Equivalence Matching Results for SNOMED-NCIT (Neoplas)
 
Unsupervised (No Train Maps)
Semi-supervised (Optional 30% Train Maps)
⇕System ⇕Precision ⇕Recall ⇕F-score ⇕MRR ⇕Hits@1 ⇕Use Train Maps ⇕Precision ⇕Recall ⇕F-score ⇕MRR ⇕Hits@1
AMD 0.836 0.481 0.610 0.792 0.528 0.633
BERTMap 0.557 0.762 0.643 0.954 0.928 0.562 0.771 0.650 0.962 0.938
BERTMapLt 0.831 0.687 0.752 0.891 0.859 0.775 0.688 0.729 0.891 0.859
LogMap 0.870 0.586 0.701 0.823 0.583 0.683
LogMapBio 0.748 0.795 0.771 0.675 0.793 0.729
LogMapLt 0.951 0.517 0.670 0.931 0.514 0.662
Matcha 0.838 0.551 0.665 0.782 0.545 0.642
Matcha-DL 0.811 0.514 0.629 0.829 0.806 0.806 0.714 0.757 0.829 0.806
OLaLa 0.540 0.546 0.543 0.451 0.545 0.493
SORBETMtch†∗ 0.626 0.642 0.634 0.731 0.605 0.662

Note: Mixed results for SORBETMtch, where unsupervised results come from MELT and semi-supervised results come from direct result submission.