<![CDATA[Student satisfaction with campus facilities and services is an important indicator in assessing the quality of higher education. This study develops a system for classifying student reviews related to facilities and services at PGRI Private Universities using the Bidirectional Long Short-Term Memory (BiLSTM) method. The research data consist of 3,067 student reviews collected through Google Forms and Google Maps, covering five service aspects: Physical Facilities and Core Infrastructure; Academic Support Facilities and Learning Resources; Administrative Services and Staff; Facilities and Environment Supporting Non-Academic Activities; and Security and Accessibility. The BiLSTM method with 300-dimensional FastText word embeddings is employed to classify reviews into service aspect categories and satisfaction levels (Very Satisfied, Satisfied, Moderately Satisfied, Less Satisfied, and Not Satisfied). The model architecture comprises two BiLSTM layers with 128 and 64 units, respectively, along with a dropout mechanism to reduce the risk of overfitting. Model performance is evaluated using a confusion matrix, precision, recall, F1-score, and overall accuracy. The results indicate that the BiLSTM model is able to classify service aspects with good accuracy, although the performance of satisfaction level classification is still affected by data imbalance and the similarity of expressions across satisfaction categories. Overall, the proposed system can provide automated analysis of student reviews and serve as a decision-support tool for universities in understanding service quality based on review data in a more objective and structured manner.]]>
