WISCA: A Consensus-Based Approach to Harmonizing Interpretability in Tabular Datasets
Machine Learning and Knowledge Extraction · 2026-04-10 · 8 (4) · 97
DOI: 10.3390/make8040097 ↗At a glance
Journal article in Machine Learning and Knowledge Extraction, published online on 2026-04-10. Written by Antonio Jesús Banegas-Luna and 2 co-authors. Cited 1 times so far (Crossref); 56 references. The full text is freely available (open access, CC BY 4.0).
Abstract (original language)
While predictive accuracy is often prioritized in machine learning (ML) models, interpretability remains essential in scientific and high-stakes domains. However, diverse interpretability algorithms frequently yield conflicting explanations, highlighting the need for consensus to harmonize results. In this study, six ML models were trained on six synthetic datasets with known ground truths, utilizing various model-agnostic interpretability techniques, as well as gradient-based and counterfactual-based explainers. Consensus explanations were generated using established methods and a novel approach: WISCA (Weighted Scaled Consensus Attributions), which integrates class probability and normalized attributions. WISCA consistently aligned with the most reliable individual method, underscoring the value of robust consensus strategies in improving explanation reliability.
Cite
APA
Banegas-Luna, A. J., Pérez-Sánchez, H., & Martínez-Cortés, C. (2026). WISCA: A Consensus-Based Approach to Harmonizing Interpretability in Tabular Datasets. Machine Learning and Knowledge Extraction, 8(4), 97. https://doi.org/10.3390/make8040097
Vancouver
Banegas-Luna AJ, Pérez-Sánchez H, Martínez-Cortés C. WISCA: A Consensus-Based Approach to Harmonizing Interpretability in Tabular Datasets. Machine Learning and Knowledge Extraction. 2026;8(4):97. doi:10.3390/make8040097
BibTeX
@article{BanegasLuna2026,
author = {Banegas-Luna, Antonio Jesús and Pérez-Sánchez, Horacio and Martínez-Cortés, Carlos},
title = {WISCA: A Consensus-Based Approach to Harmonizing Interpretability in Tabular Datasets},
journal = {Machine Learning and Knowledge Extraction},
year = {2026},
volume = {8},
number = {4},
pages = {97},
doi = {10.3390/make8040097}
}
EUVIMED