Explainability Requirements for Time Series Forecasts: A Study in the Energy Domain

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Research Organisations

External Research Organisations

  • Kraft-Wärme-Kopplung GmbH
View graph of relations

Details

Original languageEnglish
Title of host publicationProceedings
Subtitle of host publication32nd IEEE International Requirements Engineering Conference, RE 2024
EditorsGrischa Liebel, Irit Hadar, Paola Spoletini
PublisherIEEE Computer Society
Pages229-239
Number of pages11
ISBN (electronic)9798350395112
ISBN (print)979-8-3503-9512-9
Publication statusPublished - 2024
Event32nd IEEE International Requirements Engineering Conference, RE 2024 - Reykjavik, Iceland
Duration: 24 Jun 202428 Jun 2024

Publication series

NameProceedings of the IEEE International Conference on Requirements Engineering
ISSN (Print)1090-705X
ISSN (electronic)2332-6441

Abstract

With the rise of artificial intelligence in industry, many companies rely on machine learning methods such as time series forecasting. By processing data from the past, such systems can provide predictions for data in the future. In practice, however, there is often skepticism about the quality of the forecasts. Explainability has been identified as a means to address this skepticism and foster trust. While there are already different methods to explain time series forecasts, it is unclear which of these explanations are actually useful for stakeholders. To investigate the need for explanations for time series forecasts, we conducted a study at a mid-sized German company in the energy domain. Throughout the study, 23 participants were shown five examples of different explanation types. For each type of explanation, we tested if it actually helped our participants to better understand the forecasts. We found that visual explanations including decision trees and feature importance charts were able to improve domain experts' understanding of time series forecasts. Textual explanations tended to lead to confusion rather than empowerment. While the exact findings and preferable types of explanations may vary between companies, our concrete results can provide a starting point for in-depth analyses in other environments.

Keywords

    Empirical Research, Explainability, Re-quirements Elicitation, Time Series Forecasting

ASJC Scopus subject areas

Cite this

Explainability Requirements for Time Series Forecasts: A Study in the Energy Domain. / Droste, Jakob; Fuchs, Ronja; Deters, Hannah et al.
Proceedings : 32nd IEEE International Requirements Engineering Conference, RE 2024. ed. / Grischa Liebel; Irit Hadar; Paola Spoletini. IEEE Computer Society, 2024. p. 229-239 (Proceedings of the IEEE International Conference on Requirements Engineering).

Research output: Chapter in book/report/conference proceedingConference contributionResearchpeer review

Droste, J, Fuchs, R, Deters, H, Klunder, J & Schneider, K 2024, Explainability Requirements for Time Series Forecasts: A Study in the Energy Domain. in G Liebel, I Hadar & P Spoletini (eds), Proceedings : 32nd IEEE International Requirements Engineering Conference, RE 2024. Proceedings of the IEEE International Conference on Requirements Engineering, IEEE Computer Society, pp. 229-239, 32nd IEEE International Requirements Engineering Conference, RE 2024, Reykjavik, Iceland, 24 Jun 2024. https://doi.org/10.1109/RE59067.2024.00030
Droste, J., Fuchs, R., Deters, H., Klunder, J., & Schneider, K. (2024). Explainability Requirements for Time Series Forecasts: A Study in the Energy Domain. In G. Liebel, I. Hadar, & P. Spoletini (Eds.), Proceedings : 32nd IEEE International Requirements Engineering Conference, RE 2024 (pp. 229-239). (Proceedings of the IEEE International Conference on Requirements Engineering). IEEE Computer Society. https://doi.org/10.1109/RE59067.2024.00030
Droste J, Fuchs R, Deters H, Klunder J, Schneider K. Explainability Requirements for Time Series Forecasts: A Study in the Energy Domain. In Liebel G, Hadar I, Spoletini P, editors, Proceedings : 32nd IEEE International Requirements Engineering Conference, RE 2024. IEEE Computer Society. 2024. p. 229-239. (Proceedings of the IEEE International Conference on Requirements Engineering). doi: 10.1109/RE59067.2024.00030
Droste, Jakob ; Fuchs, Ronja ; Deters, Hannah et al. / Explainability Requirements for Time Series Forecasts : A Study in the Energy Domain. Proceedings : 32nd IEEE International Requirements Engineering Conference, RE 2024. editor / Grischa Liebel ; Irit Hadar ; Paola Spoletini. IEEE Computer Society, 2024. pp. 229-239 (Proceedings of the IEEE International Conference on Requirements Engineering).
Download
@inproceedings{f741e3f0d98945f4bf294024aa44caaa,
title = "Explainability Requirements for Time Series Forecasts: A Study in the Energy Domain",
abstract = "With the rise of artificial intelligence in industry, many companies rely on machine learning methods such as time series forecasting. By processing data from the past, such systems can provide predictions for data in the future. In practice, however, there is often skepticism about the quality of the forecasts. Explainability has been identified as a means to address this skepticism and foster trust. While there are already different methods to explain time series forecasts, it is unclear which of these explanations are actually useful for stakeholders. To investigate the need for explanations for time series forecasts, we conducted a study at a mid-sized German company in the energy domain. Throughout the study, 23 participants were shown five examples of different explanation types. For each type of explanation, we tested if it actually helped our participants to better understand the forecasts. We found that visual explanations including decision trees and feature importance charts were able to improve domain experts' understanding of time series forecasts. Textual explanations tended to lead to confusion rather than empowerment. While the exact findings and preferable types of explanations may vary between companies, our concrete results can provide a starting point for in-depth analyses in other environments.",
keywords = "Empirical Research, Explainability, Re-quirements Elicitation, Time Series Forecasting",
author = "Jakob Droste and Ronja Fuchs and Hannah Deters and Jil Klunder and Kurt Schneider",
note = "Publisher Copyright: {\textcopyright} 2024 IEEE.; 32nd IEEE International Requirements Engineering Conference, RE 2024 ; Conference date: 24-06-2024 Through 28-06-2024",
year = "2024",
doi = "10.1109/RE59067.2024.00030",
language = "English",
isbn = "979-8-3503-9512-9",
series = "Proceedings of the IEEE International Conference on Requirements Engineering",
publisher = "IEEE Computer Society",
pages = "229--239",
editor = "Grischa Liebel and Irit Hadar and Paola Spoletini",
booktitle = "Proceedings",
address = "United States",

