Acquiring Semantic Knowledge for User Model Updates via Human-Agent Alignment Dialogues

Pei Yu Chen; Myrthe L. Tielman; Dirk K.J. Heylen; Catholijn M. Jonker; M. Birna Van Riemsdijk

doi:10.3233/FAIA230077

Acquiring Semantic Knowledge for User Model Updates via Human-Agent Alignment Dialogues

Pei Yu Chen^*, Myrthe L. Tielman, Dirk K.J. Heylen, Catholijn M. Jonker, M. Birna Van Riemsdijk

^*Corresponding author for this work

Interactive Intelligence

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

3 Citations (Scopus)

135 Downloads (Pure)

Abstract

For personal assistive technologies to effectively support users, they need a user model that records information about the user, such as their goals, values, and context. Knowledge-based techniques can model the relationships between these concepts, enabling the support agent to act in accordance with the user's values. However, user models require updating over time to accommodate changes and continuously align with what the user deems important. In our work, we propose and investigate the use of human-agent alignment dialogues for establishing whether user model updates are needed and acquiring the necessary information for these updates. In this paper, we perform an exploratory qualitative focus group study in which we investigate participants' opinions about written examples of alignment dialogues, as a foundation for their design. Transcripts were analyzed using thematic analysis. A main theme that emerged concerns the potential impact of agent utterances on the user's feelings about themselves and about the agent.

Original language	English
Title of host publication	HHAI 2023
Subtitle of host publication	Augmenting Human Intellect - Proceedings of the 2nd International Conference on Hybrid Human-Artificial Intelligence
Editors	Paul Lukowicz, Sven Mayer, Janin Koch, John Shawe-Taylor, Ilaria Tiddi
Publisher	IOS Press
Pages	93-107
Number of pages	15
ISBN (Electronic)	9781643683942
DOIs	https://doi.org/10.3233/FAIA230077
Publication status	Published - 2023
Event	2nd International Conference on Hybrid Human Artificial Intelligence - Munich, Germany Duration: 26 Jun 2023 → 30 Jun 2023 Conference number: 2 https://hhai-conference.org/2023/

Publication series

Name	Frontiers in Artificial Intelligence and Applications
Volume	368
ISSN (Print)	0922-6389
ISSN (Electronic)	1879-8314

Conference

Conference	2nd International Conference on Hybrid Human Artificial Intelligence
Abbreviated title	HHAI 2023
Country/Territory	Germany
City	Munich
Period	26/06/23 → 30/06/23
Internet address	https://hhai-conference.org/2023/

Keywords

Behaviour support technology
Conversational agents
Dialogue
Human-agent alignment
User modelling
Values

Access to Document

10.3233/FAIA230077

FAIA-368-FAIA230077Final published version, 467 KBLicence: CC BY-NC

Cite this

Chen, P. Y., Tielman, M. L., Heylen, D. K. J., Jonker, C. M., & Van Riemsdijk, M. B. (2023). Acquiring Semantic Knowledge for User Model Updates via Human-Agent Alignment Dialogues. In P. Lukowicz, S. Mayer, J. Koch, J. Shawe-Taylor, & I. Tiddi (Eds.), HHAI 2023: Augmenting Human Intellect - Proceedings of the 2nd International Conference on Hybrid Human-Artificial Intelligence (pp. 93-107). (Frontiers in Artificial Intelligence and Applications; Vol. 368). IOS Press. https://doi.org/10.3233/FAIA230077

Chen, Pei Yu ; Tielman, Myrthe L. ; Heylen, Dirk K.J. et al. / Acquiring Semantic Knowledge for User Model Updates via Human-Agent Alignment Dialogues. HHAI 2023: Augmenting Human Intellect - Proceedings of the 2nd International Conference on Hybrid Human-Artificial Intelligence. editor / Paul Lukowicz ; Sven Mayer ; Janin Koch ; John Shawe-Taylor ; Ilaria Tiddi. IOS Press, 2023. pp. 93-107 (Frontiers in Artificial Intelligence and Applications).

@inproceedings{0ae6c2d65d754c62b87ed533302f5ce9,

title = "Acquiring Semantic Knowledge for User Model Updates via Human-Agent Alignment Dialogues",

abstract = "For personal assistive technologies to effectively support users, they need a user model that records information about the user, such as their goals, values, and context. Knowledge-based techniques can model the relationships between these concepts, enabling the support agent to act in accordance with the user's values. However, user models require updating over time to accommodate changes and continuously align with what the user deems important. In our work, we propose and investigate the use of human-agent alignment dialogues for establishing whether user model updates are needed and acquiring the necessary information for these updates. In this paper, we perform an exploratory qualitative focus group study in which we investigate participants' opinions about written examples of alignment dialogues, as a foundation for their design. Transcripts were analyzed using thematic analysis. A main theme that emerged concerns the potential impact of agent utterances on the user's feelings about themselves and about the agent.",

keywords = "Behaviour support technology, Conversational agents, Dialogue, Human-agent alignment, User modelling, Values",

author = "Chen, {Pei Yu} and Tielman, {Myrthe L.} and Heylen, {Dirk K.J.} and Jonker, {Catholijn M.} and {Van Riemsdijk}, {M. Birna}",

