Software systems that adapt their services to characteristics of individual users have already proven to be more effective and/or usable than non-adaptive systems. User-adaptive systems rely on user modeling systems for exhibiting personalized behavior. Quite a few user modeling systems have been developed during the past fifteen years. The decisions as to what useful services/functionalities of these systems are were mostly based on intuition and/or experience gained from studying the literature of a few user-adaptive applications. Results from neighboring disciplines and commercial developments have been largely ignored. Empirical evaluations of the practical applicability of user modeling systems were hardly ever carried out. This book is different: the author takes an interdisciplinary and application-oriented approach, defines meaningful requirements on user modeling servers, gives an overview of existing systems, pinpoints their deficiencies, develops a very novel architecture for user modeling servers, implements it, and tests its utility both within an application project and in empirically founded performance experiments. His excellent synthesis of scientific and industrial concerns (which rests on research in data bases, distributed systems, human-computer interaction, user modeling, statistics, and e-commerce) and his very convincing solutions make this book a worthwhile reading both for researchers and for industrial practitioners.