Fig 1.
Overall workflow of math item classification system architecture.
SVM: support vector machine, SGD: stochastic gradient descent, XGBoost: extreme gradient boosting, AUC-ROC: Area under the receiver operating characteristic curve, LMS: Learning management system.
Table 1.
Samples of final master table and main variables.
Table 2.
Optimal parameter value results obtained through grid-search cross-validation.
Table 3.
Data on students.
Table 4.
Indices for statistical t-test analysis results.
Fig 2.
Evaluation metrics of model performances.
Fig 3.
AUC-ROC curve for each difficulty level, namely “high”, “medium”, and “low”.
Table 5.
Comparison of evaluation indicators of different machine-learning models.