Automated Learning of Pricing and Bundling Strategies in Information Economies

Christopher H. Brooks and Edmund H. Durfee, University of Michigan

The emergence of the Internet and the potential of software agents for conducting electronic commerce presents a new set of challenges for producers of information goods. We discuss how a producer of information goods can learn a price schedule to charge and a set of goods to offer while contending with an unknown and changing consumer population and competition from other producers. We take a decision- theoretic approach, emphasizing the cost of learning and the need for a producer to quickly find an acceptable solution, due to the dynamics of the problem and the small number of potential interactions with a consumer population. We also discuss the use of taxonomic information as a guide when selecting which goods to offer.

This page is copyrighted by AAAI. All rights reserved. Your use of this site constitutes acceptance of all of AAAI's terms and conditions and privacy policy.