INNOVATORS at SemEval-2021 Task-11: A Dependency Parsing and BERT-based model for Extracting Contribution Knowledge from Scientific Papers

Proceedings of the 15th International Workshop on Semantic Evaluation (SemEval-2021), pages 502–510 Bangkok, Thailand (online), August 5–6, 2021. ©2021 Association for Computational Linguistics 502 INNOVATORS at SemEval-2021 Task-11: A Dependency Parsing and BERT-based model for Extracting Contribution Knowledge from Scientific Papers Hardik Arora†, Tirthankar Ghosal*, Sandeep Kumar†, Suraj Patwal†, Phil Gooch‡ †Indian Institute of Technology Patna, India ∗Institute of Formal and Applied Linguistics, Charles University, Czech Republic †(hardik 1901ce15,sandeep 1911mc12,suraj 1911mt12)@iitp.ac.in ∗ghosal@ufal.mff.cuni.cz, ‡phil@scholarcy.com Abstract In this work, we describe our system sub- mission to the SemEval 2021 Task 11: NLP Contribution Graph Challenge. We attempt all the three sub-tasks in the challenge and report our results. Subtask 1 aims to identify the contributing sentences in a given publi- cation. Subtask 2 follows from Subtask 1 to extract the scientific term and predicate phrases from the identified contributing sen- tences. The final Subtask 3 entails extracting triples (subject, predicate, object) from the phrases and categorizing them under one or more defined information units. With the NLPContributionGraph Shared Task, the organizers formalized the building of a scholarly contributions-focused graph over NLP scholarly articles as an automated task. Our approaches include a BERT-based classification model for identifying the con- tributing sentences in a research publication, a rule-based dependency parsing for phrase extraction, followed by a CNN-based model for information units classification and a set of rules for triples extraction. The quantitative results show that we obtain the 5th, 5th, and 7th rank respectively in three evaluation phases. We make our codes available at https://github.com/HardikArora17/ SemEval-2021-INNOVATORS 1 Introduction Thousands of papers are published by the scientific community every day. It is now increasingly be- coming difficult to browse the huge pool of papers to identify relevant work and thereby keep up with the latest research findings. Scientific literature is growing at an exponential rate and researchers today face the problem to identify the latest state- of-the-art contributions. Keeping track of recent advancements is becoming a tedious exercise, if not practically impossible. The Open Research Knowledge Graph (ORKG) (Jaradeh et al., 2019) is posited as a solution to keeping track of research progress minus the cognitive overload that reading dozens of full papers imposes. It aims to build a comprehensive knowledge graph that publishes scholarly publications’ research contributions per paper, where the contributions are interconnected via the graph even across documents. As described in D’Souza et al. (2021), with the ORKG comparisons feature, researchers are no longer faced with the daunting cognitive ingestion obstacle from manually scouring through dozens of papers of unstructured content in their field. This process traditionally would take several days or months; using the ORKG contributions compari- son tabulated view, the task is reduced to just a few minutes. Assuming the individual paper contribu- tions are structured in the ORKG, they can then deconstruct the graph, tap into the aspects they are interested in, and can enhance it for their purposes. Further, they can select multiple such paper graphs and click a button to generate their tabulated com- parison. This presents …

Content truncated to 500 words