In the beginning of the work we explain some of the reasons why are recommender systems necessary in the world we live today, as well, we explain field of machine learning since methods that stems from the field of machine learning are used by recommender systems in order to resolve problems that are encounter. First we explain recommender systems whose recommend is based on content, whilst as we continue with work we are more into explaining collaborative filtering recommender system whose recommend is based on collaboration, that is, common actions among users which makes recommender system better. As collaborative filtering recommender systems are used more and more in real world, we explain them in more detail. As we have to explain system recommender that are based on collaborative filtering, we have to explain algorithms such as linear regression and gradient descent, as well we explain some of the problems that algorithms encounter. Practically, algorithms are explained through recommender system that recommend movies. We explain how we can get top-n recommends for particular user, how to find similar movies and how to solve problem of new user. In the chapter that follows, we explain evaluation methods for recommender systems. In the end of the work, we explain program code that is used in order to implement recommender system that is explained in third heading, as well, we explain evaluation of the implemented system.