A Neural Network Based Text Classification with Attention Mechanism

Abstract

Text classification is a basic task in Natural Language Processing field. As neural networks gains great breakthroughs in computer vision and speech recognition, neural networks based model, such as convolutional neural networks and recurrent neural networks, is also proved to be powerful in many Natural Language Processing tasks compared to traditional approaches. Based on recurrent neural networks with attention mechanism, a new model incorporated enhanced text representation by means of convolutional neural networks is proposed to deal with text classification task. The experimental results show that the proposed model gains higher accuracy compared to the common attention based recurrent neural networks model.