الفهرس | Only 14 pages are availabe for public view |
Abstract The Internet-based auction is a dynamic branch of E-Commerce, However, the lack of reputation and trust systems due to the anonymity and misrepresentation of the relation among the participants. This causes many features of fraud that occurs in three period times of the online auction process. Hence in this thesis we propose a web-based multi-agent system platform for online auction that allows detecting the shill bidders, develop the rules of key performance indicators ( KPIs) to make a detective formula for shill bidder during the auction process, apply the proposed model on a dataset of the information of the users and their historical transactions and present a prototype of the proposed multi-agent system. Currently, the proposed model is a multi-agent which provides a reliable, scalable and easy to manage the environment to decrease the shill bidding during bidding process by detecting malicious users based on their historical transactions. Moreover, we use the KPIs rules to create a fraud rate formula can detect the shill bidder according to their historical transactions. We use the analytical hierarchal process (AHP) technique to calculate the weights of the concerned parameters in the KPI’s formula. |