Recently a new E-learning application, called Learning Content Management System (LCMS) has been developed as part of a new learning paradigm that focuses on learning needs of individuals. This paper presents a model and a method for discovering and contextualizing browsing patterns from access log database in an LCMS system for personalized learning. The method is based on data mining and statistic techniques to help identify significant browsing events associated with specific contexts. A case study is conducted by applying the method on a log database collected from an on-line course practiced in a web-based LCMS system, called IDEAL. The preliminary results show its potential in supporting personalized content recommendation.