HyperAI超神経

プラットフォーム

Drug Drug Interaction Extraction On Ddi

評価指標

F1

Micro F1

評価結果

このベンチマークにおける各モデルのパフォーマンス結果

			Paper Title
DESC+MOL+SciBERT	0.8408	84.08	Using Drug Descriptions and Molecular Structures for Drug-Drug Interaction Extraction from Literature
SciFive-large	0.8367	83.67	SciFive: a text-to-text transformer model for biomedical literature
PubMedBERT	0.8236	82.36	Domain-Specific Language Model Pretraining for Biomedical Natural Language Processing
BioBERT	0.8088	80.88	BioBERT: a pre-trained biomedical language representation model for biomedical text mining
CharacterBERT (base, medical)	-	80.38	CharacterBERT: Reconciling ELMo and BERT for Word-Level Open-Vocabulary Representations From Characters
ELECTRAMed	-	79.13	ELECTRAMed: a new pre-trained language representation model for biomedical NLP
Hierarchy Bi-LSTMs +Att.+SDP	0.729	72.9	Drug–drug interaction extraction via hierarchical RNNs on sequence and shortest dependency paths
MOL+CNN	0.7255	72.55	Enhancing Drug-Drug Interaction Extraction from Texts by Molecular Structure Information
Graph Kernel-based method	0.684	68.4	A graph kernel based on context vectors for extracting drug–drug interactions
Feature-based SVM	0.670	67.0	Extracting drug–drug interactions from literature using a rich feature-based linear kernel approach

0 of 10 row(s) selected.

Drug Drug Interaction Extraction On Ddi | SOTA | HyperAI超神経