Towards successful automated negotiations based on Neural Networks

Mobile intelligent agents may assist the rapid and wide market penetration of services and products offered via e-commerce channels, as they improve the performance and sophistication of systems in the e-business domain. In this framework, the design and evaluation of agents handling automated negotiations on behalf of their human or corporate owners is a challenging research field. This paper proposes to enhance such agents with learning techniques, in order to achieve more profitable results for the parties they represent. As mobile agents have reduced processing capabilities and have to remotely communicate with their peers or migrate to foreign network nodes, the learning mechanisms they employ should require minimal resources. The proposed learning technique is based on neural networks (NNs) and is quite lightweight. It aims to reduce the cases of unsuccessful negotiations and maximize the client´s utility. The designed NN-assisted negotiation strategy has been empirically evaluated via numerous experiments