Corpus ID: 201103713

Tale of tails using rule augmented sequence labeling for event extraction

@article{Patel2019TaleOT,
  title={Tale of tails using rule augmented sequence labeling for event extraction},
  author={H. Patel and Nandan Rathod and Ayush Maheshwari and R. Kumar and Ganesh Ramakrishnan and P. Bhattacharyya},
  journal={ArXiv},
  year={2019},
  volume={abs/1908.07018}
}
  • H. Patel, Nandan Rathod, +3 authors P. Bhattacharyya
  • Published 2019
  • Computer Science
  • ArXiv
  • The problem of event extraction is a relatively difficult task for low resource languages due to the non-availability of sufficient annotated data. Moreover, the task becomes complex for tail (rarely occurring) labels wherein extremely less data is available. In this paper, we present a new dataset (InDEE-2019) in the disaster domain for multiple Indic languages, collected from news websites. Using this dataset, we evaluate several rule-based mechanisms to augment deep learning based models. We… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 12 REFERENCES
    Joint Event Extraction via Recurrent Neural Networks
    172
    Event Extraction Using Distant Supervision
    37
    Snorkel: Rapid Training Data Creation with Weak Supervision
    242
    Harnessing Deep Neural Networks with Logic Rules
    303
    Revisiting Unreasonable Effectiveness of Data in Deep Learning Era
    508
    Text Chunking using Transformation-Based Learning
    1287
    Bidirectional LSTM-CRF Models for Sequence Tagging
    1418
    Enriching Word Vectors with Subword Information
    3516
    Long Short-Term Memory
    29973
    A Domain-independent Rule-based Framework for Event Extraction
    55