Addressing the Limitations of News Recommendation Systems: Incorporating User Demographics for Enhanced Personalization

News recommendation schemes use features of the news and user information to suggest relevant news items based on their interests. However, existing schemes are not very effective for new users who have little activity history. To address this issue, we have developed a hybrid news recommender system that combines content-based and collaborative filtering approaches, along with user demographic data. We conducted an extensive experiment using a dataset of news articles, user ratings, and demographic data to evaluate the system's effectiveness. We assessed the system's performance through two experiments. The first experiment focused on individual user similarity and achieved a precision of 68.05%, a recall of 42.46%, and an average F1 score of 52.1%. The second experiment examined user similarity within the same category, resulting in a precision of 93.75%, a recall of 40.25%, and an F1 score of 56.31%. These results indicate that the proposed model performs better in the second experiment compared to the first.