}

Download

TY - GEN

T1 - Explainability Requirements for Time Series Forecasts

T2 - 32nd IEEE International Requirements Engineering Conference, RE 2024

AU - Droste, Jakob

AU - Fuchs, Ronja

AU - Deters, Hannah

AU - Klunder, Jil

AU - Schneider, Kurt

N1 - Publisher Copyright: © 2024 IEEE.

PY - 2024

Y1 - 2024

N2 - With the rise of artificial intelligence in industry, many companies rely on machine learning methods such as time series forecasting. By processing data from the past, such systems can provide predictions for data in the future. In practice, however, there is often skepticism about the quality of the forecasts. Explainability has been identified as a means to address this skepticism and foster trust. While there are already different methods to explain time series forecasts, it is unclear which of these explanations are actually useful for stakeholders. To investigate the need for explanations for time series forecasts, we conducted a study at a mid-sized German company in the energy domain. Throughout the study, 23 participants were shown five examples of different explanation types. For each type of explanation, we tested if it actually helped our participants to better understand the forecasts. We found that visual explanations including decision trees and feature importance charts were able to improve domain experts' understanding of time series forecasts. Textual explanations tended to lead to confusion rather than empowerment. While the exact findings and preferable types of explanations may vary between companies, our concrete results can provide a starting point for in-depth analyses in other environments.

AB - With the rise of artificial intelligence in industry, many companies rely on machine learning methods such as time series forecasting. By processing data from the past, such systems can provide predictions for data in the future. In practice, however, there is often skepticism about the quality of the forecasts. Explainability has been identified as a means to address this skepticism and foster trust. While there are already different methods to explain time series forecasts, it is unclear which of these explanations are actually useful for stakeholders. To investigate the need for explanations for time series forecasts, we conducted a study at a mid-sized German company in the energy domain. Throughout the study, 23 participants were shown five examples of different explanation types. For each type of explanation, we tested if it actually helped our participants to better understand the forecasts. We found that visual explanations including decision trees and feature importance charts were able to improve domain experts' understanding of time series forecasts. Textual explanations tended to lead to confusion rather than empowerment. While the exact findings and preferable types of explanations may vary between companies, our concrete results can provide a starting point for in-depth analyses in other environments.

KW - Empirical Research

KW - Explainability

KW - Re-quirements Elicitation

KW - Time Series Forecasting

UR - http://www.scopus.com/inward/record.url?scp=85202751217&partnerID=8YFLogxK

U2 - 10.1109/RE59067.2024.00030

DO - 10.1109/RE59067.2024.00030

M3 - Conference contribution

AN - SCOPUS:85202751217

SN - 979-8-3503-9512-9

T3 - Proceedings of the IEEE International Conference on Requirements Engineering

SP - 229

EP - 239

BT - Proceedings

A2 - Liebel, Grischa

A2 - Hadar, Irit

A2 - Spoletini, Paola

PB - IEEE Computer Society

Y2 - 24 June 2024 through 28 June 2024

ER -

By the same author(s)