AI RESEARCH
Class-Dependent Hybrid Data Augmentation for Multiclass Migraine Classification under Severe Class Imbalance
arXiv CS.LG
•
ArXi:2605.23453v1 Announce Type: new We conducted a reproducibility-oriented re-evaluation of prior migraine classification studies, correcting for data leakage and metric bias. We then Correcting methodological flaws reduces previously inflated performance estimates, with the corrected macro-F1 baseline standing at 0.71. The proposed framework consistently outperformed individual augmenters in macro-F1 averaged across the eight evaluated classifiers (0.862 vs.