Learning Feature Representations for Keyphrase Extraction

AAAI 2018 pp. 8077-8078

doi:10.1609/AAAI.V32I1.12144 /aaai/2018/florescu2018aaai-learning/

Abstract

In supervised approaches for keyphrase extraction, a candidate phrase is encoded with a set of hand-crafted features and machine learning algorithms are trained to discriminate keyphrases from non-keyphrases. Although the manually-designed features have shown to work well in practice, feature engineering is a difficult process that requires expert knowledge and normally does not generalize well. In this paper, we present SurfKE, a feature learning framework that exploits the text itself to automatically discover patterns that keyphrases exhibit. Our model represents the document as a graph and automatically learns feature representation of phrases. The proposed model obtains remarkable improvements in performance over strong baselines.

PDF AAAI Semantic Scholar

Cite

Text

Florescu and Jin. "Learning Feature Representations for Keyphrase Extraction." AAAI Conference on Artificial Intelligence, 2018. doi:10.1609/AAAI.V32I1.12144

Markdown

[Florescu and Jin. "Learning Feature Representations for Keyphrase Extraction." AAAI Conference on Artificial Intelligence, 2018.](https://mlanthology.org/aaai/2018/florescu2018aaai-learning/) doi:10.1609/AAAI.V32I1.12144

BibTeX

@inproceedings{florescu2018aaai-learning,
  title     = {{Learning Feature Representations for Keyphrase Extraction}},
  author    = {Florescu, Corina and Jin, Wei},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2018},
  pages     = {8077-8078},
  doi       = {10.1609/AAAI.V32I1.12144},
  url       = {https://mlanthology.org/aaai/2018/florescu2018aaai-learning/}
}