Explaining Bayesian Networks in Natural Language: State of the Art andChallenges

Conor  Hennessy; Alberto Bugarin; Ehud Reiter

Explaining Bayesian Networks in Natural Language: State of the Art andChallenges

Conor Hennessy, Alberto Bugarin, Ehud Reiter

Research output: Chapter in Book/Report/Conference proceeding › Published conference contribution

4 Downloads (Pure)

Abstract

In order to increase trust in the usage ofBayesian networks and to cement their roleas a model which can aid in critical deci-sion making, the challenge of explainabilitymust be faced. Previous attempts at explain-ing Bayesian networks have largely focusedon graphical or visual aids. In this paper weaim to highlight the importance of a naturallanguage approach to explanation and to dis-cuss some of the previous and state of the artattempts of the textual explanation of BayesianNetworks. We outline several challenges thatremain to be addressed in the generation andvalidation of natural language explanations ofBayesian Networks. This can serve as a re-search agenda for future work on natural lan-guage explanations of Bayesian Networks.

Original language	English
Title of host publication	2nd Workshop on Interactive Natural Language Technologyfor Explainable Artificial Intelligence
Subtitle of host publication	Proceedings of NL4XAI
Publisher	The Association for Computational Linguistics
Pages	28-33
Number of pages	6
ISBN (Print)	978-1-952148-56-9
Publication status	Published - 18 Dec 2020
Event	2nd Workshop on Interactive Natural Language Technology for Explainable Artificial Intelligence - Duration: 18 Dec 2020 → 18 Dec 2020 https://sites.google.com/view/nl4xai2020/program

Workshop

Workshop	2nd Workshop on Interactive Natural Language Technology for Explainable Artificial Intelligence
Period	18/12/20 → 18/12/20
Internet address	https://sites.google.com/view/nl4xai2020/program

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

Conor_et_al_NL4XAI_ExplainingBayesianNetworks_VoR
ACL materials are Copyright © 1963–2021 ACL; other materials are copyrighted by their respective copyright holders. Materials prior to 2016 here are licensed under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 International License. Permission is granted to make copies for the purposes of teaching and research. Materials published in or after 2016 are licensed on a Creative Commons Attribution 4.0 International License. https://creativecommons.org/licenses/by/4.0/
Final published version, 249 KBLicence: CC BY

https://www.aclweb.org/anthology/2020.nl4xai-1.7/Licence: CC BY
https://www.aclweb.org/anthology/2020.nl4xai-1.7/Licence: Unspecified

Cite this

Explaining Bayesian Networks in Natural Language : State of the Art andChallenges. / Hennessy, Conor ; Bugarin, Alberto; Reiter, Ehud.

2nd Workshop on Interactive Natural Language Technologyfor Explainable Artificial Intelligence: Proceedings of NL4XAI. The Association for Computational Linguistics, 2020. p. 28-33.

Research output: Chapter in Book/Report/Conference proceeding › Published conference contribution

Hennessy, C, Bugarin, A & Reiter, E 2020, Explaining Bayesian Networks in Natural Language: State of the Art andChallenges. in 2nd Workshop on Interactive Natural Language Technologyfor Explainable Artificial Intelligence: Proceedings of NL4XAI. The Association for Computational Linguistics, pp. 28-33, 2nd Workshop on Interactive Natural Language Technology for Explainable Artificial Intelligence, 18/12/20. <https://www.aclweb.org/anthology/2020.nl4xai-1.7/>

