CRAN:A Hybrid CNN-RNN Attention-Based Model for Text Classification

摘要：

　　Text classification is one of the fundamental tasks in the field of natural language processing.The CNN-based approaches and RNN-based approaches have shown different capabilities in representing a piece of text.In this paper,we propose a hybrid CNN-RNN attention-based neural network,named CRAN,which combines the convolutional neural network and recurrent neural network effectively with the help of the attention mechanism.We validate the proposed model on several largescale datasets(i.e.,eight multi-class text classification and five multi-label text classification tasks),and compare with the state-of-the-art models.Experimental results show that CRAN can achieve the state-of-the-art performance on most of the datasets.In particular,CRAN yields better performance with much fewer parameters compared with a very deep convolutional networks with 29 layers,which proves its effectiveness and efficiency.

关键词： Text classification Convolutional neural network Recurrent neural network Attention mechanism

作者: Long Guo Dongxiang Zhang Lei Wang HanWang Bin Cui

作者单位: Peking University,Beijing,China University of Electronic Science and Technology of China,Chengdu,China

会议类型: 国际会议

会议名称: The 37th International Conference on Conceptual Modeling(第37届概念建模国际会议(ER2018)

会议地点: 西安

会议语种:英文

页码: 571-585

在线出版日期: 2018-10-22（万方平台首次上网日期，不代表论文的发表时间）