Recommender systems have been gaining popularity and appreciation over the past few years and they kept growing towards a semantic web. Internet users search for more and more facilities to get information and recommendations based on their preferences, experience and expectations. Nowadays, there are many recommender systems on the web for music, movies, diets, products, etc. Some of them use very efficient recommending techniques (ex. Amazon), while others are very simple, based on algorithms that do not always provide relevant or interesting recommendations. The solution we propose is a recommender system for running professionals and amateurs, which is able to provide information to users regarding the workout and the diet that best suits them, based on their profile information, preferences and declared purpose. The solution mixes a social dimension derived from an expanding community with expert knowledge defined within an ontology. Moreover, our model addresses adaptability in terms of personal profile, professional results and unfortunate events that might occur during workouts.
