A Genetic Algorithm Based on Metropolis Rule and Its Application in Multi-Lateral Multi-Issue Negotiation
Yang Yi-xian · Dianzi xuebao · 2009
To make the negotiation agents gain satisfying result and negotiate more efficiently in multi-agent based e-commerce,a genetic algorithm based on metropolis rule is presented and the algorithm is applied in multi-lateral multi-issue simultaneous bidding negotiation.After 1000 times of experiments,simple genetic algorithm averagely needs 155 runs to gain the satisfying result,while the genetic algorithm based on metropolis rule averagely needs 76 runs to gain the satisfying result.The experimental result show the genetic algorithm based on metropolis rule can gain the optimal negotiate result more efficiently in multi-literal multi-issue negotiation.