This thesis addressed the shortcomings of existing recommender systems, particularly the cold-start problem and ineffective information distribution encountered in traditional recommendation methods. The primary objective was to develop a hybrid recommender system that combined content-based filtering and collaborative filtering techniques to deliver personalized content for news articles and books effectively. The methodology adopted involved integrating user preferences with historical interaction data to enhance the accuracy and relevance of recommendations. The system utilized a comprehensive database comprising user interaction histories and content metadata. Implementation was carried out using high-level programming languages, specifically Python, along with relevant libraries for data analysis and machine learning. The resulting hybrid system successfully managed the cold-start issue and significantly improved the distribution of personalized information. Its modular architecture facilitated easy maintenance and scalability, ensuring performance stability despite increasing numbers of users and expanding content. Experimental evaluations demonstrated that the hybrid recommender system substantially improved recommendation accuracy and enhanced user satisfaction compared to traditional recommendation methods. This research provided valuable insights into designing adaptive recommender systems with broad implications for various digital content delivery applications.