The Concept of System for Automated Scientific Literature Reviews Generation

ABSTRACT: We present a concept of system which is aimed to create a literature review of scientific articles having a small sketch of statements as the input. Key elements of the system include transformer-based BERT encoder, deep LSTM decoder and a loss function which combines auto-encoder loss and forces generated summaries to be in the input text domain. We propose to use PMC open access subset for model learning.

SUBMITTER: Krzhizhanovskaya V 

PROVIDER: S-EPMC7304034 | biostudies-literature | 2020 May

REPOSITORIES: biostudies-literature