Abstract

This paper presents a conceptual based approach for improving a Web site content. Usually Web Usage Mining (WUM) techniques study the visitors' browsing behavior to obtain interesting knowledge. However, most of the work in the area leave behind the semantic information of web pages. We propose to combine the Concept-Based Knowledge Discovery in Text with the visitors sessions to perform the personalization task. This way, it is possible to obtain information about which are the users' goals when browsing a web site. Moreover, it is possible to give better browsing recomendations and help managers improving the content of their Web site. We test this idea on a real Web site to show its effectiveness. © Springer-Verlag Berlin Heidelberg 2006.