Int J Performability Eng ›› 2019, Vol. 15 ›› Issue (3): 834-841.doi: 10.23940/ijpe.19.03.p12.834841

Previous Articles     Next Articles

Dual-Channel Attention Model for Text Sentiment Analysis

Hui Li, Yuanyuan Zheng*, and Pengju Ren   

  1. School of Physics and Electronic Information, Henan Polytechnic University, Jiaozuo, 454000, China
  • Submitted on ; Revised on ;
  • Contact: 13273919602@163.com
  • About author:Hui Li received his doctorate degree in information and communication engineering in 2008 from Nanjing University of Science and Technology. He is currently working as a teacher at Henan Polytechnic University. His research interests are chaotic communications.Yuanyuan Zheng is a graduate student in the School of Physics and Electronic Information at Henan Polytechnic University. Her research interests are natural language processing and deep learning.Pengju Ren is an undergraduate student in the School of Physics and Electronic Information at Henan Polytechnic University. His research interests are image processing and data mining.

Abstract: Focused on the issue that text information cannot be fully extracted by the single-channel neural network model, the Dual-Channel Attention Model (DCAM) is proposed for text sentiment analysis. Firstly, text is represented in the form of a matrix using a word vector trained by Word2Vec. Secondly, the matrix is used as input data and sent to Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) networks for feature extraction. Thirdly, an attention model is introduced to extract important feature information. Finally, the text features are merged, and the classification layer is used to classify the sentiment. The model is evaluated on a Chinese corpus. According to the experimental results, the accuracy of the proposed model can reach 92.7%, which is obviously superior to other single-channel neural network models.

Key words: convolutional neural network, long short-term memory, attention model, sentiment analysis, dual-channel