Subsumption Matching Results for SNOMED-NCIT (Neoplas)
⇕System ⇕Use Training Maps ⇕MRR ⇕Hits@1 ⇕Hits@5 ⇕Hits@10
BERTSubs (IC) 0.6610.4830.8930.980
BERTSubs (IC) 0.5590.3560.8320.940
OWL2Vec* + RF 0.673 0.5030.9330.980
SORBETMtch 0.685 0.557 0.859 0.899
Word2Vec + RF 0.597 0.4090.8660.966
Word2Vec + RF (del) 0.532 0.3690.7450.805

Note: (del) is a trial setting that uses the pruned target ontology similar to OAEI 2022.

Note: OWL2Vec* uses structure, literal and mixture documents, without using pre-trained Word2Vec. RF indicates Random Forest Classifier with n_estimators=50.