Argument Component Classification with Context-LSTM

DOI

10.2991/gcmce-17.2017.23How to use a DOI?

Keywords

Argumentation Mining, Argument Component Classification, LSTM

Abstract

Argumentation Mining (AM) is an emerging field of natural language understanding. It extends sentiment analysis, topic modeling and other existing text mining methods, aimed at tap the latent logical relationship between sentences. At present, Argument Component Classification (ACC) is a challenging subtask in AM. Existing popular SVM based models heavily rely on artificially constructed and domain dependent features, and they cannot nicely model the context relations among sentences. In this paper, we propose an ACC method based on Context-LSTM, which do not require any artificial features. Moreover, Context-LSTM can model the contextual information of the current sentence very well. We conduct experiments on two well-annotated corpora and get state-of-art results both.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Chaoming Wang
AU  - Xudong Chen
PY  - 2017/06
DA  - 2017/06
TI  - Argument Component Classification with Context-LSTM
BT  - Proceedings of the 2017 Global Conference on Mechanics and Civil Engineering (GCMCE 2017)
PB  - Atlantis Press
SP  - 115
EP  - 121
SN  - 2352-5401
UR  - https://doi.org/10.2991/gcmce-17.2017.23
DO  - 10.2991/gcmce-17.2017.23
ID  - Wang2017/06
ER  -