Robustness under noise: assessing the impact of perturbed key attributes on machine learning models
Cristina Padró-Ferragut, Fernando Martínez-Plumed, María José Ramírez-Quintana
International Journal of Data Science and Analytics, Vol. 22, Issue 1 (2026)
10.1007/s41060-025-01006-4
Abstract
Abstract Machine learning models often struggle when confronted with the complexities of real-world data, such as noise and incomplete information, which pose significant challenges for practical applications. This paper presents a systematic framework for assessing the robustness of machine learning models against these adversities. In particular, we investigate how noise affecting the most critical attribute of a dataset impacts on model predictions. Our methodology uses a variety of datasets, a wide range of machine learning models, and a customizable noise induction function, allowing for comprehensive evaluations tailored to specific application needs. Through an extensive experimental analysis, we identify the vulnerabilities of models when their critical attributes are perturbed. In addition, by applying hierarchical clustering, we categorise machine learning families based on their robustness, establishing a taxonomy that highlights the susceptibility of different models to noise. Our findings provide important insights into the interplay between model performance and noise levels and improve the understanding of model behaviour under adverse conditions.
Topics
Field: Computer Science · Subfield: Artificial Intelligence
Keywords
Robustness (evolution),Key (lock),Noise (video),Noisy data,Variety (cybernetics)
All Available Versions
- PDF Landing page — International Journal of Data Science and Analytics publishedVersion cc-by
- Landing page — RiuNet (Universitat Politècnica de València) submittedVersion cc-by
Citations by Year
| 2026 |
|---|
| 1 |
Continue on JournalsHub
Put what you just read into practice with these free tools.