An Attention-Gated Convolutional Neural Network for Sentence Classification
Abstract
The classification of sentences is very challenging, since sentences contain the limited contextual information. In this paper, we proposed an Attention-Gated Convolutional Neural Network (AGCNN) for sentence classification, which generates attention weights from the feature's context windows of different sizes by using specialized convolution encoders. It makes full use of limited contextual information to extract and enhance the influence of important features in predicting the sentence's category. Experimental results demonstrated that our model can achieve up to 3.1% higher accuracy than standard CNN models, and gain competitive results over the baselines on four out of the six tasks. Besides, we designed an activation function, namely, Natural Logarithm rescaled Rectified Linear Unit (NLReLU). Experiments showed that NLReLU can outperform ReLU and is comparable to other well-known activation functions on AGCNN.
Cite
@article{arxiv.1808.07325,
title = {An Attention-Gated Convolutional Neural Network for Sentence Classification},
author = {Yang Liu and Lixin Ji and Ruiyang Huang and Tuosiyu Ming and Chao Gao and Jianpeng Zhang},
journal= {arXiv preprint arXiv:1808.07325},
year = {2018}
}
Comments
Accepted for publication in the Intelligent Data Analysis journal, 19 pages, 4 figures and 5 tables