Table 7. Comparative performance metrics.

Method	Accuracy (%)	Precision (%)	Recall (%)	F1-Score (%)	Specificity (%)	AUC-ROC Score
Proposed Model (Deep Learning + STFT)	92.9	91.0	95.3	93.1	90.6	0.95
Random Forest (Traditional ML) [28]	74.0	83.0	63.0	72.0	86.0	0.86
AdaBoost (Traditional ML) [28]	74.0	75.0	74.0	75.0	75.0	0.75
CNN (Convolutional Neural Networks) [10]	85.0	87.0	80.0	83.0	82.0	0.88
Attention-Based Model (Hybrid DL) [11]	89.0	90.0	85.0	87.5	88.0	0.91
Ensemble Learning Model (SAMP) [Feng et al., 2024] [50]	88.5	89.1	87.8	88.4	87.0	0.90
StackDPPred (Ensemble Learning) [Arif et al., 2024] [51]	90.2	91.4	88.9	90.1	89.3	0.92