Artificial Intelligence-Based Approach for Misogyny and Sarcasm Detection from Arabic Texts

Table 8

Detection evaluation results of Arabic sentiment compared with different classifiers: multiclass scenario.

Classifier model	Accuracy (%)	Precision (%)	Recall (%)	F1-score (%)
PAC	71	68	71	69
LRC	73	73	73	68
RFC	68	46	66	55
LSVC	73	69	73	70
DTC	63	62	63	63
KNNC	68	59	68	57
ARABERT	77	—	—	75

The bold values represent the highest values of evaluation metrics achieved by the corresponding classifier model.