Coalitional Strategies for Efficient Individual Prediction Explanation

Gabriel Ferrettini (), Elodie Escriva (), Julien Aligon (), Jean-Baptiste Excoffier () and Chantal Soulé-Dupuy ()
Additional contact information
Gabriel Ferrettini: Université de Toulouse-Capitole, IRIT, (CNRS/UMR 5505)
Elodie Escriva: Kaduceo
Julien Aligon: Université de Toulouse-Capitole, IRIT, (CNRS/UMR 5505)
Jean-Baptiste Excoffier: Kaduceo
Chantal Soulé-Dupuy: Université de Toulouse-Capitole, IRIT, (CNRS/UMR 5505)

Information Systems Frontiers, 2022, vol. 24, issue 1, No 4, 49-75

Abstract: Abstract As Machine Learning (ML) is now widely applied in many domains, in both research and industry, an understanding of what is happening inside the black box is becoming a growing demand, especially by non-experts of these models. Several approaches had thus been developed to provide clear insights of a model prediction for a particular observation but at the cost of long computation time or restrictive hypothesis that does not fully take into account interaction between attributes. This paper provides methods based on the detection of relevant groups of attributes -named coalitions- influencing a prediction and compares them with the literature. Our results show that these coalitional methods are more efficient than existing ones such as SHapley Additive exPlanation (SHAP). Computation time is shortened while preserving an acceptable accuracy of individual prediction explanations. Therefore, this enables wider practical use of explanation methods to increase trust between developed ML models, end-users, and whoever impacted by any decision where these models played a role.

Keywords: Data analysis; Machine learning; Interpretability; Explainable Artificial Intelligence (XAI); Prediction explanation (search for similar items in EconPapers)
Date: 2022
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (1)

Downloads: (external link)
http://link.springer.com/10.1007/s10796-021-10141-9 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:spr:infosf:v:24:y:2022:i:1:d:10.1007_s10796-021-10141-9

Ordering information: This journal article can be ordered from
http://www.springer.com/journal/10796

DOI: 10.1007/s10796-021-10141-9

Access Statistics for this article

Information Systems Frontiers is currently edited by Ram Ramesh and Raghav Rao

More articles in Information Systems Frontiers from Springer
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().