Towards the use of Commitments in Multi-agent decision support systems

Isabel Sassoon, Nadin Kokciyan, Simon Parsons, Elizabeth I Sklar

Research output: Contribution to conferencePaperpeer-review

Abstract

When a decision support system relies on information coming from different sources to reason about different recommendations it may be desirable to compute, record, track and engineer the influence that each source has on the elements which form part of the reasoning process. Different sources of information may have varying levels of reliability and, consequently, decision makers may want to reduce the potential impact of inputs from less reliable sources. In this paper, we introduce an approach for managing variable-quality inputs to a human multi-agent decision support system by using computational argumentation and commitments to model the relationships between sources and the data they provide, in a context-sensitive way. The methodology is illustrated with a clinical case study.
Original languageEnglish
Number of pages4
Publication statusPublished - 15 Dec 2018
EventThe International Workshop on Dialogue, Explanation and Argumentation in Human-Agent Interaction - Southampton, United Kingdom
Duration: 15 Dec 201815 Dec 2018
https://sites.google.com/view/dexahai-18/home?authuser=0

Conference

ConferenceThe International Workshop on Dialogue, Explanation and Argumentation in Human-Agent Interaction
Abbreviated titleDEXAHAI 2018
Country/TerritoryUnited Kingdom
CitySouthampton
Period15/12/1815/12/18
Internet address

Keywords

  • Computing methodologies
  • Multi-agent systems
  • Knowl- edge representation and reasoning
  • Distributed artificial intelligence
  • Human-centered computing
  • Human computer interaction (HCI)
  • Human Agent Interaction
  • Multi-Agent Decision Support
  • Compu- tational Argumentation
  • Commitments

Fingerprint

Dive into the research topics of 'Towards the use of Commitments in Multi-agent decision support systems'. Together they form a unique fingerprint.

Cite this