Scientific Discourse Tagging for Evidence Extraction

Proceedings of the 16th Conference of the European Chapter of the Association for Computational Linguistics, pages 2550–2562 April 19 - 23, 2021. ©2021 Association for Computational Linguistics 2550 Scientific Discourse Tagging for Evidence Extraction Xiangci Li1∗ Gully Burns2 Nanyun Peng3 1 University of Texas at Dallas, 2 Chan Zuckerburg Initiative 3 University of California Los Angeles lixiangci8@gmail.com,gully.burns@chanzuckerburg.com violetpeng@cs.ucla.edu Abstract Evidence plays a crucial role in any biomed- ical research narrative, providing justification for some claims and refutation for others. We seek to build models of scientific argument us- ing information extraction methods from full- text papers. We present the capability of au- tomatically extracting text fragments from pri- mary research papers that describe the evi- dence presented in that paper’s figures, which arguably provides the raw material of any sci- entific argument made within the paper. We apply richly contextualized deep representa- tion learning pre-trained on biomedical do- main corpus to the analysis of scientific dis- course structures and the extraction of “evi- dence fragments” (i.e., the text in the results section describing data presented in a speci- fied subfigure) from a set of biomedical exper- imental research articles. We first demonstrate our state-of-the-art scientific discourse tagger on two scientific discourse tagging datasets and its transferability to new datasets. We then show the benefit of leveraging scientific discourse tags for downstream tasks such as claim-extraction and evidence fragment detec- tion. Our work demonstrates the potential of using evidence fragments derived from figure spans for improving the quality of scientific claims by cataloging, indexing and reusing ev- idence fragments as independent documents. 1 Introduction Primary experimental articles (i.e., papers that de- scribe original experimental work) provide the cru- cial raw material for all other subsequent scientific research. However, the drastically growing number of scientific literature makes it increasingly diffi- cult for domain experts to efficiently utilize them. Automatic information extraction from biomedical ∗ Work performed while the author is interning at the Information Sciences Institute, University of Southern Cali- fornia lc3 , the mammalian atg8 homolog , undergoes a set of modifications resulting in conversion from lc3i to lc3ii during autophagy 42 . [fact] to further test the function of rag in autophagy [goal] we examined the lc3 modification in hek293 cells . [method] expression of raga ql and ragc sn inhibited lc3 conversion in response to amino acid starvation ( fig. 7e ) . [result] furthermore , expression of raga tn and ragc ql enhanced lc3 conversion even in the presence of amino acids . [result] these results are consistent with the data observed in drosophila and further demonstrate a role of the rag gtpases in autophagy regulation in response to nutrient signals [implication] Figure 1: An example paragraph tagged with scientific discourse tags on each clause in SciDT dataset (Dasigi et al., 2017). The text is tokenized and converted to lower case. literature is a crucial step to help researchers to achieve this goal. Extracting important information from biomedi- cal literature to facilitate and accelerate scientific discovery has been …

Content truncated to 500 words