The movie recommender model have become integral components of numerous online retail platforms. Recommendation systems personalize user experiences by helping them discover their preferred content within the vast ocean of media available online. However, a significant hurdle in achieving this personalization lies in capturing the evolving characteristics of user preferences over time. This chapter introduces content-oriented movie recommendation system, aiming to grasp the evolving user preferences over time through user modeling and anticipate their favored films. The method suggested offers a user-centric approach, integrating the content characteristics of rated movies (for each user) into a flexible clustering model. By incorporating the Dirichlet process mixture model (DPMM), this function enhances the process of inferring user preferences. This refined understanding of user tastes allows for the generation of highly relevant recommendation lists. To assess the efficacy of this function, our model implemented this proposed method and utilized the TMDB 5000 Movies dataset for our experiments. The results demonstrate that this recommendation system performs competitively, achieving superior results compared to existing methods.