ML Anthology

Con-S2V: A Generic Framework for Incorporating Extra-Sentential Context into Sen2Vec

Tanay Kumar Saha, Shafiq R. Joty, Mohammad Al Hasan
ECML-PKDD 2017 pp. 753-769
doi:10.1007/978-3-319-71249-9_45 /ecmlpkdd/2017/saha2017ecmlpkdd-cons2v/

Abstract

We present a novel approach to learn distributed representation of sentences from unlabeled data by modeling both content and context of a sentence. The content model learns sentence representation by predicting its words. On the other hand, the context model comprises a neighbor prediction component and a regularizer to model distributional and proximity hypotheses, respectively. We propose an online algorithm to train the model components jointly. We evaluate the models in a setup, where contextual information is available. The experimental results on tasks involving classification, clustering, and ranking of sentences show that our model outperforms the best existing models by a wide margin across multiple datasets. Code related to this chapter is available at: https://github.com/tksaha/con-s2v/tree/jointlearning Data related to this chapter are available at: https://www.dropbox.com/sh/ruhsi3c0unn0nko/AAAgVnZpojvXx9loQ21WP_MYa?dl=0

PDF ECML-PKDD Semantic Scholar

Cite

Text

Saha et al. "Con-S2V: A Generic Framework for Incorporating Extra-Sentential Context into Sen2Vec." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2017. doi:10.1007/978-3-319-71249-9_45

Markdown

[Saha et al. "Con-S2V: A Generic Framework for Incorporating Extra-Sentential Context into Sen2Vec." European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, 2017.](https://mlanthology.org/ecmlpkdd/2017/saha2017ecmlpkdd-cons2v/) doi:10.1007/978-3-319-71249-9_45

BibTeX

@inproceedings{saha2017ecmlpkdd-cons2v,
  title     = {{Con-S2V: A Generic Framework for Incorporating Extra-Sentential Context into Sen2Vec}},
  author    = {Saha, Tanay Kumar and Joty, Shafiq R. and Al Hasan, Mohammad},
  booktitle = {European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases},
  year      = {2017},
  pages     = {753-769},
  doi       = {10.1007/978-3-319-71249-9_45},
  url       = {https://mlanthology.org/ecmlpkdd/2017/saha2017ecmlpkdd-cons2v/}
}