Using hyperlinks to improve multilingual partial parsers

    Abstract

    Syntactic annotation is costly and not available for the vast majority of the world's languages. We show that sometimes we can do away with less labeled data by exploiting more readily available forms of mark-up. Specifically, we revisit an idea from Valentin Spitkovsky's work (2010), namely that hyperlinks typically bracket syntactic constituents or chunks. We strengthen his results by showing that not only can hyperlinks help in low resource scenarios, exemplified here by Quechua, but learning from hyperlinks can also improve state-of-the-art NLP models for English newswire. We also present out-of-domain evaluation on English Ontonotes 4.0.

    Original languageEnglish
    Title of host publicationProceedings of the 15th International Conference on Parsing Technologies
    PublisherAssociation for Computational Linguistics
    Publication date2017
    Pages67-71
    Publication statusPublished - 2017
    Event15th International Conference on Parsing Technologies - Pisa, Italy
    Duration: 20 Sept 201722 Sept 2017

    Conference

    Conference15th International Conference on Parsing Technologies
    Country/TerritoryItaly
    CityPisa
    Period20/09/201722/09/2017

    Fingerprint

    Dive into the research topics of 'Using hyperlinks to improve multilingual partial parsers'. Together they form a unique fingerprint.

    Cite this