. 2022 Sep 21;5(3):e33460. doi: 10.2196/33460

Table 2.

Machine learning classification results of models trained on automatic transcripts compared to results of models trained on manually corrected transcripts.

Task and model type		Automatic transcripts AUROC^a	Manually corrected transcripts AUROC	Change in AUROC^b
Picture description
	RF^c	0.617	0.687	0.070^d
	GNB^e	0.662	0.725	0.063^d
	LR^f	0.671	0.743	0.072^d
	BERT^g	0.618	0.686	0.068^d
Experience description
	RF	0.503	0.636	0.133^d
	GNB	0.549	0.677	0.128^d
	LR	0.543	0.674	0.131^d
	BERT	0.630	0.650	0.020^d

^aAUROC: area under the receiver operating characteristic curve.

^bPositive change in AUROC indicates that the manually corrected transcript model outperformed the automatic transcript model.

^cRF: random forest.

^dIndicates P<.001.

^eGNB: Gaussian naive Bayes.

^fLR: logistic regression.

^gBERT: Bidirectional Encoder Representations from Transformers.