@inproceedings{3175cdd1ac1046799226c262d9364e6e,

title = "Explaining Bayesian Networks in Natural Language: State of the Art andChallenges",

abstract = "In order to increase trust in the usage ofBayesian networks and to cement their roleas a model which can aid in critical deci-sion making, the challenge of explainabilitymust be faced. Previous attempts at explain-ing Bayesian networks have largely focusedon graphical or visual aids. In this paper weaim to highlight the importance of a naturallanguage approach to explanation and to dis-cuss some of the previous and state of the artattempts of the textual explanation of BayesianNetworks. We outline several challenges thatremain to be addressed in the generation andvalidation of natural language explanations ofBayesian Networks. This can serve as a re-search agenda for future work on natural lan-guage explanations of Bayesian Networks.",

author = "Conor Hennessy and Alberto Bugarin and Ehud Reiter",

note = "Acknowledgments: This project has received funding from the Eu-ropean Union{\textquoteright}s Horizon 2020 research and inno-vation programme under the Marie Sk{\l}odowska-Curie Grant Agreement No. 860621. It was also funded by the Spanish Ministry for Science, In-novation and Universities, the Galician Ministryof Education, University and Professional Train-ing and the European Regional Development Fund(grants TIN2017-84796-C2-1-R, ED431C2018/29and ED431G2019/04); 2nd Workshop on Interactive Natural Language Technology for Explainable Artificial Intelligence ; Conference date: 18-12-2020 Through 18-12-2020",

year = "2020",

month = dec,

day = "18",

language = "English",

isbn = "978-1-952148-56-9",

pages = "28--33",

booktitle = "2nd Workshop on Interactive Natural Language Technologyfor Explainable Artificial Intelligence",

publisher = "The Association for Computational Linguistics",

url = "https://sites.google.com/view/nl4xai2020/program",

}

TY - GEN

T1 - Explaining Bayesian Networks in Natural Language

T2 - 2nd Workshop on Interactive Natural Language Technology for Explainable Artificial Intelligence

AU - Hennessy, Conor

AU - Bugarin, Alberto

AU - Reiter, Ehud

N1 - Acknowledgments: This project has received funding from the Eu-ropean Union’s Horizon 2020 research and inno-vation programme under the Marie Skłodowska-Curie Grant Agreement No. 860621. It was also funded by the Spanish Ministry for Science, In-novation and Universities, the Galician Ministryof Education, University and Professional Train-ing and the European Regional Development Fund(grants TIN2017-84796-C2-1-R, ED431C2018/29and ED431G2019/04)

PY - 2020/12/18

Y1 - 2020/12/18

N2 - In order to increase trust in the usage ofBayesian networks and to cement their roleas a model which can aid in critical deci-sion making, the challenge of explainabilitymust be faced. Previous attempts at explain-ing Bayesian networks have largely focusedon graphical or visual aids. In this paper weaim to highlight the importance of a naturallanguage approach to explanation and to dis-cuss some of the previous and state of the artattempts of the textual explanation of BayesianNetworks. We outline several challenges thatremain to be addressed in the generation andvalidation of natural language explanations ofBayesian Networks. This can serve as a re-search agenda for future work on natural lan-guage explanations of Bayesian Networks.

AB - In order to increase trust in the usage ofBayesian networks and to cement their roleas a model which can aid in critical deci-sion making, the challenge of explainabilitymust be faced. Previous attempts at explain-ing Bayesian networks have largely focusedon graphical or visual aids. In this paper weaim to highlight the importance of a naturallanguage approach to explanation and to dis-cuss some of the previous and state of the artattempts of the textual explanation of BayesianNetworks. We outline several challenges thatremain to be addressed in the generation andvalidation of natural language explanations ofBayesian Networks. This can serve as a re-search agenda for future work on natural lan-guage explanations of Bayesian Networks.

UR - https://sites.google.com/view/nl4xai2020/program

M3 - Published conference contribution

SN - 978-1-952148-56-9

SP - 28

EP - 33

BT - 2nd Workshop on Interactive Natural Language Technologyfor Explainable Artificial Intelligence

PB - The Association for Computational Linguistics

Y2 - 18 December 2020 through 18 December 2020

ER -

Explaining Bayesian Networks in Natural Language: State of the Art andChallenges

Abstract

Workshop

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this