Predicting the Ranking of Engineering Mechanics Students Using the Bayesian Model

Abstract

GPA plays an important role in the entire learning process of students. The value of GPA not only reflects students’ current grades but also affects their future progress, motivation, and opportunities. It is worth noting that the grades of specific courses also have an impact on GPA. Therefore, it is necessary to predict students’ performance in future courses based on their current grades. In this paper, a Bayesian model is employed to classify course grades and estimate the probability of these grades being affected by other factors in the first semester, enabling the prediction of subsequent performance. The Bayesian approach integrates prior knowledge of grade distributions through four key steps: establishing a prior probability distribution, using a likelihood function to relate grades to academic ability, combining prior and new evidence to compute posterior probabilities, and forecasting next-semester results. These predictions support timely academic interventions and adjustments to teaching strategies. By utilizing data such as assignment and exam scores, a Bayesian classification model can analyze and predict outcomes. The actual grades of students in the second semester are used to validate the predictive accuracy of the model.