Question Classification Based on Hybrid Neural Networks

DOI

10.2991/iceeecs-16.2016.11How to use a DOI?

Keywords

text classification, deep learning, CNN, LSTM

Abstract

Question classification is an important step in question answering system. There are many previous work based on machine learning about question classification. Although they are effective and practical, most of them require finding specific features to train the designed classifier. So it cannot be in common use in all the situations. Recently, deep learning methods has shown its remarkable strength in natural language processing. Deep neural networks like CNN and LSTM are helpful for sentence classification. The CNN is able to extract high-level local features while the LSTM can remember and discard information according to the context. Thus we implement a Hybrid Neural Network to handle different parts of a query. Both CNN and LSTM are utilized to achieve question classification. We conduct experiments comparing with methods based on machine learning and deep learning. The experimental results show the effectiveness and efficiency of our methods.

Copyright

© 2016, 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  - Zhongcheng Zhou
AU  - Xiang Zhu
AU  - Zhonghe He
AU  - Yinchuan Qu
PY  - 2016/12
DA  - 2016/12
TI  - Question Classification Based on Hybrid Neural Networks
BT  - Proceedings of the 2016 4th International Conference on Electrical & Electronics Engineering and Computer Science (ICEEECS 2016)
PB  - Atlantis Press
SP  - 44
EP  - 52
SN  - 2352-538X
UR  - https://doi.org/10.2991/iceeecs-16.2016.11
DO  - 10.2991/iceeecs-16.2016.11
ID  - Zhou2016/12
ER  -