This study used Natural Language Processing (NLP) and Artificial Intelligence (AI) to analyze ten years of teacher evaluations. The research leveraged VADER and NRC for sentiment analysis and LDA for topic modeling to extract key themes. The Google Gemini AI model then generated actionable recommendations for pedagogical improvement. Analysis of 9,052 textual comments revealed a predominantly positive (71%) to neutral (27%) comments, and LDA identified eight distinct topics. The AI-driven analysis successfully provided targeted suggestions for pedagogical enhancement, offering a pathway toward data-informed professional growth for educators. However, multilingual feedback presented challenges for comprehensive analysis.
Article: AI-Driven Insights from Student Feedback for Teacher Improvement (2025 research brief)
Key point: Using AI to analyse evaluation data helps extract actionable, structured feedback that informs pedagogical improvement and professional growth.
Quote: AI analysis provided targeted recommendations for pedagogical enhancement based on sentiment and topic analysis of evaluation comments.
🔗 Note: Content from ResearchGate summary (link may require access). https://www.researchgate.net/publication/393666660_AI-Driven_Insights_from_Student_Feedback_for_Teacher_Improvement