• TUCS Publication Series
• How to Publish in TUCS Series
• Publication Search
• Publication Graph
• Publication Input
• Publication Input Guide
• JuFo Browser

Do We Really Need All Those Rich Linguistic Features? A Neural Network-Based Approach to Implicit Sense Labeling

Niko Schenk, Christian Chiarcos, Kathrin Donandt, Samuel Rönnqvist, Evgeny A. Stepanov, Giuseppe Riccardi, Do We Really Need All Those Rich Linguistic Features? A Neural Network-Based Approach to Implicit Sense Labeling. In: Nianwen Xue, Hwee Tou Ng, Sameer Pradhan, Attapol Rutherford, Bonnie Webber, Chuan Wang, Hongmin Wang (Eds.), Proceedings of the CoNLL-16 shared task, 41–49, Association for Computational Linguistics, 2016.

http://dx.doi.org/10.18653/v1/K16-2005

Abstract:

We describe our contribution to the CoNLL 2016 Shared Task on shallow discourse parsing. Our system extends the two best parsers from previous year’s competition by integration of a novel
implicit sense labeling component. It is grounded
on a highly generic, language-independent feedforward neural network architecture incorporating weighted word embeddings for argument spans which obviates the need for (traditional) hand-crafted features. Despite its simplicity, our system overall outperforms all results from 2015 on 5 out of 6 evaluation sets for English and achieves an absolute improvement in
F1-score of 3.2% on the PDTB test section for non-explicit sense classification.

Files:

Full publication in PDF-format

BibTeX entry:

@INPROCEEDINGS{inpScChDoRxStRi16a,
  title = {Do We Really Need All Those Rich Linguistic Features? A Neural Network-Based Approach to Implicit Sense Labeling},
  booktitle = {Proceedings of the CoNLL-16 shared task},
  author = {Schenk, Niko and Chiarcos, Christian and Donandt, Kathrin and Rönnqvist, Samuel and Stepanov, Evgeny A. and Riccardi, Giuseppe},
  editor = {Xue, Nianwen and Ng, Hwee Tou and Pradhan, Sameer and Rutherford, Attapol and Webber, Bonnie and Wang, Chuan and Wang, Hongmin},
  publisher = {Association for Computational Linguistics},
  pages = {41–49},
  year = {2016},
}

Belongs to TUCS Research Unit(s): Data Mining and Knowledge Management Laboratory

Publication Forum rating of this publication: level 1

Edit publication