1. 09/28/2007 CIS Dept., UMass Dartmouth 1 Trustworthy Agent-Based Online Auction Systems Prof. Haiping Xu Concurrent Software Systems Laboratory Computer and Information Science Department University of Massachusetts Dartmouth

2. 09/28/2007CIS Dept., UMass Dartmouth2 Online Auctions Different types of auctions Increase-price auction (English auction) Decrease-price auction (Dutch auction) Second-price sealed-bid auction (Vickrey auction) English auction has become the most popular one in online auction houses (e.g., eBay). However, it is time-consuming for a human user to search and place bids on an auctioned item. There is a pressing need to introduce agent technology into online auction systems.

3. 09/28/2007CIS Dept., UMass Dartmouth3 Agent-Based Online Auction System Client Auction House (Server) GUI Agent Auction Agent Security Agent Main Agent Search Agent Selling/Bidding Agent Database It consists of an auction house and a number of clients. It is designed as a multi-agent system. The auction house is managed by auction house administrator. Agents at the client side work on behalf of human users. Security agent monitors online auction transactions for any undesired bidding activities, e.g., shilling behaviors.

4. 09/28/2007CIS Dept., UMass Dartmouth4 Shilling Behaviors A shill bidding is a deliberate activity of placing bids in order to artificially raise the price of an auctioned item. Although most of the online auction houses prohibit shilling behaviors, it is easy for malicious users to disguise themselves and put in shill bids in online auctions. According to a recent research at Carnegie Mellon University, dozens of probable fraudsters were detected at eBay using data mining techniques. It is vital for us to introduce a feasible trust management mechanism to prevent, detect and avoid trading frauds, such as shilling behaviors.

5. 09/28/2007CIS Dept., UMass Dartmouth5 An Example We call this type of shilling behavior concurrent shilling. Other types of shilling behaviors include: reserve price shilling, competitive shilling etc. Shilling behaviors become much more server in an agent- based online auction system because Detection of shill bidders can be much more difficult. Malicious users may set up bidding strategies and automatically initiate shilling activities using agent technology. While two auctions with the same type of auctioned items are running concurrently, a shill bidder might put bids in the auction with higher bidding price rather than the one with lower bidding price in order to drive up the price in one auction.

6. 09/28/2007CIS Dept., UMass Dartmouth6 Trust Management Trust and reputation management has been a promising approach to building trustworthiness in networked systems. Two major types of trust management approaches Reputation-based trust management (e.g., in eBay) Policy-based trust management (e.g., R EFEREE, KeyNote). Our approach is a combined approach, which Considers agent reputations stored in a history module. Adopts role-based access control (RBAC) mechanism based on a set of policy rules. More importantly, considers user’s real-time behaviors in agent-based online auctions.

7. 09/28/2007CIS Dept., UMass Dartmouth7 Related Publications H. Xu and Y-T Cheng Model Checking Bidding Behaviors in Internet Concurrent Auctions. International Journal of Computer Systems Science & Engineering (IJCSSE), July 2007, Vol. 22, No. 4, pp. 179-191. R. Patel, H. Xu, and A. Goel Real-Time Trust Management in Agent Based Online Auction Systems. Proceedings of the19th International Conf. on Software Engineering and Knowledge Engineering (SEKE'07), Boston, USA, July 2007, pp. 244-250. Y-T Cheng and H. Xu A Formal Approach to Detecting Shilling Behaviors in Concurrent Online Auctions. Proceedings of the 8th International Conf. on Enterprise Information Systems (ICEIS 2006), May 2006, Paphos, Cyprus, pp. 375-381. Contact Information Haiping Xu, Assistant Professor Computer and Information Science Department University of Massachusetts Dartmouth Phone : (508) 910-6427 Email: hxu@umassd.eduhxu@umassd.edu Sol M. Shatz, Professor (Collaborator) Computer Science Department University of Illinois at Chicago Phone : (312) 996-5488 Email: shatz@cs.uic.edushatz@cs.uic.edu This project was supported by the Chancellor’s Research Fund and UMass Joseph P. Healey Endowment Grants, and the U.S. National Science Foundation under grant number CNS-0715648.