Table 1 Statistical significance tests for model comparisons.
Model Comparison | Paired t-test p-value | McNemar’s test p-value | Significant? |
|---|---|---|---|
Proposed vs. CNN | 0.0012 | 0.0008 | Yes (p < 0.01) |
Proposed vs. DT-RF | 0.0007 | 0.0005 | Yes (p < 0.01) |
Proposed vs. BERT | 0.0214 | 0.0186 | Yes (p < 0.05) |
Proposed vs. RoBERTa | 0.0329 | 0.0274 | Yes (p < 0.05) |
Proposed vs. DeBERTa | 0.0412 | 0.0389 | Yes (p < 0.05) |
Proposed vs. GNN | 0.0182 | 0.0153 | Yes (p < 0.05) |
Proposed vs. Multi-Modal | 0.0297 | 0.0235 | Yes (p < 0.05) |
