Corpus ID: 199668863

Multi-class Hierarchical Question Classification for Multiple Choice Science Exams

@article{Xu2019MulticlassHQ,
  title={Multi-class Hierarchical Question Classification for Multiple Choice Science Exams},
  author={Dongfang Xu and Peter Jansen and Jaycie Martin and Zhengnan Xie and Vikas Yadav and Harish Tayyar Madabushi and Oyvind Tafjord and Peter Clark},
  journal={ArXiv},
  year={2019},
  volume={abs/1908.05441}
}
  • Dongfang Xu, Peter Jansen, +5 authors Peter Clark
  • Published in ArXiv 2019
  • Computer Science
  • Prior work has demonstrated that question classification (QC), recognizing the problem domain of a question, can help answer it more accurately. [...] Key Method We then show that a BERT-based model trained on this dataset achieves a large (+0.12 MAP) gain compared with previous methods, while also achieving state-of-the-art performance on benchmark open-domain and biomedical QC datasets. Finally, we show that using this model's predictions of question topic significantly improves the accuracy of a question…Expand Abstract

    Create an AI-powered research feed to stay up to date with new papers like this posted to ArXiv

    Figures, Tables, and Topics from this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 56 REFERENCES

    Introduction to Information Retrieval

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    Learning Question Classifiers

    • Xin Li, Dan Roth
    • Computer Science
    • COLING
    • 2002
    VIEW 8 EXCERPTS
    HIGHLY INFLUENTIAL

    Multi-label biomedical question classification

    • 2019
    VIEW 3 EXCERPTS
    HIGHLY INFLUENTIAL

    High accuracy rulebased question classification using question syntax and semantics

    • H. T. Madabushi, M. Lee
    • COLING, pages 1220–1230.
    • 2016
    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL

    MS MARCO: A Human Generated MAchine Reading COmprehension Dataset

    VIEW 3 EXCERPTS
    HIGHLY INFLUENTIAL

    Convolutional Neural Networks for Sentence Classification

    VIEW 4 EXCERPTS
    HIGHLY INFLUENTIAL