X-RAI: A Framework for the Transparent, Responsible, and Accurate Use of Machine Learning in the Public Sector

Per Rådberg Nagbøl; Oliver Müller

X-RAI: A Framework for the Transparent, Responsible, and Accurate Use of Machine Learning in the Public Sector

Per Rådberg Nagbøl, Oliver Müller

Business IT

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

Abstract

This paper reports on an Action Design Research project taking place in the Danish Business Authority focusing on quality assurance and evaluation of machine learning models in production. The design artifact is a Framework (X-RAI) which stands for Transparency (X-Ray), Responsible(R), and explainable (X-AI). X-RAI consist of four sub-frameworks: the Model Impact and Clarification Framework, Evaluation Plan Framework, Evaluation Support Framework, and Retraining Execution Framework for machine learning that builds upon the theory of interpretable AI and practical experiences tested on nine different machine learning models used by the Danish Business Authority.

Original language	English
Title of host publication	EGOV-CeDEM-ePart 2020 : Proceedings of Ongoing Research, Practitioners, Workshops, Posters, and Projects of the International Conference EGOV-CeDEM-ePart 2020
Editors	Shefali Virkar, Marijn Janssen, Ida Lindgren, Ulf Melin, Francesco Mureddu, Peter Parycek, Efthimios Tambouris, Gerhard Schwabe, Hans Jochen Scholl
Number of pages	267
Place of Publication	Sweden
Publisher	CEUR Workshop Proceedings
Publication date	2020
Pages	259
Publication status	Published - 2020
Event	IFIP EGOV-ePart-CeDEM conference - Linköping Univeristy, Linköping, Sweden Duration: 31 Aug 2020 → 2 Sept 2020 http://dgsociety.org/egov-2020/

Conference

Conference	IFIP EGOV-ePart-CeDEM conference
Location	Linköping Univeristy
Country/Territory	Sweden
City	Linköping
Period	31/08/2020 → 02/09/2020
Internet address	http://dgsociety.org/egov-2020/

Series	CEUR Workshop Proceedings
Volume	2797
ISSN	1613-0073

Keywords

Machine Learning Evaluation
Government
Interpretability

Access to Document

paper25Final published version, 1 MBLicence: CC BY

1 Other contribution

X-RAI: A Framework for a Transparent, Explainable, Responsible, and Accurate Use of Artificial Intelligence
Nagbøl, P. R., Krancher, O. & Müller, O., 2024, 24 p.
Research output: Other contribution › Research

Open Access
File

Cite this

Nagbøl, P. R., & Müller, O. (2020). X-RAI: A Framework for the Transparent, Responsible, and Accurate Use of Machine Learning in the Public Sector. In S. Virkar, M. Janssen, I. Lindgren, U. Melin, F. Mureddu, P. Parycek, E. Tambouris, G. Schwabe, & H. J. Scholl (Eds.), EGOV-CeDEM-ePart 2020: Proceedings of Ongoing Research, Practitioners, Workshops, Posters, and Projects of the International Conference EGOV-CeDEM-ePart 2020 (pp. 259). CEUR Workshop Proceedings. http://ceur-ws.org/Vol-2797/paper25.pdf

Nagbøl, Per Rådberg ; Müller, Oliver. / X-RAI: A Framework for the Transparent, Responsible, and Accurate Use of Machine Learning in the Public Sector. EGOV-CeDEM-ePart 2020: Proceedings of Ongoing Research, Practitioners, Workshops, Posters, and Projects of the International Conference EGOV-CeDEM-ePart 2020. editor / Shefali Virkar ; Marijn Janssen ; Ida Lindgren ; Ulf Melin ; Francesco Mureddu ; Peter Parycek ; Efthimios Tambouris ; Gerhard Schwabe ; Hans Jochen Scholl. Sweden : CEUR Workshop Proceedings, 2020. pp. 259 (CEUR Workshop Proceedings, Vol. 2797).

@inproceedings{578239a8be66427fa9b00e5383d76f0e,

title = "X-RAI: A Framework for the Transparent, Responsible, and Accurate Use of Machine Learning in the Public Sector",

abstract = "This paper reports on an Action Design Research project taking place in the Danish Business Authority focusing on quality assurance and evaluation of machine learning models in production. The design artifact is a Framework (X-RAI) which stands for Transparency (X-Ray), Responsible(R), and explainable (X-AI). X-RAI consist of four sub-frameworks: the Model Impact and Clarification Framework, Evaluation Plan Framework, Evaluation Support Framework, and Retraining Execution Framework for machine learning that builds upon the theory of interpretable AI and practical experiences tested on nine different machine learning models used by the Danish Business Authority. ",

keywords = "Machine Learning Evaluation, Government, Interpretability",

author = "Nagb{\o}l, {Per R{\aa}dberg} and Oliver M{\"u}ller",

year = "2020",

language = "English",

series = "CEUR Workshop Proceedings",

publisher = "CEUR Workshop Proceedings",

pages = "259",

editor = "Shefali Virkar and Marijn Janssen and Ida Lindgren and Ulf Melin and Francesco Mureddu and Peter Parycek and Efthimios Tambouris and Gerhard Schwabe and Scholl, {Hans Jochen}",

