四川大学计算机学院琚生根教授,基于卷积神经网络和自注意力机制的文本分类模型...

Abstract:

The wordlevel shallow convolutional neural network (CNN) model has achieved good performance in text classification tasks. However, shallow CNN models cant capture longrange dependencies, which affects the model's performance in text classification tasks, but simply deepening the number of layers of the model does not improve the models performance. This paper proposes a new wordlevel text classification model WordCNNAtt, which uses CNN to capture local features and position information, and captures longrange dependencies with selfattention mechanism. The accuracy of WordCNNAtt, on the five public datasets of AGNews, DBPedia, Yelp Review Polarity, Yelp Review Full, Yahoo! Answers, is 0.9%, 0.2%, 0.5%, 2.1%, and 2.0% higher than the wordlevel shallow CNN model respectively.