abstract
- This article comprehensively analyzes 70,000 student comments provided to the Faculty of Engineering using advanced Natural Language Processing (NLP) tools. The study aims to uncover valuable insights and practical applications of NLP in education. The findings reveal a nuanced perspective of student sentiment through meticulous analysis and provide actionable information for academic institutions. Key findings from the analysis include: 1. Positive Feedback Dominance: Approximately 60% of the students' comments expressed positivity, reflecting intense overall satisfaction with the Faculty of Engineering. These positive sentiments highlight the institution's commitment to academic excellence and effective teaching practices. 2. Limited Negative Sentiments: Contrasting the positive feedback, only 16% of the comments contained negative sentiments. This result suggests that negative experiences among students are relatively rare. Identifying and addressing these concerns can further enhance the quality of education provided. 3. Unclassified Comments: A notable 15% of comments proved challenging to classify, indicating the need for more advanced NLP techniques or human intervention to extract meaningful insights from this feedback portion. 4. Neutral Sentiments: The remaining comments were predominantly categorized as neutral, signifying that a significant portion of student feedback does not strongly lean toward either positivity or negativity. These comments offer opportunities for fine-tuning educational practices and improving overall student satisfaction. This study underscores the efficacy of NLP in handling large-scale feedback analysis in the academic context. The insights gained from this research are invaluable for institutions seeking to enhance their educational quality, student experiences, and overall performance. Furthermore, it demonstrates the potential for NLP to revolutionize feedback analysis and inform decision-making processes in higher education. © 2024 PICMET.