Literature mining for context-specific molecular relations using multimodal representations (COMMODAR)

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 94
  • Download : 0
Biological contextual information helps understand various phenomena occurring in the biological systems consisting of complex molecular relations. The construction of context-specific relational resources vastly relies on laborious manual extraction from unstructured literature. In this paper, we propose COMMODAR, a machine learning-based literature mining framework for context-specific molecular relations using multimodal representations. The main idea of COMMODAR is the feature augmentation by the cooperation of multimodal representations for relation extraction. We leveraged biomedical domain knowledge as well as canonical linguistic information for more comprehensive representations of textual sources. The models based on multiple modalities outperformed those solely based on the linguistic modality. We applied COMMODAR to the 14 million PubMed abstracts and extracted 9214 context-specific molecular relations. All corpora, extracted data, evaluation results, and the implementation code are downloadable at https://github.com/jae-hyun-lee/commodar. Ccs concepts center dot Computing methodologies similar to Information extraction center dot Computing methodologies similar to Neural networks center dot Applied computing similar to Biological networks.
Publisher
BMC
Issue Date
2020-10
Language
English
Article Type
Article
Citation

BMC BIOINFORMATICS, v.21

ISSN
1471-2105
DOI
10.1186/s12859-020-3396-y
URI
http://hdl.handle.net/10203/279203
Appears in Collection
BiS-Journal Papers(저널논문)RIMS Journal Papers
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0