María José Ramírez-Quintana
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
María José Ramírez-Quintana has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
- José Hernández-Orallo
VRAIN, Universitat Politècnica de València, València, Spain
28 shared publications - Cèsar Ferri
Universitat Politècnica de València
23 shared publications - Fernando Martínez-Plumed
Universitat Politècnica de València
9 shared publications V. Estruch
DSIC, Universitat Politècnica de València, Valencia, Spain
7 shared publicationsC. Ferri
5 shared publicationsAntonio Bella
4 shared publications
Latest Publications
- 25 August 2026
- Published by Springer Nature
- p. 213-228
Targeted Robustness for Regression via Top-Feature Noise
- 25 August 2026
- Published by Springer Nature
- p. 213-228
Targeted Robustness for Regression via Top-Feature Noise
- Open Access
- 23 February 2026
- Published by Springer Nature in International Journal of Data Science and Analytics
- Vol. 22 (1) , 71
Robustness under noise: assessing the impact of perturbed key attributes on machine learning models
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.
- 23 February 2026
- Published by Springer Nature in International Journal of Data Science and Analytics
- Vol. 22 (1) , 71
Open AccessRobustness under noise: assessing the impact of perturbed key attributes on machine learning models
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.
- 21 November 2025
- Published by Springer Nature
- p. 86-100
Towards an Automatic Student Supervisor System Using Declarative Languages
- 21 November 2025
- Published by Springer Nature
- p. 86-100
Towards an Automatic Student Supervisor System Using Declarative Languages
- 14 November 2024
- Published by Springer Nature
- p. 96-108
Noise Tolerance and Robustness Ranking in Machine Learning Models
- 14 November 2024
- Published by Springer Nature
- p. 96-108
Noise Tolerance and Robustness Ranking in Machine Learning Models
- Open Access
- 19 August 2024
- Published by Springer Nature in Complex & Intelligent Systems
- Vol. 10 (6) , 8287-8317
A general supply-inspect cost framework to regulate the reliability-usability trade-offs for few-shot inference
- 19 August 2024
- Published by Springer Nature in Complex & Intelligent Systems
- Vol. 10 (6) , 8287-8317
Open AccessA general supply-inspect cost framework to regulate the reliability-usability trade-offs for few-shot inference
- Open Access
- 8 August 2024
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 5 (3) , 035040
Unveiling the robustness of machine learning families
The evaluation of machine learning systems has typically been limited to performance measures on clean and curated datasets, which may not accurately reflect their robustness in real-world situations where data distribution can vary from learning to deployment, and where truthfully predict some instances could be more difficult than others. Therefore, a key aspect in understanding robustness is instance difficulty, which refers to the level of unexpectedness of system failure on a specific instance. We present a framework that evaluates the robustness of different ML models using item response theory-based estimates of instance difficulty for supervised tasks. This framework evaluates performance deviations by applying perturbation methods that simulate noise and variability in deployment conditions. Our findings result in the development of a comprehensive taxonomy of ML techniques, based on both the robustness of the models and the difficulty of the instances, providing a deeper understanding of the strengths and limitations of specific families of ML models. This study is a significant step towards exposing vulnerabilities of particular families of ML models.
- 8 August 2024
- Published by IOP Publishing in Machine Learning: Science and Technology
- Vol. 5 (3) , 035040
Open AccessUnveiling the robustness of machine learning families
The evaluation of machine learning systems has typically been limited to performance measures on clean and curated datasets, which may not accurately reflect their robustness in real-world situations where data distribution can vary from learning to deployment, and where truthfully predict some instances could be more difficult than others. Therefore, a key aspect in understanding robustness is instance difficulty, which refers to the level of unexpectedness of system failure on a specific instance. We present a framework that evaluates the robustness of different ML models using item response theory-based estimates of instance difficulty for supervised tasks. This framework evaluates performance deviations by applying perturbation methods that simulate noise and variability in deployment conditions. Our findings result in the development of a comprehensive taxonomy of ML techniques, based on both the robustness of the models and the difficulty of the instances, providing a deeper understanding of the strengths and limitations of specific families of ML models. This study is a significant step towards exposing vulnerabilities of particular families of ML models.
- Open Access
- 1 December 2022
- Published by Springer Nature in Machine Learning
- Vol. 112 (6) , 2053-2082
Can language models automate data wrangling?
The automation of data science and other data manipulation processes depend on the integration and formatting of ‘messy’ data. Data wrangling is an umbrella term for these tedious and time-consuming tasks. Tasks such as transforming dates, units or names expressed in different formats have been challenging for machine learning because (1) users expect to solve them with short cues or few examples, and (2) the problems depend heavily on domain knowledge. Interestingly, large language models today (1) can infer from very few examples or even a short clue in natural language, and (2) can integrate vast amounts of domain knowledge. It is then an important research question to analyse whether language models are a promising approach for data wrangling, especially as their capabilities continue growing. In this paper we apply different variants of the language model Generative Pre-trained Transformer (GPT) to five batteries covering a wide range of data wrangling problems. We compare the effect of prompts and few-shot regimes on their results and how they compare with specialised data wrangling systems and other tools. Our major finding is that they appear as a powerful tool for a wide range of data wrangling tasks. We provide some guidelines about how they can be integrated into data processing pipelines, provided the users can take advantage of their flexibility and the diversity of tasks to be addressed. However, reliability is still an important issue to overcome.
- 1 December 2022
- Published by Springer Nature in Machine Learning
- Vol. 112 (6) , 2053-2082
Open AccessCan language models automate data wrangling?
The automation of data science and other data manipulation processes depend on the integration and formatting of ‘messy’ data. Data wrangling is an umbrella term for these tedious and time-consuming tasks. Tasks such as transforming dates, units or names expressed in different formats have been challenging for machine learning because (1) users expect to solve them with short cues or few examples, and (2) the problems depend heavily on domain knowledge. Interestingly, large language models today (1) can infer from very few examples or even a short clue in natural language, and (2) can integrate vast amounts of domain knowledge. It is then an important research question to analyse whether language models are a promising approach for data wrangling, especially as their capabilities continue growing. In this paper we apply different variants of the language model Generative Pre-trained Transformer (GPT) to five batteries covering a wide range of data wrangling problems. We compare the effect of prompts and few-shot regimes on their results and how they compare with specialised data wrangling systems and other tools. Our major finding is that they appear as a powerful tool for a wide range of data wrangling tasks. We provide some guidelines about how they can be integrated into data processing pipelines, provided the users can take advantage of their flexibility and the diversity of tasks to be addressed. However, reliability is still an important issue to overcome.
- 27 September 2018
- Published by Springer Nature
- p. 118-127
Adapting Hierarchical Multiclass Classification to Changes in the Target Concept
- 27 September 2018
- Published by Springer Nature
- p. 118-127
Adapting Hierarchical Multiclass Classification to Changes in the Target Concept
- 30 November 2015
- Published by Springer Nature in Data Mining and Knowledge Discovery
- Vol. 30 (4) , 848-890
Binarised regression tasks: methods and evaluation metrics
- 30 November 2015
- Published by Springer Nature in Data Mining and Knowledge Discovery
- Vol. 30 (4) , 848-890
Binarised regression tasks: methods and evaluation metrics
- 14 November 2015
- Published by Springer Nature
- p. 234-244
Pentaho + R: An Integral View for Multidimensional Prediction Models
- 14 November 2015
- Published by Springer Nature
- p. 234-244
Pentaho + R: An Integral View for Multidimensional Prediction Models