Aspect sentiment analysis with heterogeneous graph neural networks |
| |
| Institution: | 1. School of Information Resource Management, Renmin University of China, Beijing 100872, China;2. Research Center for Digital Humanities, Renmin University of China, Beijing 100872, China;3. Department of Information Management, Nanjing University of Science and Technology, Nanjing, 210094, China;4. Computer Science, Worcester Polytechnic Institute, Worcester, MA 01609, USA |
| |
| Abstract: | Aspect-based sentiment analysis technologies may be a very practical methodology for securities trading, commodity sales, movie rating websites, etc. Most recent studies adopt the recurrent neural network or attention-based neural network methods to infer aspect sentiment using opinion context terms and sentence dependency trees. However, due to a sentence often having multiple aspects sentiment representation, these models are hard to achieve satisfactory classification results. In this paper, we discuss these problems by encoding sentence syntax tree, words relations and opinion dictionary information in a unified framework. We called this method heterogeneous graph neural networks (Hete_GNNs). Firstly, we adopt the interactive aspect words and contexts to encode the sentence sequence information for parameter sharing. Then, we utilized a novel heterogeneous graph neural network for encoding these sentences’ syntax dependency tree, prior sentiment dictionary, and some part-of-speech tagging information for sentiment prediction. We perform the Hete_GNNs sentiment judgment and report the experiments on five domain datasets, and the results confirm that the heterogeneous context information can be better captured with heterogeneous graph neural networks. The improvement of the proposed method is demonstrated by aspect sentiment classification task comparison. |
| |
| Keywords: | Graph attention networks Aspect sentiment analysis Opinion mining Heterogeneous graph neural network Multi-head attention mechanism Graph convolution neural networks |
| 本文献已被 ScienceDirect 等数据库收录！ |
|
