Table 8. A comparison the accuracy, precision, recall, and F1 of Criminal charge prediction using BERT and XLNET models.

Model	Accuracy	$\mathrm{P}\mathrm{r}\mathrm{e}\mathrm{c}\mathrm{i}\mathrm{s}\mathrm{i}\mathrm{o}\mathrm{n}$	$\mathrm{R}\mathrm{e}\mathrm{c}\mathrm{a}\mathrm{l}\mathrm{l}$	$\mathrm{F}1$
XLNET	98.16%	98.17%	98.16%	98.16%
BERT	97.53%	97.53%	97.53%	97.53%
Two chunking BERT	98.5%	98.5%	98.5%	98.5%
Three chunking BERT	98.95%	98.95%	98.95%	98.95%