To deal with the class imbalance problem in multi-label learning with missing labels, we propose Class Imbalance aware Missing labels Multi-label Learning, CIMML. Our proposed method handles class imbalance issue by constructing a label weight matrix with weight estimation guided by how frequently a label is present, absent, and unobserved.
class-imbalance
multi-label-learning
missing-labels
missing-label-aware-label-weights
label-correlation
-
Updated
Jan 28, 2024 - MATLAB