Decision Tree with Light Gradient Boosting Algorithm for University Teacher Performance Evaluation

University teacher performance evaluation analyze educators depending on teaching effectiveness, student engagement, and research contributors. It aims to ensure continuous professional development and quality education. Moreover, it contains technology and different teaching approach for engaging students and rapid deeper learning. However, it has challenges of engaging students with different learning styles by managing efficient content delivery. This results in gap at student participation and comprehension that minimize overall learning outcomes. This research proposes Decision Tree with Light Gradient Boosting Algorithm (DT-LGBA) to evaluate the performance for university teachers. DT-LGBA enable for accurate determination of primary factors to teach performance by effectively managing both non-linear and linear relationship which provides reliable evaluation outcomes. DT makes easy interpretation to influence teacher performance because it presents clear decision rules whereas LGBA provides effective in managing intricate non-linear relationships that makes high model performance in teaching. Therefore, this combination makes both clarity and accurate performance in classroom dynamics. The proposed DT-LGBA achieves a high teaching evaluation effect of 96.38% compared to existing techniques like Random Forest (RF) and Support Vector Machine (SVM).