This chapter presents a study applied to the analysis of the utilization of learning Web-based resources in a virtual campus. A huge amount of historical Web log data from e-learning activities, such as e-mail exchange, content consulting, forum participation, and chats is processed using a knowledge discovery approach. Data mining techniques as clustering, decision rules, independent component analysis, and neural networks, are used to search for structures or patterns in the data. The results show the detection of learning styles of the students based on a known educational framework, and useful knowledge of global and specific content on academic performance success and failure. From the discovered knowledge, a set of preliminary academic management strategies to improve the e-learning system is outlined.