Improving the learning of chemical-protein interactions from literature using transfer learning and specialized word embeddings. (12th July 2018)

• Record Type: Journal Article
• Authors: Corbett, P; Boyle, J
• Is Part Of: Database. Volume 2018(2018)
• Publication Date: 2018-07-12
• Subjects: Biology -- Databases -- Periodicals; Bioinformatics -- Periodicals; 570.285
• DOI: 10.1093/database/bay066
• Languages: English
• ISSNs: 1758-0463

Abstract

Abstract: In this paper, we explore the application of artificial neural network ('deep learning') methods to the problem of detecting chemical-protein interactions in PubMed abstracts. We present here a system using multiple Long Short Term Memory layers to analyse candidate interactions, to determine whether there is a relation and which type. A particular feature of our system is the use of unlabelled data, both to pre-train word embeddings and also pre-train LSTM layers in the neural network. On the BioCreative VI CHEMPROT test corpus, our system achieves an F score of 61.51% (56.10% precision, 67.84% recall).