Classifier-Dependent Benefits of Pseudo-Labeling for Semi-Supervised Android Malware Attribution
A systematic evaluation of pseudo-labeling across six classifiers on the CICMalDroid 2020 dataset, using five-fold stratified cross-validation and paired t-tests across five labeled ratios, finds that SSL benefit is strongly classifier-dependent.
Md Rafid Islam, Zahid Hasan, Hafizah Rahman
· 0 citations