A probabilistic model for trust and reputation

George Vogiatzis, I. MacGillivray, Maria Chli

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

This paper concerns the problem of agent trust in an electronic market place. We maintain that agent trust involves making decisions under uncertainty and therefore the phenomenon should be modelled probabilistically. We therefore propose a probabilistic framework that models agent interactions as a Hidden Markov Model (HMM). The observations of the HMM are the interaction outcomes and the hidden state is the underlying probability of a good outcome. The task of deciding whether to interact with another agent reduces to probabilistic inference of the current state of that agent given all previous interaction outcomes. The model is extended to include a probabilistic reputation system which involves agents gathering opinions about other agents and fusing them with their own beliefs. Our system is fully probabilistic and hence delivers the following improvements with respect to previous work: (a) the model assumptions are faithfully translated into algorithms; our system is optimal under those assumptions, (b) It can account for agents whose behaviour is not static with time (c) it can estimate the rate with which an agent's behaviour changes. The system is shown to significantly outperform previous state-of-the-art methods in several numerical experiments. Copyright © 2010, International Foundation for Autonomous Agents and Multiagent Systems (www.ifaamas.org). All rights reserved.

Original languageEnglish
Title of host publicationProceedings of the 9th International Conference on Autonomous Agents and Multiagent Systems
PublisherIFAAMAS
Pages225-232
Number of pages8
Volume1
ISBN (Print)978-1-61738771-5
Publication statusPublished - 2010
Event9th International Joint Conference on Autonomous Agents and Multiagent Systems - Toronto, ON, Canada
Duration: 10 May 201014 May 2010

Conference

Conference9th International Joint Conference on Autonomous Agents and Multiagent Systems
Abbreviated titleAAMAS 2010
CountryCanada
CityToronto, ON
Period10/05/1014/05/10

Fingerprint

Hidden Markov models
Optimal systems
Autonomous agents
Multi agent systems
Decision making
Statistical Models
Experiments
Uncertainty

Keywords

  • distributed artificial intelligence
  • multiagent systems
  • simulation and modelling
  • model development
  • modelling methodologies
  • algorithms
  • trust
  • reliability
  • reputation

Cite this

Vogiatzis, G., MacGillivray, I., & Chli, M. (2010). A probabilistic model for trust and reputation. In Proceedings of the 9th International Conference on Autonomous Agents and Multiagent Systems (Vol. 1, pp. 225-232). IFAAMAS.
Vogiatzis, George ; MacGillivray, I. ; Chli, Maria. / A probabilistic model for trust and reputation. Proceedings of the 9th International Conference on Autonomous Agents and Multiagent Systems. Vol. 1 IFAAMAS, 2010. pp. 225-232
@inproceedings{b8c1f89904de4d2b88910efc04f8c5d4,
title = "A probabilistic model for trust and reputation",
abstract = "This paper concerns the problem of agent trust in an electronic market place. We maintain that agent trust involves making decisions under uncertainty and therefore the phenomenon should be modelled probabilistically. We therefore propose a probabilistic framework that models agent interactions as a Hidden Markov Model (HMM). The observations of the HMM are the interaction outcomes and the hidden state is the underlying probability of a good outcome. The task of deciding whether to interact with another agent reduces to probabilistic inference of the current state of that agent given all previous interaction outcomes. The model is extended to include a probabilistic reputation system which involves agents gathering opinions about other agents and fusing them with their own beliefs. Our system is fully probabilistic and hence delivers the following improvements with respect to previous work: (a) the model assumptions are faithfully translated into algorithms; our system is optimal under those assumptions, (b) It can account for agents whose behaviour is not static with time (c) it can estimate the rate with which an agent's behaviour changes. The system is shown to significantly outperform previous state-of-the-art methods in several numerical experiments. Copyright {\circledC} 2010, International Foundation for Autonomous Agents and Multiagent Systems (www.ifaamas.org). All rights reserved.",
keywords = "distributed artificial intelligence, multiagent systems, simulation and modelling, model development, modelling methodologies, algorithms, trust, reliability, reputation",
author = "George Vogiatzis and I. MacGillivray and Maria Chli",
year = "2010",
language = "English",
isbn = "978-1-61738771-5",
volume = "1",
pages = "225--232",
booktitle = "Proceedings of the 9th International Conference on Autonomous Agents and Multiagent Systems",
publisher = "IFAAMAS",

