Measuring of strategies' similarity in automated negotiation

Abstract

Negotiation is a process between self-interested agents in ecommerce trying to reach an agreement on one or multi issues. The outcome of the negotiation depends on several parameters such as the agents' strategies and the knowledge one agent has about the opponents. One way for discovering opponent's strategy is to find the similarity between strategies. In this paper we present a simple model for measuring the similarity of negotiators' strategies. Our measure is based only on the history of the offers during the sessions of negotiation and we use a notion of Levenshtein distance. We implement this measure and experimentally show that the result of using this measure can improve the recognition of negotiation strategy. Also, this measure can be used for modeling behaviors of negotiators and predictive decision-making.