The Internet has emerged as the primary database, and technological platform for electronic business (EB), including the emergence of online retail concerns. Knowledge collection, verification, distribution, storage, and re-use are all essential elements in retail. They are required for decision-making or problem solving by expert consultants, as well as for the accumulation of customers and market knowledge for use by managers in their attempts to increase sales. Previous data mining algorithms usually assumed that input data was precise and clean, this assumes would be eliminated if the best rule for each particular situation. The Algorithm we used in this study however, proved to function even when the input data was vague and unclean. We provided an assessment model of brand trust as an example, to show that the algorithm was able to provide decision makers additional reliable information, in the hope of building a rough set theoretical model and base of resources that would better suit user demand.
ICCCI'10 Proceedings of the Second international conference on Computational collective intelligence: technologies and applications II, pp.501-510