Automated Dynamic Negotiation over Environmental Issues

Negotiation is a common means of resolving conflicts in social interactions. A popular approach for modeling social negotiation is automated negotiation. It is a distributed search in the space of potential agreements, facilitated by an agent-based model (ABM). Although automated negotiation is extensively applied in different fields of e-commerce, its application in environmental studies is still unexplored. This paper aims to lead the negotiation process over environmental issues in an efficient way where the possible agreement can be reached in few rounds of negotiation. To achieve this goal, an ABM is developed which has two significant characteristics. First, the proposer agent automatically learns the preferences of all stakeholder using the arguments and responses received from them in the rounds of negotiation. Second, the proposer accelerates the negotiation by automating the process of proposal-offering. To this end, first, the problem of proposal selection in one-to-one negotiation with each stakeholder is modeled using Markov Random Fields (MRF) and is solved using a belief propagation-based approach. Then, the proposer applies statistical analysis to identify the most optimal proposal and conducts a one-to-many negotiation.