year = "2023",

doi = "10.3233/FAIA230077",

language = "English",

series = "Frontiers in Artificial Intelligence and Applications",

publisher = "IOS Press",

pages = "93--107",

editor = "Paul Lukowicz and Sven Mayer and Janin Koch and John Shawe-Taylor and Ilaria Tiddi",

booktitle = "HHAI 2023",

address = "Netherlands",

note = "2nd International Conference on Hybrid Human Artificial Intelligence, HHAI 2023 ; Conference date: 26-06-2023 Through 30-06-2023",

url = "https://hhai-conference.org/2023/",

}

Chen, PY , Tielman, ML, Heylen, DKJ, Jonker, CM & Van Riemsdijk, MB 2023, Acquiring Semantic Knowledge for User Model Updates via Human-Agent Alignment Dialogues. in P Lukowicz, S Mayer, J Koch, J Shawe-Taylor & I Tiddi (eds), HHAI 2023: Augmenting Human Intellect - Proceedings of the 2nd International Conference on Hybrid Human-Artificial Intelligence. Frontiers in Artificial Intelligence and Applications, vol. 368, IOS Press, pp. 93-107, 2nd International Conference on Hybrid Human Artificial Intelligence, Munich, Bavaria, Germany, 26/06/23. https://doi.org/10.3233/FAIA230077

Acquiring Semantic Knowledge for User Model Updates via Human-Agent Alignment Dialogues. / Chen, Pei Yu ; Tielman, Myrthe L.; Heylen, Dirk K.J. et al.
HHAI 2023: Augmenting Human Intellect - Proceedings of the 2nd International Conference on Hybrid Human-Artificial Intelligence. ed. / Paul Lukowicz; Sven Mayer; Janin Koch; John Shawe-Taylor; Ilaria Tiddi. IOS Press, 2023. p. 93-107 (Frontiers in Artificial Intelligence and Applications; Vol. 368).

Research output: Chapter in Book/Conference proceedings/Edited volume › Conference contribution › Scientific › peer-review

TY - GEN

T1 - Acquiring Semantic Knowledge for User Model Updates via Human-Agent Alignment Dialogues

AU - Chen, Pei Yu

AU - Tielman, Myrthe L.

AU - Heylen, Dirk K.J.

AU - Jonker, Catholijn M.

AU - Van Riemsdijk, M. Birna

N1 - Conference code: 2

PY - 2023

Y1 - 2023

N2 - For personal assistive technologies to effectively support users, they need a user model that records information about the user, such as their goals, values, and context. Knowledge-based techniques can model the relationships between these concepts, enabling the support agent to act in accordance with the user's values. However, user models require updating over time to accommodate changes and continuously align with what the user deems important. In our work, we propose and investigate the use of human-agent alignment dialogues for establishing whether user model updates are needed and acquiring the necessary information for these updates. In this paper, we perform an exploratory qualitative focus group study in which we investigate participants' opinions about written examples of alignment dialogues, as a foundation for their design. Transcripts were analyzed using thematic analysis. A main theme that emerged concerns the potential impact of agent utterances on the user's feelings about themselves and about the agent.

AB - For personal assistive technologies to effectively support users, they need a user model that records information about the user, such as their goals, values, and context. Knowledge-based techniques can model the relationships between these concepts, enabling the support agent to act in accordance with the user's values. However, user models require updating over time to accommodate changes and continuously align with what the user deems important. In our work, we propose and investigate the use of human-agent alignment dialogues for establishing whether user model updates are needed and acquiring the necessary information for these updates. In this paper, we perform an exploratory qualitative focus group study in which we investigate participants' opinions about written examples of alignment dialogues, as a foundation for their design. Transcripts were analyzed using thematic analysis. A main theme that emerged concerns the potential impact of agent utterances on the user's feelings about themselves and about the agent.

KW - Behaviour support technology

KW - Conversational agents

KW - Dialogue

KW - Human-agent alignment

KW - User modelling

KW - Values

UR - http://www.scopus.com/inward/record.url?scp=85171440252&partnerID=8YFLogxK

U2 - 10.3233/FAIA230077

DO - 10.3233/FAIA230077

M3 - Conference contribution

AN - SCOPUS:85171440252

T3 - Frontiers in Artificial Intelligence and Applications

SP - 93

EP - 107

BT - HHAI 2023

A2 - Lukowicz, Paul

A2 - Mayer, Sven

A2 - Koch, Janin

A2 - Shawe-Taylor, John

A2 - Tiddi, Ilaria

PB - IOS Press

T2 - 2nd International Conference on Hybrid Human Artificial Intelligence

Y2 - 26 June 2023 through 30 June 2023

ER -

Chen PY , Tielman ML, Heylen DKJ, Jonker CM, Van Riemsdijk MB. Acquiring Semantic Knowledge for User Model Updates via Human-Agent Alignment Dialogues. In Lukowicz P, Mayer S, Koch J, Shawe-Taylor J, Tiddi I, editors, HHAI 2023: Augmenting Human Intellect - Proceedings of the 2nd International Conference on Hybrid Human-Artificial Intelligence. IOS Press. 2023. p. 93-107. (Frontiers in Artificial Intelligence and Applications). doi: 10.3233/FAIA230077

Acquiring Semantic Knowledge for User Model Updates via Human-Agent Alignment Dialogues

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this