Appendix 2—table 1. Summary of the Validation Statistics.

	First validation dataset (n = 420)								Second validation dataset (n = 100)
Reference Type	True positive	False positive	False negative	True negative	Ratio	Precision	Recall	F0.5 score	True positive	False negative
Original	80	6	7	11	0.31	0.93	0.92	0.92	38	1
Pubmed	37	1	5	6	0.30	0.97	0.88	0.92	5	1
DOI	115	6	0	6	0.19	0.95	1.00	0.97	43	1
Title	38	2	0	16	0.12	0.95	1.00	0.97	7	0
Split	8	0	15	27	0.08	1.00	0.35	0.52	4	0
Publishing details	1	0	1	32	0.01	1.00	0.50	0.67	0	0
Total	279	15	28	98	1.00	-	-	-	97	3
Corrected total	-	-	-	-	-	0.96	0.89	0.91	-	-