An Empirical Study of Span Modeling in Science NER

Xiaorui Jiang

    Research output: Chapter in Book/Report/Conference proceedingConference proceedingpeer-review

    1 Citation (Scopus)
    170 Downloads (Pure)

    Abstract

    Little evaluation has been performed on the many modeling options for span-based approaches. This paper investigates the performances of a wide range of span and context representation methods and their combinations with a focus on scientific named entity recognition (science NER). While some most common classical span encodings and their combination prove to be effective, few conclusions can be derived to context representations.

    Original languageEnglish
    Title of host publicationLinking Theory and Practice of Digital Libraries
    Subtitle of host publication25th International Conference on Theory and Practice of Digital Libraries, TPDL 2021, Virtual Event, September 13–17, 2021, Proceedings
    EditorsGerd Berget, Mark Michael Hall, Daniel Brenn, Sanna Kumpulainen
    PublisherSpringer
    Pages41-48
    Number of pages8
    ISBN (Electronic)978-3-030-86324-1
    ISBN (Print)978-3-030-86323-4
    DOIs
    Publication statusPublished - 7 Sept 2021
    Event25th International Conference on Theory and Practice of Digital Libraries - Online, United Kingdom
    Duration: 13 Sept 202117 Sept 2021
    Conference number: 25
    http://www.tpdl.eu/tpdl2021/

    Publication series

    NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
    Volume12866 LNCS
    ISSN (Print)0302-9743
    ISSN (Electronic)1611-3349

    Conference

    Conference25th International Conference on Theory and Practice of Digital Libraries
    Country/TerritoryUnited Kingdom
    Period13/09/2117/09/21
    Internet address

    Keywords

    • Context representation
    • SciBERT
    • Scientific named entity recognition
    • Span representation
    • Span-based model

    ASJC Scopus subject areas

    • Theoretical Computer Science
    • General Computer Science

    Fingerprint

    Dive into the research topics of 'An Empirical Study of Span Modeling in Science NER'. Together they form a unique fingerprint.

    Cite this