Show simple item record

dc.contributor.authorChen, Jun
dc.contributor.authorHoehndorf, Robert
dc.contributor.authorElhoseiny, Mohamed
dc.contributor.authorZhang, Xiangliang
dc.date.accessioned2020-04-16T11:33:45Z
dc.date.available2020-04-16T11:33:45Z
dc.date.issued2020-04-07
dc.identifier.urihttp://hdl.handle.net/10754/662553
dc.description.abstractIn natural language processing, relation extraction seeks to rationally understand unstructured text. Here, we propose a novel SpanBERT-based graph convolutional network (DG-SpanBERT) that extracts semantic features from a raw sentence using the pre-trained language model SpanBERT and a graph convolutional network to pool latent features. Our DG-SpanBERT model inherits the advantage of SpanBERT on learning rich lexical features from large-scale corpus. It also has the ability to capture long-range relations between entities due to the usage of GCN on dependency tree. The experimental results show that our model outperforms other existing dependency-based and sequence-based models and achieves a state-of-the-art performance on the TACRED dataset.
dc.publisherarXiv
dc.relation.urlhttps://arxiv.org/pdf/2004.03636
dc.rightsArchived with thanks to arXiv
dc.titleEfficient long-distance relation extraction with DG-SpanBERT
dc.typePreprint
dc.contributor.departmentBio-Ontology Research Group (BORG)
dc.contributor.departmentComputational Bioscience Research Center (CBRC)
dc.contributor.departmentComputer Science
dc.contributor.departmentComputer Science Program
dc.contributor.departmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
dc.contributor.departmentKing Abduallah University of Science and Technology Thuwal 23955, Saudi Arabia.
dc.contributor.departmentMachine Intelligence & kNowledge Engineering Lab
dc.eprint.versionPre-print
dc.identifier.arxivid2004.03636
kaust.personChen, Jun
kaust.personHoehndorf, Robert
kaust.personElhoseiny, Mohamed
kaust.personZhang, Xiangliang
refterms.dateFOA2020-04-16T11:43:08Z


Files in this item

Thumbnail
Name:
Preprintfile1.pdf
Size:
344.5Kb
Format:
PDF
Description:
Pre-print

This item appears in the following Collection(s)

Show simple item record