Visual Specification of Layered Bidding Strategies for Autonomous Bidding Agents Benjamin J. Ford, Haiping Xu, Christopher K. Bates Computer and Information Science Department University of Massachusetts Dartmouth, North Dartmouth, MA 02747, USA Email: {u_bford, hxu, u_cbates}@umassd.edu Sol M. Shatz Computer Science Department University of Illinois at Chicago, Chicago, IL 60607, USA Email: shatz@uic.edu Abstract—In an agent-based online auction system, a bidding agent can automatically place bids on behalf of a human user according to a user-specified bidding strategy. Current implementations of bidding agents only support a set of simple predefined bidding strategies. In this paper, we introduce a formal bidding strategy model that supports specification of complex bidding strategies for autonomous bidding agents. The formal model is defined as a layered bidding strategy model (LBSM), which can be represented using notations adapted from UML activity diagrams. For real-time and efficient reasoning, the formal model is converted into a rule-based bidding strategy model (RBSM) represented in bidding strategy language (BSL), which can be directly executed by a reasoning module of an autonomous bidding agent. We present an algorithm for converting an LBSM to a rule-based bidding strategy model, and an algorithm to drive the reasoning engine. Finally, we develop a prototype agent-based online auction system using JADE, and demonstrate how layered bidding strategies can be precisely specified, and how our approach may support analysis of impacts on bidding histories by using different bidding strategies in agent-based online auctions. Index Terms—online auction; software agent; bidding strategy; UML diagram; rule-based model; shill bidder I. INTRODUCTION Online auction houses, such as eBay, have seen an increasing amount of transactions since their debut. As the number of transactions increases, researchers have been investigating the mechanisms and benefits of automating online auction activities. One major form of such automation is agent-based online auctions, which are Internet auctions running partially or entirely through the use of software agents, where software agents can act on behalf of human users, such as buyers, sellers, and auction house administrators [1-3]. In an agent-based online auction system, a bidding agent can automatically place bids on behalf of a human user according to a user-specified bidding strategy [4-6]. A bidding strategy consists of a set of bidding activities and conditions. During an online auction, when certain conditions become true, appropriate bidding activities (e.g., increasing the bid amount or placing a bid) can be automatically performed by the bidding agent. While there have been previous efforts on designing optimal bidding strategies [7-9], work on specifying bidding strategies for bidding agents is more rare. Current implementations of bidding agents only support a set of simple predefined bidding strategies [10-12]. One other strategy specification framework utilizes a logic-based approach [13]; however, that approach lacks the flexibility necessary for specifying large and complex strategies. In order to support user-specified bidding strategies for autonomous bidding agents, there is a pressing need for a feasible way for allowing users to specify bidding strategies that effectively represent the user’s bidding plans. In this paper, we introduce a model-based approach that supports specification of complex and layered bidding strategies for autonomous bidding agents (we use the terminologies of bidding agent and autonomous bidding agent interchangeably in the rest of this paper). Our approach divides a complex strategy into various modular layers. Simple strategies at lower layers can be assimilated into a larger and more complex strategy at a higher layer. For real-time and efficient reasoning, the formal model is converted into a rule-based bidding strategy model represented in bidding strategy language (BSL). Thus the rule-based strategy model can be directly executed by a reasoning module of a bidding agent using a reasoning engine. This work extends our previous research on specification of flexible and complex bidding strategies in agent-based online auctions [14]. In this paper, we further provide formal definitions of our layered bidding strategy model, present the interface of a visual strategy builder (VSB) that supports visual specification of layered bidding strategies for autonomous bidding agents, and analyze new experimental results generated using our approach. Since our approach adapts notations from UML activity diagrams [15-16] for representing bidding Manuscript received August 14, 2009; revised October 9, 2009; accepted October 11, 2009. Corresponding author: Haiping Xu, Email: hxu@umassd.edu 940 JOURNAL OF COMPUTERS, VOL. 5, NO. 6, JUNE 2010 © 2010 ACADEMY PUBLISHER doi:10.4304/jcp.5.6.940-950