The increase in Single Tuition Fee (UKT) has become increasingly concerning for many prospective students and parents in Indonesia. The significant rise in UKT has led many prospective students to decide not to pursue higher education, resulting in a decline in college enrollment interest. This study aims to analyze the impact of the UKT increase on the declining interest in higher education among prospective students using the Long Short-Term Memory (LSTM) algorithm in Machine Learning. Based on sentiment data from YouTube and Instagram regarding the UKT increase issue, the LSTM model is implemented to predict college enrollment interest based on changes in UKT. The results of this study indicate a significant correlation between the UKT increase and the decline in new student enrollment. The Long Short-Term Memory (LSTM) algorithm model effectively predicts the impact of tuition fee increases on prospective students' interest in higher education. This is supported by an accuracy of 84%, showcasing the model's ability to recognize historical patterns in the data used optimally.