Gilbert Saporta
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
Gilbert Saporta has a total of 26 co-authors. Here are the top co-authors by number of shared publications:
- Genane Youness
Laboratoire LINEACT CESI, IDFC, Nanterre, France
3 shared publications - Ndèye Niang
CEDRIC lab, Conservatoire National des Arts et Métiers, Paris, France
3 shared publications Huiwen Wang
School of Economics and Management and Beijing Advanced Innovation Center for Big Data and Brain Computing, Beihang University, Beijing, China
2 shared publications- Mouhamadou Lamine Ndao
Recherche et Innovation LINEACT CESI, IDFC Nanterre France
2 shared publications - Ruiping Liu
School of Applied Science, Beijing Information Science and Technology University, Beijing, China
2 shared publications - Agostino Di Ciaccio
University of Rome La Sapienza
1 shared publication
Latest Publications
- 18 August 2025
- Published by Springer Nature
- p. 270-286
A Divisive Unsupervised Feature Selection Approach for Explainable Remaining Useful Life Prediction
- 18 August 2025
- Published by Springer Nature
- p. 270-286
A Divisive Unsupervised Feature Selection Approach for Explainable Remaining Useful Life Prediction
- Open Access
- 26 June 2025
- Published by Springer Nature in Advances in Data Analysis and Classification
- p. 1-4
Correction to: Sparse correspondence analysis for large contingency tables
- 26 June 2025
- Published by Springer Nature in Advances in Data Analysis and Classification
- p. 1-4
Open AccessCorrection to: Sparse correspondence analysis for large contingency tables
- Open Access
- 13 May 2024
- Published by University of Florida George A Smathers Libraries in The International FLAIRS Conference Proceedings
- Vol. 37
Enhancing Explainability in Predictive Maintenance : Investigating the Impact of Data Preprocessing Techniques on XAI Effectiveness
In predictive maintenance, the complexity of the data often requires the use of Deep Learning models. These models, called “black boxes”, have proved their worth in predicting the Remaining Useful Life (RUL) of industrial machines. However, the inherent opacity of these models requires the incorporation of post-hoc explanation methods to enhance transparency. The quality of the explanations provided is then assessed using so-called evaluation metrics. Modeling is a whole process that includes an important data preprocessing phase, with parameter selection such as time window, smoothing parameter, or rectified RUL when dealing with multivariate time series dataset. We propose to analyze the impact of these preprocessing methods on the quality of explanations provided by the local post-hoc models LIME, KernelSHAP, and L2X, utilizing six evaluation metrics: stability, consistency, congruence, selectivity, completeness, and acumen. This analysis will be based on NASA's Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) dataset with the LSTM model. Our findings reveal that the choice of specific pre-processing parameters can significantly improve predictive performance. Furthermore, the quality of explanations depends on the selection of explicability methods. In addition, a factorial analysis of the evaluation metrics reveals that they do not all point in the same direction.Indeed, understanding the nuanced relationships between evaluation metrics is essential for a comprehensive and accurate assessment of explainability methods.
- 13 May 2024
- Published by University of Florida George A Smathers Libraries in The International FLAIRS Conference Proceedings
- Vol. 37
Open AccessEnhancing Explainability in Predictive Maintenance : Investigating the Impact of Data Preprocessing Techniques on XAI Effectiveness
In predictive maintenance, the complexity of the data often requires the use of Deep Learning models. These models, called “black boxes”, have proved their worth in predicting the Remaining Useful Life (RUL) of industrial machines. However, the inherent opacity of these models requires the incorporation of post-hoc explanation methods to enhance transparency. The quality of the explanations provided is then assessed using so-called evaluation metrics. Modeling is a whole process that includes an important data preprocessing phase, with parameter selection such as time window, smoothing parameter, or rectified RUL when dealing with multivariate time series dataset. We propose to analyze the impact of these preprocessing methods on the quality of explanations provided by the local post-hoc models LIME, KernelSHAP, and L2X, utilizing six evaluation metrics: stability, consistency, congruence, selectivity, completeness, and acumen. This analysis will be based on NASA's Commercial Modular Aero-Propulsion System Simulation (C-MAPSS) dataset with the LSTM model. Our findings reveal that the choice of specific pre-processing parameters can significantly improve predictive performance. Furthermore, the quality of explanations depends on the selection of explicability methods. In addition, a factorial analysis of the evaluation metrics reveals that they do not all point in the same direction.Indeed, understanding the nuanced relationships between evaluation metrics is essential for a comprehensive and accurate assessment of explainability methods.
- 2 January 2023
- Published by Springer Nature in Advances in Data Analysis and Classification
- Vol. 17 (4) , 1037-1056
Sparse correspondence analysis for large contingency tables
- 2 January 2023
- Published by Springer Nature in Advances in Data Analysis and Classification
- Vol. 17 (4) , 1037-1056
Sparse correspondence analysis for large contingency tables
- 1 January 2023
- Published by Springer Nature
- p. 131-154
Old and New Perspectives on Optimal Scaling
- 1 May 2020
- Published by Elsevier in Applied Radiation and Isotopes
- Vol. 159, 109092
A Bayesian framework to update scaling factors for radioactive waste characterization
- 1 May 2020
- Published by Elsevier in Applied Radiation and Isotopes
- Vol. 159, 109092
A Bayesian framework to update scaling factors for radioactive waste characterization
- 19 March 2018
- Published by Springer Nature in International Journal of Data Science and Analytics
- Vol. 6 (3) , 201-204
Training data scientists: a few challenges
- 19 March 2018
- Published by Springer Nature in International Journal of Data Science and Analytics
- Vol. 6 (3) , 201-204
Training data scientists: a few challenges
- 1 June 2017
- Published by Elsevier in Computational Statistics & Data Analysis
- Vol. 110, 103-114
Variable selection for multiply-imputed data with penalized generalized estimating equations
- 1 June 2017
- Published by Elsevier in Computational Statistics & Data Analysis
- Vol. 110, 103-114
Variable selection for multiply-imputed data with penalized generalized estimating equations
- 1 January 2016
- Published by PERSEE Program in Statistique et Enseignement
- Vol. 7 (1) , 111-119
Le certificat "Analyste de données massives" du Conservatoire National des Arts et Métiers
Dans cet article nous présentons le certificat de spécialisation «Analyste de données massives » du Conservatoire National des Arts et Métiers (CNAM). Après une présentation du CNAM et ses missions, le contenu pédagogique de la formation, le public auquel elle s’adresse et ses spécificités qui la distinguent des enseignements classiques de statistique et d’informatique sont présentés. Enfin, deux brefs témoignages sur l’apport de cette formation aux professionnels des entreprises sont joints.
- 1 January 2016
- Published by PERSEE Program in Statistique et Enseignement
- Vol. 7 (1) , 111-119
Le certificat "Analyste de données massives" du Conservatoire National des Arts et Métiers
Dans cet article nous présentons le certificat de spécialisation «Analyste de données massives » du Conservatoire National des Arts et Métiers (CNAM). Après une présentation du CNAM et ses missions, le contenu pédagogique de la formation, le public auquel elle s’adresse et ses spécificités qui la distinguent des enseignements classiques de statistique et d’informatique sont présentés. Enfin, deux brefs témoignages sur l’apport de cette formation aux professionnels des entreprises sont joints.