Abstract
The “negotiation dance”, as Raiffa calls the dynamic pattern of the bidding, has an important influence on the outcome of the negotiation. The current practice of evaluating a negotiation strategy is to focus on fairness and quality aspects of the agreement. In this article we present the framework DANS (Dynamics Analysis of Negotiation Strategies) for the analysis of the dynamic patterns of the bidding as a means to evaluate the strengths and weaknesses of negotiation strategies for bidding. The method provides the tools to perform a detailed and quantified analysis of a negotiation between two agents in terms of dynamic properties of the negotiation trace. The classification of negotiation steps in the dance plays a central role in the analysis. The method can be applied to tournaments, but can also be used to analyze single 1-on-1 negotiation sessions. The sessions can be played by humans or by software agents. Using DANS we show that some strategies are sensitive to the bidding behaviour of the opponent, and some depend on a correct model of the opponent. DANS helped us in discovering that domain characteristics are important for the analysis of strategies. Some strategies rely heavily on some domain assumptions. Furthermore, the results illustrate that having domain knowledge is not always enough to avoid making unintentional steps. The method is demonstrated in the analysis of three strategies from the literature ABMP, Trade-Off and Bayesian Agent.