An Adaptive Approach for Decision Making Tactics in Automated Negotiation - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Applied Intelligence Année : 2013

An Adaptive Approach for Decision Making Tactics in Automated Negotiation

A. Bahrammirzaee
  • Fonction : Auteur
A. C. Chohra
  • Fonction : Auteur
K. K. Madani
  • Fonction : Auteur

Résumé

In this paper, an adaptive decision making approach of three families of tactics has been proposed for bilateral negotiation: the time dependent tactics, behavior dependent tactics, and time independent tactics. These tactics are more adaptive to the environment’s changes (reservation interval, time deadline, opponent behavior). The suggested time dependent tactics take advantage from round based time continuity and dynamics aspects (features) integrated in their modelling. For suggested behavior dependent tactics, a new formalization based on the percentage of change was introduced, which helps agents to be more prudent in the environments with incomplete information comparing to previous behavior dependent tactics suggested by Faratin et al. (Int. J. Robotics Auton. Syst. 24(3–4):159–182, 1998). Concerning the new family of tactics which are completely independent from time, the agents compute their offers based on their reservation interval. These tactics are useful when there is no time deadline and, in addition, when the behavior of opponent agents doesn’t follow any negotiation equilibrium. Moreover, new experimental measures are suggested which are more useful for final evaluation. The experiments conducted in this paper, prove the applicability of all three families of tactic.

Dates et versions

hal-01569662 , version 1 (27-07-2017)

Identifiants

Citer

A. Bahrammirzaee, A. C. Chohra, K. K. Madani. An Adaptive Approach for Decision Making Tactics in Automated Negotiation. Applied Intelligence, 2013, 39 (3), pp.583-606. ⟨10.1007/s10489-013-0434-8⟩. ⟨hal-01569662⟩
41 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More