The search for potentially interesting products in an e-marketplace: An agent-to-agent argumentation approach

In existing e-marketplaces, buyers use search engines to specify conditions and to find products that match those conditions exactly. However, some products that may be potentially interesting to buyers are nearly impossible to find. This research aims to design a multi-agent e-marketplace in which buyers and sellers delegate agents to argue about products via an argumentation mechanism. This mechanism enables sellers to persuade buyers to consider purchasing products; it also helps buyers in their search for potentially interesting products. This research adopts Semantic Web technology to express agents' ontologies and proposes an information-gathering approach that supports argumentation. A prototype was developed and a laboratory experiment was conducted to assess the approach's performance. The experimental results show that the proposed mechanism is able to help buyers search for both exactly and potentially interesting products. The study also explores the reasons why searching for potentially interesting products is crucial to e-marketplaces. The proposed prototype and its attendant approach can inspire both existing and initiative e-marketplaces to design advanced product searching and recommendation mechanisms. (C) 2009 Elsevier Ltd. All rights reserved.