The add-value of cases on WUM plans recommendation

Abstract

Web Usage Mining is nowadays extremely useful to a diverse and growing number of users, from all types of organizations trying hard to reach the goals of their Web sites. However, inexperienced users, in particular, face several difficulties on developing and applying this kind of mining processes. One crucial and challenging task is selecting proper mining methods to deal with clickstream data analysis problems. We have been engaged on designing, developing and implementing a case based reasoning system, specifically devoted to assist users on knowledge discovery from clickstream data. The system’s main aim is to recommend the most suited mining plans, according to the nature of the problem under analysis. In this paper we present such system, giving emphasis to the retrieving of similar cases using a preliminary constructed case base.