The development of ICT and AI contributes to the rapid formation of the digital environment in modern organisations. As higher education institutions expand their activities, rely on digital technologies to produce content for social media. Measuring the effectiveness of this content requires assessing its audience engagement. In this study, machine learning, including Random Forest Regressor, Gradient Boosting Regressor, and Support Vector Regression, were applied to assess and predict engagement levels of Social Media in Higher Education Institutions across the Baltic States. The regression models were trained using collected data from the official Facebook pages of universities. Efficiency was measured using the coefficient of determination (R2), mean absolute error (MAE), and root mean square error (RMSE). The results show that machine learning-based prediction is an effective tool for higher education institutions, enabling them to identify the most successful content and thereby attract more followers.