内容简介：
With a Foreword by Alfred Kobsa. The research on user modeling has developed a broad range of elaborated algorithms and techniques to support user adaptive applications with information about their users. Several empirical studies demonstrate their benefits in areas like education and training, online help for complex software, dynamic information delivery, etc. The main obstacles to wide-spread use of such systems have always been users' objections and to some extent laws regarding the usage of person-related data. This book offers a comprehensive analysis of security requirements for user modeling systems proceeding from requirements of general information systems and taking into account particular issues of user modeling. Solutions for these requirements are discussed in several ways. Existing solutions in user modeling systems are matched with these requirements. Solutions with a drawback to user modeling are displayed and weighed. New solutions for secrecy and integrity are developed and combined to a reference architecture for security in user modeling. Based on the solutions for encryption, authentication, and authorization, methods for empowering the user to define and enforce his individual requirements towards privacy and anonymity are described in detail. The reference architecture can serve as a default architecture for many user adaptive systems. It offers a modular approach that can adapt to different user requirements and protection goals.