Web Recommendations are a technique frequently used to improve the navigation and the presentation of data on E-Commerce websites. Many algorithms have been developed in order to automatically generate potentially interesting web recommendations. Each of these algorithms however has its own known drawbacks. In this dissertation we introduce a combined adaptive algorithm, which employs online optimization strategies to overcome these drawbacks. We present a comparative analysis of our approach and several other recommendation approaches using real- world evaluations and show, that our algorithm is more successful in attracting user interest in form of additional clicks and purchases. Based on the gained experience, we extend our approach to the case, when the data presented on a website are integrated from several data sources. In this setting we recognize an additional problem - the problem of data integration: we need to integrate both product data and additional semantic information. A special attention is given to the matching of the product categories. This research work was done within the Database Group at the University of Leipzig, Germany.