Cbr Based Learning in E-auctions

The advent of e-commerce has brought about a radical change in the process of auctions that can be achieved by using agents. One of the important capabilities of agent is learning from the environment. In this paper, the authors are proposing case based learning for agents in online e-auctions. Case-based reasoning (CBR) is a problem solving paradigm based on the principle that similar problems have similar solutions has inherent learning capability. In auctions, CBR has been proposed to store past histories of similar auctions with their solutions which helps agent to learn from past experience. Proposed bidding agent that uses CBR (CBR-agent) participates in the auction and performs better than those bidders who have no past knowledge about similar auctions. Empirical evaluation is done and the performance of CBRagent is calculated by comparing its success percentage and average winning price with that of other agents participating in auctions.