Large scale analysis of gender bias and sexism in song lyrics

EPJ Data Science

Table 4 Performance of the sexism classifier on the external dataset for three classification thresholds and \(N_{B}=1\). Metrics for both classes (Sexist and Not sexist) and the corresponding macro average are shown. Right column shows the performance of a naive baseline that always predicts the sexist class

Metric	Class	Classif. threshold			Baseline
Metric	Class	0.50	0.725	0.90	Baseline
Precision	Sexist	0.62	0.68	0.78	0.41
	Non-sexist	0.78	0.79	0.71	0.00
	Macro avg.	0.70	0.73	0.74	0.20
Recall	Sexist	0.70	0.69	0.45	1.00
	Non-sexist	0.71	0.78	0.91	0.00
	Macro avg.	0.70	0.73	0.68	0.50
F1-score	Sexist	0.66	0.68	0.57	0.58
	Non-sexist	0.74	0.78	0.80	0.00
	Macro avg.	0.70	0.73	0.69	0.29