Table 7.

Overall performance of all runs in Task 1—Named entity recognition where the set of high-level labels in Figure 7 is used.

Run	Exact-match			Relaxed-match
	P	R	F	P	R	F
Melaxtech-run1	0.9774	0.9774	0.9774	0.9906	0.9901	0.9903
Melaxtech-run2	0.9789	0.9732	0.9760	0.9910	0.9849	0.9879
Melaxtech-run3	0.9775	0.9714	0.9744	0.9905	0.9838	0.9871
VinAI-run2*	0.9704	0.9670	0.9687	0.9920	0.9901	0.9911
Lasige_BioTM-run1	0.9571	0.9706	0.9638	0.9886	0.9943	0.9915
VinAI-run1	0.9635	0.9579	0.9607	0.9899	0.9854	0.9877
Baseline	0.9657	0.9288	0.9469	0.9861	0.9519	0.9687
BiTeM-run1	0.9573	0.9277	0.9423	0.9907	0.9770	0.9838
NextMove/Minesoft-run2	0.9460	0.9330	0.9394	0.9773	0.9611	0.9691
NextMove/Minesoft-run1	0.9458	0.9330	0.9393	0.9773	0.9610	0.9691
BiTeM-run2	0.9323	0.9357	0.9340	0.9845	0.9962	0.9903
NextMove/Minesoft-run3	0.9201	0.8970	0.9084	0.9571	0.9308	0.9438
NLP@VCU-run1	0.9016	0.8835	0.8925	0.9855	0.9814	0.9834
NLP@VCU-run2	0.9007	0.8799	0.8902	0.9882	0.9798	0.9840
NLP@VCU-run3	0.8960	0.8805	0.8882	0.9858	0.9869	0.9863
KFU_NLP-run1	0.9125	0.8570	0.8839	0.9911	0.9465	0.9683
BiTeM-run3	0.9073	0.8496	0.8775	0.9894	0.9355	0.9617
KFU_NLP-run2	0.8735	0.8481	0.8606	0.988	0.9569	0.9722
KFU_NLP-run3	0.8332	0.8160	0.8245	0.9789	0.9516	0.9651
OntoChem-run1	0.9029	0.6796	0.7755	0.9611	0.7226	0.8249
AUKBC-run1	0.7542	0.4544	0.5671	0.9833	0.5977	0.7435
AUKBC-run2	0.6605	0.2581	0.3712	0.9290	0.3612	0.5201
SSN_NLP-run2	0.3174	0.2084	0.2516	0.9491	0.5324	0.6822
SSN_NLP-run1	0.3179	0.2076	0.2512	0.9505	0.5304	0.6808
JU_INDIA-run1	0.2019	0.1180	0.1489	0.5790	0.3228	0.4145
JU_INDIA-run2	0.0557	0.0262	0.0357	0.4780	0.2149	0.2965
JU_INDIA-run3	0.0557	0.0262	0.0357	0.4780	0.2149	0.2965

Here, P, R, and F represents the Precision, Recall, and F₁-score, respectively. For each metric, we highlight the best result in bold and the second best result in italic. The results are ordered by their performance in terms of F₁-score under exact-match.

^*This run was received after evaluation phase and thus was not included in official results.