Fraunhofer-Gesellschaft

Publica

Hier finden Sie wissenschaftliche Publikationen aus den Fraunhofer-Instituten.

A Study on Trust in Black Box Models and Post-hoc Explanations

 
: El Bekri, Nadia; Kling, J.; Huber, M.

:
Postprint urn:nbn:de:0011-n-5436585 (559 KByte PDF)
MD5 Fingerprint: ee32707f3e6711ffddc57e6e20b3ae79
The original publication is available at springerlink.com
Created on: 15.05.2020


Alvarez, Francisco M. (Ed.):
14th International Conference on Soft Computing Models in Industrial and Environmental Applications, SOCO 2019. Proceedings : Seville, Spain, May 13-15, 2019
Cham: Springer Nature, 2019 (Advances in Intelligent Systems and Computing 950)
ISBN: 978-3-030-20054-1 (Print)
ISBN: 978-3-030-20055-8 (Online)
pp.35-46
International Conference on Soft Computing Models in Industrial and Environmental Applications (SOCO) <14, 2019, Seville>
English
Conference Paper, Electronic Publication
Fraunhofer IOSB ()
Fraunhofer IPA ()
machine learning; black box; explainability; interpretability; trust

Abstract
Machine learning algorithms that construct complex prediction models are increasingly used for decision-making due to their high accuracy, e.g., to decide whether a bank customer should receive a loan or not. Due to the complexity, the models are perceived as black boxes. One approach is to augment the models with post-hoc explainability. In this work, we evaluate three different explanation approaches based on the users’ initial trust, the users’ trust in the provided explanation, and the established trust in the black box by a within-subject design study.

: http://publica.fraunhofer.de/documents/N-543658.html