}

Vogiatzis, G, MacGillivray, I & Chli, M 2010, A probabilistic model for trust and reputation. in Proceedings of the 9th International Conference on Autonomous Agents and Multiagent Systems. vol. 1, IFAAMAS, pp. 225-232, 9th International Joint Conference on Autonomous Agents and Multiagent Systems, Toronto, ON, Canada, 10/05/10.

A probabilistic model for trust and reputation. / Vogiatzis, George; MacGillivray, I.; Chli, Maria.

Proceedings of the 9th International Conference on Autonomous Agents and Multiagent Systems. Vol. 1 IFAAMAS, 2010. p. 225-232.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - A probabilistic model for trust and reputation

AU - Vogiatzis, George

AU - MacGillivray, I.

AU - Chli, Maria

PY - 2010

Y1 - 2010

N2 - This paper concerns the problem of agent trust in an electronic market place. We maintain that agent trust involves making decisions under uncertainty and therefore the phenomenon should be modelled probabilistically. We therefore propose a probabilistic framework that models agent interactions as a Hidden Markov Model (HMM). The observations of the HMM are the interaction outcomes and the hidden state is the underlying probability of a good outcome. The task of deciding whether to interact with another agent reduces to probabilistic inference of the current state of that agent given all previous interaction outcomes. The model is extended to include a probabilistic reputation system which involves agents gathering opinions about other agents and fusing them with their own beliefs. Our system is fully probabilistic and hence delivers the following improvements with respect to previous work: (a) the model assumptions are faithfully translated into algorithms; our system is optimal under those assumptions, (b) It can account for agents whose behaviour is not static with time (c) it can estimate the rate with which an agent's behaviour changes. The system is shown to significantly outperform previous state-of-the-art methods in several numerical experiments. Copyright © 2010, International Foundation for Autonomous Agents and Multiagent Systems (www.ifaamas.org). All rights reserved.

AB - This paper concerns the problem of agent trust in an electronic market place. We maintain that agent trust involves making decisions under uncertainty and therefore the phenomenon should be modelled probabilistically. We therefore propose a probabilistic framework that models agent interactions as a Hidden Markov Model (HMM). The observations of the HMM are the interaction outcomes and the hidden state is the underlying probability of a good outcome. The task of deciding whether to interact with another agent reduces to probabilistic inference of the current state of that agent given all previous interaction outcomes. The model is extended to include a probabilistic reputation system which involves agents gathering opinions about other agents and fusing them with their own beliefs. Our system is fully probabilistic and hence delivers the following improvements with respect to previous work: (a) the model assumptions are faithfully translated into algorithms; our system is optimal under those assumptions, (b) It can account for agents whose behaviour is not static with time (c) it can estimate the rate with which an agent's behaviour changes. The system is shown to significantly outperform previous state-of-the-art methods in several numerical experiments. Copyright © 2010, International Foundation for Autonomous Agents and Multiagent Systems (www.ifaamas.org). All rights reserved.

KW - distributed artificial intelligence

KW - multiagent systems

KW - simulation and modelling

KW - model development

KW - modelling methodologies

KW - algorithms

KW - trust

KW - reliability

KW - reputation

UR - http://www.scopus.com/inward/record.url?scp=80055030975&partnerID=8YFLogxK

UR - http://www.aamas-conference.org/Proceedings/aamas2010/pdf/01%20Full%20Papers/04_03_FP_0533.pdf

M3 - Conference contribution

SN - 978-1-61738771-5

VL - 1

SP - 225

EP - 232

BT - Proceedings of the 9th International Conference on Autonomous Agents and Multiagent Systems

PB - IFAAMAS

ER -

Vogiatzis G, MacGillivray I, Chli M. A probabilistic model for trust and reputation. In Proceedings of the 9th International Conference on Autonomous Agents and Multiagent Systems. Vol. 1. IFAAMAS. 2010. p. 225-232