Evaluating Project-Based Teaching Reform in Heat Transfer Courses Using Logistic Regression, Random Forest, and Gradient Boosting

We present a framework for measuring how well chemical‑engineering programs work, especially when they use project‑based learning in heat‑transfer courses. The data come from student evaluations of an inverse reheating furnace project. Using supervised machine learning, we classify student satisfaction and performance. We built three predictive models—Logistic Regression, Random Forest, and Gradient Boosting—and compared them. The models predict student views on engineering authenticity, knowledge integration, programming practice, and decision‑making. The process includes cleaning the data, turning Likert‑scale responses into features, training the models, and tuning their hyperparameters. We assess performance with Accuracy, Weighted F1‑score, Precision, Recall, and Area Under the ROC Curve (AUC). By combining predictive modeling with standard statistical

analysis, the project offers a clear way to measure the impact of teaching reforms in engineering education. The resulting analytical pipeline gives curriculum designers practical insights to improve heat‑transfer instruction and boost student engagement with real‑world thermal problems.

Message: Hi FE, I need help with “Evaluating Project-Based Teaching Reform in Heat Transfer Courses Using Logistic Regression, Random Forest, and Gradient Boosting” in “Chemical Engineering”