Qikai Lin

Educational records alone may not capture changes in students’ classroom affective states. This study proposes Bi-LSTM-SCPM, a student comprehensive performance model (SCPM) integrating bidirectional long short-term memory (Bi-LSTM) with mixture of experts (MoE) regression. The dataset comprised 14,645 facial images from 71 students across three stages of 12-week computer-assisted translation courses. ResNet50 achieved 93.8% accuracy in positive–negative facial state classification. Bi-LSTM integrated stage scores derived from instructor-designated key periods and Bayesian smoothing into a student-level affective score (ZHCJ). ZHCJ was combined with educational information and prerequisite course performance to predict final grades. Across 30 matched student-level splits, MoE achieved an R² of 0.943 ± 0.002 and an MAE of 1.144 ± 0.128 points, compared with 0.927 ± 0.002 and 1.274 ± 0.149 points for XGBoost. Both differences remained significant after Holm adjustment (p < 0.001). ZHCJ also outperformed summary statistics and direct stage score inputs under both predictors. The setting ϖ = 2 and k = 20 performed best within the nine-setting grid. These findings support the predictive value of integrating temporal affective representations and educational information within the study cohort.

Predicting academic performance in face-to-face computing and software engineering courses is hindered by the limited availability of fine-grained process data. This study proposes STEAP, a camera-based static–temporal fusion framework that integrates longitudinal classroom behavior sensing with conventional educational records. Classroom videos from 375 undergraduates enrolled in four computing-related courses were collected over nine teaching weeks. Camera-derived observable behaviors were organized into student-level weekly sequences and transformed into outcome-independent longitudinal representations. Multiple machine-learning classifiers were subsequently applied to predict students’ academic performance. Checkpoint-specific predictions were conducted at Weeks 3, 6, and 9, with each prediction using only the classroom behavioral information available up to the corresponding time point. Using the complete nine-week Temporal representation together with the pre-course Background variables, XGBoost achieved the strongest classification performance among the evaluated models, with an Accuracy of 0.867, a Macro F1 of 0.862, and an At-risk Recall of 0.924. The checkpoint analyses further indicated that classroom behavioral information collected during the early course stage already provided useful predictive information without incorporating behavioral observations from subsequent weeks. After further integrating pre-course background variables and regular assessment information, the final fusion model achieved an Accuracy of 0.896 and a Macro F1 of 0.895. Overall, longitudinal camera-derived classroom behavior provides complementary predictive information beyond conventional educational information and supports the feasibility of earlier academic-risk identification at different course checkpoints.