FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W4378515217> ?p ?o ?g. }

• W4378515217 endingPage "586" @default.
• W4378515217 startingPage "586" @default.
• W4378515217 abstract "The identification of drug–drug and chemical–protein interactions is essential for understanding unpredictable changes in the pharmacological effects of drugs and mechanisms of diseases and developing therapeutic drugs. In this study, we extract drug-related interactions from the DDI (Drug–Drug Interaction) Extraction-2013 Shared Task dataset and the BioCreative ChemProt (Chemical–Protein) dataset using various transfer transformers. We propose BERTGAT that uses a graph attention network (GAT) to take into account the local structure of sentences and embedding features of nodes under the self-attention scheme and investigate whether incorporating syntactic structure can help relation extraction. In addition, we suggest T5slim_dec, which adapts the autoregressive generation task of the T5 (text-to-text transfer transformer) to the relation classification problem by removing the self-attention layer in the decoder block. Furthermore, we evaluated the potential of biomedical relation extraction of GPT-3 (Generative Pre-trained Transformer) using GPT-3 variant models. As a result, T5slim_dec, which is a model with a tailored decoder designed for classification problems within the T5 architecture, demonstrated very promising performances for both tasks. We achieved an accuracy of 91.15% in the DDI dataset and an accuracy of 94.29% for the CPR (Chemical–Protein Relation) class group in ChemProt dataset. However, BERTGAT did not show a significant performance improvement in the aspect of relation extraction. We demonstrated that transformer-based approaches focused only on relationships between words are implicitly eligible to understand language well without additional knowledge such as structural information." @default.
• W4378515217 created "2023-05-28" @default.
• W4378515217 creator A5000233346 @default.
• W4378515217 creator A5046307698 @default.
• W4378515217 creator A5061768302 @default.
• W4378515217 date "2023-05-12" @default.
• W4378515217 modified "2023-09-30" @default.
• W4378515217 title "Biomedical Relation Extraction Using Dependency Graph and Decoder-Enhanced Transformer Model" @default.
• W4378515217 cites W2061833373 @default.
• W4378515217 cites W2128768874 @default.
• W4378515217 cites W2170146596 @default.
• W4378515217 cites W2264517602 @default.
• W4378515217 cites W2892094955 @default.
• W4378515217 cites W2910406387 @default.
• W4378515217 cites W2911489562 @default.
• W4378515217 cites W2963250244 @default.
• W4378515217 cites W2964026782 @default.
• W4378515217 cites W2964217331 @default.
• W4378515217 cites W2970771982 @default.
• W4378515217 cites W2979826702 @default.
• W4378515217 cites W3009926341 @default.
• W4378515217 cites W3027982260 @default.
• W4378515217 cites W3034999214 @default.
• W4378515217 cites W3104578551 @default.
• W4378515217 cites W3166204619 @default.
• W4378515217 cites W3166593409 @default.
• W4378515217 cites W3192478068 @default.
• W4378515217 cites W3198824379 @default.
• W4378515217 cites W3213241618 @default.
• W4378515217 cites W4221153690 @default.
• W4378515217 cites W4243798656 @default.
• W4378515217 cites W4285158728 @default.
• W4378515217 cites W4285225959 @default.
• W4378515217 cites W4297458751 @default.
• W4378515217 cites W4309728330 @default.
• W4378515217 doi "https://doi.org/10.3390/bioengineering10050586" @default.
• W4378515217 hasPubMedId "https://pubmed.ncbi.nlm.nih.gov/37237656" @default.
• W4378515217 hasPublicationYear "2023" @default.
• W4378515217 type Work @default.
• W4378515217 citedByCount "2" @default.
• W4378515217 countsByYear W43785152172023 @default.
• W4378515217 crossrefType "journal-article" @default.
• W4378515217 hasAuthorship W4378515217A5000233346 @default.
• W4378515217 hasAuthorship W4378515217A5046307698 @default.
• W4378515217 hasAuthorship W4378515217A5061768302 @default.
• W4378515217 hasBestOaLocation W43785152171 @default.
• W4378515217 hasConcept C119857082 @default.
• W4378515217 hasConcept C121332964 @default.
• W4378515217 hasConcept C124101348 @default.
• W4378515217 hasConcept C132525143 @default.
• W4378515217 hasConcept C149782125 @default.
• W4378515217 hasConcept C153180895 @default.
• W4378515217 hasConcept C153604712 @default.
• W4378515217 hasConcept C154945302 @default.
• W4378515217 hasConcept C159877910 @default.
• W4378515217 hasConcept C16311509 @default.
• W4378515217 hasConcept C165801399 @default.
• W4378515217 hasConcept C195807954 @default.
• W4378515217 hasConcept C33923547 @default.
• W4378515217 hasConcept C41008148 @default.
• W4378515217 hasConcept C41608201 @default.
• W4378515217 hasConcept C62520636 @default.
• W4378515217 hasConcept C66322947 @default.
• W4378515217 hasConcept C80444323 @default.
• W4378515217 hasConceptScore W4378515217C119857082 @default.
• W4378515217 hasConceptScore W4378515217C121332964 @default.
• W4378515217 hasConceptScore W4378515217C124101348 @default.
• W4378515217 hasConceptScore W4378515217C132525143 @default.
• W4378515217 hasConceptScore W4378515217C149782125 @default.
• W4378515217 hasConceptScore W4378515217C153180895 @default.
• W4378515217 hasConceptScore W4378515217C153604712 @default.
• W4378515217 hasConceptScore W4378515217C154945302 @default.
• W4378515217 hasConceptScore W4378515217C159877910 @default.
• W4378515217 hasConceptScore W4378515217C16311509 @default.
• W4378515217 hasConceptScore W4378515217C165801399 @default.
• W4378515217 hasConceptScore W4378515217C195807954 @default.
• W4378515217 hasConceptScore W4378515217C33923547 @default.
• W4378515217 hasConceptScore W4378515217C41008148 @default.
• W4378515217 hasConceptScore W4378515217C41608201 @default.
• W4378515217 hasConceptScore W4378515217C62520636 @default.
• W4378515217 hasConceptScore W4378515217C66322947 @default.
• W4378515217 hasConceptScore W4378515217C80444323 @default.
• W4378515217 hasFunder F4320322028 @default.
• W4378515217 hasIssue "5" @default.
• W4378515217 hasLocation W43785152171 @default.
• W4378515217 hasLocation W43785152172 @default.
• W4378515217 hasOpenAccess W4378515217 @default.
• W4378515217 hasPrimaryLocation W43785152171 @default.
• W4378515217 hasRelatedWork W2947903144 @default.
• W4378515217 hasRelatedWork W2949602493 @default.
• W4378515217 hasRelatedWork W2996917304 @default.
• W4378515217 hasRelatedWork W3046335499 @default.
• W4378515217 hasRelatedWork W3093872895 @default.
• W4378515217 hasRelatedWork W3157910026 @default.
• W4378515217 hasRelatedWork W3203028781 @default.
• W4378515217 hasRelatedWork W3204214691 @default.
• W4378515217 hasRelatedWork W4206160067 @default.
• W4378515217 hasRelatedWork W4378515217 @default.

• next