [1]
Research Profile
The research profile centers on machine learning methodologies involving evolutionary search, feature selection, and ensemble construction. The published work demonstrates particular interest in multi-objective optimization, where competing requirements such as predictive accuracy and structural simplicity are considered simultaneously. This orientation is evident in both high-dimensional feature-selection research and recent work on heterogeneous regression ensembles. [1] [2]
Research Contributions
A notable contribution is the development of permutation-based subset evaluation using a multi-objective evolutionary algorithm for high-dimensional feature selection. The approach evaluates groups of attributes rather than isolated variables and simultaneously considers predictive degradation and subset cardinality. Related work extends evolutionary optimization to heterogeneous ensembles by jointly selecting learners, learner-specific features, and aggregation weights within an integrated framework. [1] [2]
Publications
The publication record includes recent peer-reviewed studies addressing evolutionary machine learning and optimization. The article on permutation-based multi-objective evolutionary feature selection was published in Knowledge and Information Systems in 2026 and investigates high-dimensional classification and regression datasets. A second 2026 publication in Algorithms studies simultaneous optimization of heterogeneous ensembles, feature subsets, and aggregation weights for regression. [1] [2]
Research Impact
The reported research has relevance to machine learning applications where feature dimensionality, predictive accuracy, and model complexity must be balanced. The feature-selection study evaluates its methodology across 27 high-dimensional datasets and reports reductions in feature counts alongside predictive-performance improvements. The ensemble study similarly investigates predictive error and structural sparsity through multi-objective optimization. [1] [2]
Award Suitability
Based on the supplied bibliometric indicators and documented publications, Gracia Sanchez Carpena presents a research profile aligned with an innovative research recognition in machine learning. Her recent work demonstrates methodological development rather than solely application, with evolutionary optimization used to address feature-selection and ensemble-design problems. The combination of 27 documents, 587 citations, and an h-index of 13 further contextualizes the supplied scholarly record. [1] [2]
Conclusion
Gracia Sanchez Carpena’s documented research demonstrates a coherent focus on machine learning, evolutionary optimization, feature selection, and ensemble learning. Her recent publications contribute methods for addressing high-dimensional data and jointly optimizing predictive architectures. The combination of methodological research, peer-reviewed publication, and the supplied bibliometric record provides a reasonable scholarly basis for consideration for an Innovative Research Award. [1] [2]
External Links
References
- Espinosa, R., Sánchez, G., Palma, J., & Jiménez, F. (2026). Permutation-based multi-objective evolutionary feature selection for high-dimensional data. Knowledge and Information Systems, 68, 116.
https://doi.org/10.1007/s10115-026-02734-0
- Galván, J., Sánchez, G., & Jiménez, F. (2026). Simultaneous multi-objective evolutionary optimization of heterogeneous ensembles, learner-specific feature subsets, and aggregation weights. Algorithms, 19(8), 681.
https://doi.org/10.3390/a19080681
