The problem of modeling and predicting Web users' browsing pattern has gained increasing attention. In this book, we present our methods for clustering and making recommendations to Web users and the applications to a real dataset generated by a Web-based knowledge management system, Livelink. The problem of clustering Web users and access sequences presents two unique challenges: the immense volume of data and the sequentiality of user navigation patterns. We propose to model user access sequences as stochastic processes, and a Mixture of Markov Model (MMM) based approach is taken to capture the sequential relationships inherent in user access histories. Several important issues that arise in constructing the model are addressed. The first issue lies in the complexity of the MMMs. To improve the efficiency of building/maintaining the models, we develop a light-weight adaptive algorithm to update the model parameters without evoking overhaul computations. The second issue involves the proper selection of training data. We investigate two different training data selection strategies and perform extensive experiments to compare their effectiveness in our application domain.