Evaluating the Utility of Web-Based Consumer Support Tools Using Rough Sets

Many popular e-commerce sites provide decision support tools to assist potential customers. Preliminary research indicates that web usage mining analyses may help to assess the utility of these tools and highlight possible areas for improvement. This paper describes a new procedure for assessing the utility of web-based support tools using techniques in rough set theory. The authors evaluated this procedure in a study of two such support tools, one developed by the US-EPA and the other developed by one of the authors. Results provided interesting insights on the utility of both tools and indicated that both tools could be improved. Details of the new procedure, results obtained from its evaluation, and its implications for future work are described.