Multistage Fuzzy Decision Making in Bilateral Negotiation with Finite Termination Times

J. Richter; R. Kowalczyk; Matthias Klusch

In: Ann E. Nicholson; X. Li (Hrsg.). AI 2009: Advances in Artificial Intelligence, Proceedings of the 22nd Australasian Joint Conference on Advances in Artificial Intelligence. Australian Joint Conference on Artificial Intelligence (AI-2009), 22nd, December 1-4, Melbourne, Australia, Pages 21-30, Lecture Notes in Computer Science (LNCS), Vol. 5866, ISBN 978-3-642-10438-1, Springer-Verlag, 2009.


In this paper we model the negotiation process as a multistage fuzzy decision problem where the agents preferences are represented by a fuzzy goal and fuzzy constraints. The opponent is represented by a fuzzy Markov decision process in the form of offer-response patterns which enables utilization of limited and uncertain information, e.g. the characteristics of the concession behaviour. We show that we can obtain adaptive negotiation strategies by only using the negotiation threads of two past cases to create and update the fuzzy transition matrix. The experimental evaluation demonstrates that our approach is adaptive towards different negotiation behaviours and that the fuzzy representation of the preferences and the transition matrix allows for application in many scenarios where the available information, preferences and constraints are soft or imprecise.

Deutsches Forschungszentrum für Künstliche Intelligenz
German Research Center for Artificial Intelligence