Fig 1.
Classification model for predicting student performance GDO-RBFN.
Fig 2.
Selection of instances.
Fig 3.
The model of RBFN.
Table 1.
Higher education students performance evaluation dataset feature description.
Table 2.
Student performance dataset feature description.
Table 3.
Confusion matrix.
Table 4.
Classify by overall characteristics. The best results are in bold and the second best results are in italics (%).
Table 5.
Classify by personal issues. The best results are in bold and the second best results are in italics (%).
Table 6.
Classified by family issues. The best results are in bold and the second best results are in italics (%).
Table 7.
Classified by educational habits. The best results are in bold and the second best results are in italics (%).
Table 8.
Classify according to overall characteristics after amplification. The best results are in bold and the second best results are in italics (%).
Table 9.
Categorize by personal questions after amplification. The best results are in bold and the second best results are in italics (%).
Table 10.
Categorize by family issues after amplification. The best results are in bold and the second best results are in italics (%).
Table 11.
Expand and classify according to individual educational habits. The best results are in bold and the second best results are in italics (%).
Table 12.
Reliability verification of synthetic data.
Fig 4.
GDO-RBFN is affected by oversampling rate.
Table 13.
Effect of k-value in GDO on the accuracy of classification of educational habitus features, and the best results are in bold (%).