In this study, we aim to predict user engagement on Facebook using machine learning algorithms. We collected data from Facebook users, including attributes such as userid, age, dob_day, dob_year, dob_month, gender, tenure, friend_count, friendships_initiated, likes, likes_received, mobile_likes, mobile_likes_received, www_likes, and www_likes_received. We trained six different machine learning algorithms, including Naive Bayes, Logistic Regression, Random Forest, K-Neighbors Classifier, Extreme Gradient Boost, and Modified Extreme Gradient Boost, to predict user engagement. We evaluated the performance of each algorithm using various performance metrics such as accuracy, precision, recall, and F1-score. Our results show that the Modified Extreme Gradient Boost algorithm outperforms all other algorithms in terms of accuracy, with an accuracy score of 0.92.