Finding information on a large web site can be a difficult and time-consuming process. Recommender systems can help users find information by providing them with personalized suggestions. In this paper the creation of recommendation system was emphasized to achieve the personalization on the website. Recommender systems typically use techniques from collaborative filtering, in which proximity measures between users are formulated to generate recommendations, or content-based filtering, in which users are compared directly to items. Our approach used similarity measures between users. User-based collaborative filtering gave personalized recommendations by finding similar users. Item-Based collaborative filtering recommended similar items. Different algorithms were compared. However, the applied testing procedure did not employ equal conditions for both approaches. The aim of this report was to give an evaluation on both the approaches by employing a fair testing procedure on the data gathered. Test results and their dependency to the employed algorithms were interpreted. The experiments are carried out by building the recommendation engine through the Taste library in Java — a fast and flexible engine for collaborative filtering.