Equivalence Matching Results for NCIT-DOID (Disease)
 
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 0.885 0.691 0.777 0.858 0.770 0.811
BERTMap 0.888 0.878 0.883 0.959 0.937 0.831 0.883 0.856 0.960 0.938
BERTMapLt 0.919 0.772 0.839 0.890 0.861 0.888 0.770 0.825 0.890 0.861
LogMap 0.934 0.668 0.779 0.908 0.664 0.767
LogMapBio 0.860 0.962 0.908 0.811 0.959 0.879
LogMapLt 0.983 0.575 0.725 0.976 0.575 0.723
Matcha 0.882 0.756 0.814 0.839 0.750 0.792
Matcha-DL 0.847 0.586 0.693 0.870 0.844 0.847 0.834 0.841 0.870 0.844
OLaLa 0.913 0.864 0.888 0.880 0.861 0.870
SORBETMtch†∗ 0.920 0.907 0.913 0.925 0.882 0.903

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