Convolutional recurrent neural network for question answering

Question answering has become a very important task for natural language understanding as most natural language processing problems can be posed as a question answering problem. Recurrent neural network (RNN) is a standard baseline model for various sequence prediction tasks including question answe...

Full description

Saved in:
Bibliographic Details
Published in2017 3rd International Conference on Electrical Information and Communication Technology (EICT) pp. 1 - 6
Main Authors Zaman, M. M. Arefin, Mishu, Sadia Zaman
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.12.2017
Subjects
Convolution
convolutional neural network
Deep learning
Kernel
Knowledge discovery
Natural language processing
question answering
recurrent neural network
Recurrent neural networks
Task analysis
Online AccessGet full text

Cover

More Information
Summary:Question answering has become a very important task for natural language understanding as most natural language processing problems can be posed as a question answering problem. Recurrent neural network (RNN) is a standard baseline model for various sequence prediction tasks including question answering. Recurrent networks can represent global information for a long period of time but they do not preserve local information very well. To address this problem we propose a model which is a combination of recurrent and convolutional network that can be trained end to end using backpropagation. Our experiments on bAbI dataset demonstrate that this model can achieve significant improvement over RNN model for question answering task.
ISBN:9781538623053
1538623056
DOI:10.1109/EICT.2017.8275236