Table 2.

Accuracy of the NLP model all categories

Variable name	Spans	NLP model (Gate)			Fine-tuned NLP model (Prodigy)
		Precision	Recall	F1	Precision	Recall	F1
Dosage	75	0.84	1.0	0.91	0.94	0.93	0.93 (0.02)
Route	6	0.79	1.0	0.88	0.40	0.33	0.36 (0.52)
Medication	152	0.69	0.92	0.79	0.90	0.91	0.90 (0.11)
Frequency	88	0.73	0.64	0.68	0.80	0.70	0.75 (0.07)
Duration	60	0.51	0.71	0.59	0.59	0.66	0.62 (0.03)
History	9	1.0	0.40	0.57	0.00	0.00	0.00 (0.57)
Form	24	0.60	0.38	0.46	0.84	0.91	0.86 (0.40)
Diagnosis	33	1.0	0.17	0.29	0.53	0.48	0.50 (0.21)
Assessment scales	6	1.0	0.13	0.22	0.66	0.33	0.44 (0.22)
Symptoms	272	0.93	0.08	0.14	0.65	0.71	0.68 (0.54)
Time info	173	1.0	0.03	0.06	0.73	0.78	0.75 (0.72)
Adverse effects	10	0.00	0.00	0.00	0.25	0.10	0.14 (0.14)
Response	30	0.00	0.00	0.00	0.47	0.30	0.36 (0.36)
TOTAL		0.73	0.43	0.54	0.74	0.73	0 .74 ( 0 .20)

Total precision, recall and F1 are a micro-averaged measures

NLP, natural language processing.