booktitle = "EGOV-CeDEM-ePart 2020",

note = "IFIP EGOV-ePart-CeDEM conference ; Conference date: 31-08-2020 Through 02-09-2020",

url = "http://dgsociety.org/egov-2020/",

}

Nagbøl, PR & Müller, O 2020, X-RAI: A Framework for the Transparent, Responsible, and Accurate Use of Machine Learning in the Public Sector. in S Virkar, M Janssen, I Lindgren, U Melin, F Mureddu, P Parycek, E Tambouris, G Schwabe & HJ Scholl (eds), EGOV-CeDEM-ePart 2020: Proceedings of Ongoing Research, Practitioners, Workshops, Posters, and Projects of the International Conference EGOV-CeDEM-ePart 2020. CEUR Workshop Proceedings, Sweden, CEUR Workshop Proceedings, vol. 2797, pp. 259, IFIP EGOV-ePart-CeDEM conference , Linköping, Sweden, 31/08/2020. <http://ceur-ws.org/Vol-2797/paper25.pdf>

X-RAI: A Framework for the Transparent, Responsible, and Accurate Use of Machine Learning in the Public Sector. / Nagbøl, Per Rådberg; Müller, Oliver.
EGOV-CeDEM-ePart 2020: Proceedings of Ongoing Research, Practitioners, Workshops, Posters, and Projects of the International Conference EGOV-CeDEM-ePart 2020. ed. / Shefali Virkar; Marijn Janssen; Ida Lindgren; Ulf Melin; Francesco Mureddu; Peter Parycek; Efthimios Tambouris; Gerhard Schwabe; Hans Jochen Scholl. Sweden: CEUR Workshop Proceedings, 2020. p. 259 (CEUR Workshop Proceedings, Vol. 2797).

Research output: Conference Article in Proceeding or Book/Report chapter › Article in proceedings › Research › peer-review

TY - GEN

T1 - X-RAI: A Framework for the Transparent, Responsible, and Accurate Use of Machine Learning in the Public Sector

AU - Nagbøl, Per Rådberg

AU - Müller, Oliver

PY - 2020

Y1 - 2020

N2 - This paper reports on an Action Design Research project taking place in the Danish Business Authority focusing on quality assurance and evaluation of machine learning models in production. The design artifact is a Framework (X-RAI) which stands for Transparency (X-Ray), Responsible(R), and explainable (X-AI). X-RAI consist of four sub-frameworks: the Model Impact and Clarification Framework, Evaluation Plan Framework, Evaluation Support Framework, and Retraining Execution Framework for machine learning that builds upon the theory of interpretable AI and practical experiences tested on nine different machine learning models used by the Danish Business Authority.

AB - This paper reports on an Action Design Research project taking place in the Danish Business Authority focusing on quality assurance and evaluation of machine learning models in production. The design artifact is a Framework (X-RAI) which stands for Transparency (X-Ray), Responsible(R), and explainable (X-AI). X-RAI consist of four sub-frameworks: the Model Impact and Clarification Framework, Evaluation Plan Framework, Evaluation Support Framework, and Retraining Execution Framework for machine learning that builds upon the theory of interpretable AI and practical experiences tested on nine different machine learning models used by the Danish Business Authority.

KW - Machine Learning Evaluation

KW - Government

KW - Interpretability

M3 - Article in proceedings

T3 - CEUR Workshop Proceedings

SP - 259

BT - EGOV-CeDEM-ePart 2020

A2 - Virkar, Shefali

A2 - Janssen, Marijn

A2 - Lindgren, Ida

A2 - Melin, Ulf

A2 - Mureddu, Francesco

A2 - Parycek, Peter

A2 - Tambouris, Efthimios

A2 - Schwabe, Gerhard

A2 - Scholl, Hans Jochen

PB - CEUR Workshop Proceedings

CY - Sweden

T2 - IFIP EGOV-ePart-CeDEM conference

Y2 - 31 August 2020 through 2 September 2020

ER -

Nagbøl PR, Müller O. X-RAI: A Framework for the Transparent, Responsible, and Accurate Use of Machine Learning in the Public Sector. In Virkar S, Janssen M, Lindgren I, Melin U, Mureddu F, Parycek P, Tambouris E, Schwabe G, Scholl HJ, editors, EGOV-CeDEM-ePart 2020: Proceedings of Ongoing Research, Practitioners, Workshops, Posters, and Projects of the International Conference EGOV-CeDEM-ePart 2020. Sweden: CEUR Workshop Proceedings. 2020. p. 259. (CEUR Workshop Proceedings, Vol. 2797).

X-RAI: A Framework for the Transparent, Responsible, and Accurate Use of Machine Learning in the Public Sector

Abstract

Conference

Keywords

Access to Document

Fingerprint

Research output

X-RAI: A Framework for a Transparent, Explainable, Responsible, and Accurate Use of Artificial Intelligence

Cite this