Modeling information quality in agent-based e-commerce systems

Abstract

In agent-based e-commerce systems, most buying agents rely primarily on ranking algorithms to make purchase decisions. These algorithms are based on one or a combination of several features of goods information, such as price, seller's reputation, etc.; however, information quality is usually ignored. This leads to the problem that product information is retrieved without proper regard to their quality. Realizing the importance and necessity of information quality for buying agents, this thesis introduces an Information Quality (IQ)-based ranking approach. A model which combines information quality assessment with user expectation is proposed. An approach on representation and interpretation of information quality is also developed. Two sets of experiments are conducted to verify the applicability of the proposed model and explore the effectiveness of individual information quality dimensions. The experimental results show that information quality based ranking does lead to better user satisfaction. Keywords: Information quality, agent system, e-commerce