An Attention-Gated Convolutional Neural Network for Sentence Classification
This addresses the challenge of limited contextual information in sentence classification, offering an incremental improvement for NLP applications.
The paper tackles sentence classification by proposing an Attention-Gated Convolutional Neural Network (AGCNN) that uses attention weights from context windows to enhance important features, achieving up to 3.1% higher accuracy than standard CNNs and competitive results on most tasks.
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.