BMC Medical Informatics and Decision Making

Table 2 NER models results

From: Automatic data extraction to support meta-analysis statistical analysis: a case study on breast cancer

(a) BioBERT model results
	BioBERT			BioBERT_split
Sub-category	Precision	Recall	F1	Precision	Recall	F1
Total-participants	0.95	0.95	0.95	0.94	0.94	0.94
Intervention-participants	0.80	0.91	0.85	0.78	0.93	0.85
Control-participants	0.87	0.91	0.89	0.85	0.91	0.88
Age	0.66	0.97	0.79	0.66	0.96	0.78
Eligibility	0.75	0.77	0.76	0.77	0.74	0.76
Ethnicity	0.82	0.89	0.86	0.82	0.96	0.88
Condition	0.86	0.81	0.84	0.84	0.75	0.79
Location	0.75	0.85	0.80	0.73	0.81	0.77
Intervention	0.85	0.82	0.84	0.85	0.82	0.84
Control	0.78	0.80	0.79	0.77	0.76	0.77
Outcome	0.82	0.81	0.81	0.84	0.80	0.82
Outcome-measure	0.79	0.90	0.84	0.81	0.88	0.84
bin-abs-iv	0.75	0.78	0.77	0.81	0.78	0.79
bin-abs-cv	0.79	0.87	0.83	0.77	0.80	0.79
bin-percent-iv	0.87	0.88	0.87	0.83	0.86	0.84
bin-percent-cv	0.88	0.90	0.89	0.87	0.82	0.84
cont-mean-iv	0.78	0.90	0.83	0.80	0.86	0.83
cont-mean-cv	0.86	0.86	0.86	0.81	0.84	0.83
cont-median-iv	0.70	0.80	0.75	0.70	0.86	0.78
cont-median-cv	0.76	0.81	0.78	0.83	0.74	0.78
cont-sd-iv	0.68	0.93	0.79	0.80	0.85	0.82
cont-sd-cv	0.76	0.84	0.80	0.72	0.85	0.78
cont-q1-iv	0.00	0.00	0.00	0.00	0.00	0.00
cont-q1-cv	0.00	0.00	0.00	0.00	0.00	0.00
cont-q3-iv	0.00	0.00	0.00	0.00	0.00	0.00
cont-q3-cv	0.00	0.00	0.00	0.00	0.00	0.00

(b) BlueBERT model results
	BlueBERT			BlueBERT_split
Sub-category	Precision	Recall	F1	Precision	Recall	F1
Total-participants	0.94	0.91	0.92	0.95	0.92	0.94
Intervention-participants	0.72	0.90	0.80	0.73	0.91	0.81
Control-participants	0.81	0.85	0.83	0.79	0.89	0.84
Age	0.67	0.97	0.79	0.66	0.97	0.79
Eligibility	0.73	0.74	0.73	0.73	0.70	0.72
Ethnicity	0.90	0.72	0.80	0.91	0.78	0.84
Condition	0.90	0.70	0.79	0.82	0.77	0.79
Location	0.77	0.67	0.71	0.76	0.76	0.76
Intervention	0.80	0.81	0.81	0.84	0.83	0.83
Control	0.72	0.68	0.70	0.78	0.71	0.74
Outcome	0.81	0.79	0.80	0.81	0.80	0.80
Outcome-measure	0.73	0.84	0.78	0.76	0.86	0.81
bin-abs-iv	0.77	0.75	0.76	0.67	0.76	0.71
bin-abs-cv	0.75	0.79	0.77	0.72	0.84	0.78
bin-percent-iv	0.74	0.85	0.79	0.79	0.81	0.80
bin-percent-cv	0.83	0.73	0.78	0.82	0.79	0.80
cont-mean-iv	0.72	0.74	0.73	0.61	0.81	0.69
cont-mean-cv	0.77	0.74	0.75	0.73	0.76	0.74
cont-median-iv	0.65	0.78	0.71	0.67	0.62	0.64
cont-median-cv	0.80	0.66	0.72	0.75	0.66	0.70
cont-sd-iv	0.62	0.68	0.65	0.59	0.60	0.59
cont-sd-cv	0.67	0.68	0.67	0.56	0.70	0.63
cont-q1-iv	0.00	0.00	0.00	0.00	0.00	0.00
cont-q1-cv	0.00	0.00	0.00	0.00	0.00	0.00
cont-q3-iv	0.00	0.00	0.00	0.00	0.00	0.00
cont-q3-cv	0.00	0.00	0.00	0.00	0.00	0.00

(c) Longformer model results
Sub-category	Precision	Recall	F1
Total-participants	0.96	0.94	0.95
Intervention-participants	0.79	0.92	0.85
Control-participants	0.89	0.89	0.89
Age	0.78	0.98	0.87
Eligibility	0.89	0.86	0.88
Ethnicity	0.75	0.83	0.78
Condition	0.83	0.79	0.81
Location	0.91	0.79	0.85
Intervention	0.86	0.85	0.86
Control	0.81	0.86	0.83
Outcome	0.85	0.86	0.86
Outcome-measure	0.85	0.95	0.90
bin-abs-iv	0.83	0.83	0.83
bin-abs-cv	0.84	0.85	0.84
bin-percent-iv	0.85	0.90	0.88
bin-percent-cv	0.88	0.85	0.87
cont-mean-iv	0.85	0.87	0.86
cont-mean-cv	0.78	0.91	0.84
cont-median-iv	0.65	0.76	0.70
cont-median-cv	0.75	0.76	0.75
cont-sd-iv	0.83	0.86	0.85
cont-sd-cv	0.77	0.92	0.84
cont-q1-iv	0.00	0.00	0.00
cont-q1-cv	0.00	0.00	0.00
cont-q3-iv	0.00	0.00	0.00
cont-q3-cv	0.00	0.00	0.00

Bold texts represent the best score for each sub-category

Back to article page

ISSN: 1472-6947

Contact us

General enquiries: journalsubmissions@springernature.com