Multi-label F1
A common metric for evaluating multi-label classification that computes precision and recall for each label and then averages them (macro-F1) or averages across samples (micro-F1); it treats all errors alike and cannot reflect semantic differences.
Timeline (last 90 days)
Existing affective computing benchmarks rarely evaluate social signals that co-occur in conversations, and traditional multi-label F1 ignores semantic associations and annotation disagreement.
Multi-label F1 computes precision and recall for each label separately and then takes the average (macro-F1) or averages over samples (micro-F1), but treats all errors alike and cannot reflect differences in semantic hierarchy.
All Facts (2)
Existing affective computing benchmarks rarely evaluate social signals that co-occur in conversations, and traditional multi-label F1 ignores semantic associations and annotation disagreement.
50%UnverifiedMulti-label F1 computes precision and recall for each label separately and then takes the average (macro-F1) or averages over samples (micro-F1), but treats all errors alike and cannot reflect differences in semantic hierarchy.
50%