Exploring the Relationship Between Feature Attribution Methods and Model Performance

Priscylla Silva, Claudio Silva, Luis Gustavo Nonato
Proceedings of the 2024 AAAI Conference on Artificial Intelligence, PMLR 257:29-38, 2024.

Abstract

Machine learning and deep learning models are pivotal in educational contexts, particularly in predicting student success. Despite their widespread application, a significant gap persists in comprehending the factors influencing these models’ predictions, especially in explainability within education. This work addresses this gap by employing nine distinct explanation methods and conducting a comprehensive analysis to explore the correlation between the agreement among these methods in generating explanations and the predictive model’s performance. Applying Spearman’s correlation, our findings reveal a very strong correlation between the model’s performance and the level of agreement observed among the explanation methods.

Cite this Paper

BibTeX
@InProceedings{pmlr-v257-silva24a, title = {Exploring the Relationship Between Feature Attribution Methods and Model Performance}, author = {Silva, Priscylla and Silva, Claudio and Nonato, Luis Gustavo}, booktitle = {Proceedings of the 2024 AAAI Conference on Artificial Intelligence}, pages = {29--38}, year = {2024}, editor = {Ananda, Muktha and Malick, Debshila Basu and Burstein, Jill and Liu, Lydia T. and Liu, Zitao and Sharpnack, James and Wang, Zichao and Wang, Serena}, volume = {257}, series = {Proceedings of Machine Learning Research}, month = {26--27 Feb}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v257/main/assets/silva24a/silva24a.pdf}, url = {https://proceedings.mlr.press/v257/silva24a.html}, abstract = {Machine learning and deep learning models are pivotal in educational contexts, particularly in predicting student success. Despite their widespread application, a significant gap persists in comprehending the factors influencing these models’ predictions, especially in explainability within education. This work addresses this gap by employing nine distinct explanation methods and conducting a comprehensive analysis to explore the correlation between the agreement among these methods in generating explanations and the predictive model’s performance. Applying Spearman’s correlation, our findings reveal a very strong correlation between the model’s performance and the level of agreement observed among the explanation methods.} }
Endnote
%0 Conference Paper %T Exploring the Relationship Between Feature Attribution Methods and Model Performance %A Priscylla Silva %A Claudio Silva %A Luis Gustavo Nonato %B Proceedings of the 2024 AAAI Conference on Artificial Intelligence %C Proceedings of Machine Learning Research %D 2024 %E Muktha Ananda %E Debshila Basu Malick %E Jill Burstein %E Lydia T. Liu %E Zitao Liu %E James Sharpnack %E Zichao Wang %E Serena Wang %F pmlr-v257-silva24a %I PMLR %P 29--38 %U https://proceedings.mlr.press/v257/silva24a.html %V 257 %X Machine learning and deep learning models are pivotal in educational contexts, particularly in predicting student success. Despite their widespread application, a significant gap persists in comprehending the factors influencing these models’ predictions, especially in explainability within education. This work addresses this gap by employing nine distinct explanation methods and conducting a comprehensive analysis to explore the correlation between the agreement among these methods in generating explanations and the predictive model’s performance. Applying Spearman’s correlation, our findings reveal a very strong correlation between the model’s performance and the level of agreement observed among the explanation methods.
APA
Silva, P., Silva, C. & Nonato, L.G.. (2024). Exploring the Relationship Between Feature Attribution Methods and Model Performance. Proceedings of the 2024 AAAI Conference on Artificial Intelligence, in Proceedings of Machine Learning Research 257:29-38 Available from https://proceedings.mlr.press/v257/silva24a.html.

Related Material