Personalized recommendations for series and movies are an important aspect of online streaming services. This project s objective was to implement and evaluate a series of algorithms used by movie recommendation systems. Four recommendation algorithms were analyzed, two by method of collaborative filtering and two by method of content-based filtering. In addition to the analysis of these, a new hybrid method was developed using two of the priorly analyzed algorithms: a collaborative filtering algorithm based on matrix factorization by singular value decomposition (SVD) and a content-based filtering algorithm using a similarity calculation of textual information between movies. The new method was implemented and analyzed, exhibiting better results than all previous methods. The evaluation data was based on movie ratings given by users of the MovieLens social platform.