Convolutional Neural Networks for Text Classification with Multi-size Convolution and Multi-type Pooling

Abstract

Translate Article

Text classification is a very important problem in Nature Language Processing (NLP). The text classification based on shallow machine-learning models takes too much time and energy to extract features of data, but only obtains poor performance. Recently, deep learning methods are widely used in text classification and result in good performance. In this paper, we propose a Convolutional Neural Network (CNN) with multi-size convolution and multi-type pooling for text classification. In our method, we adopt CNNs to extract features of the texts and then select the important information of these features through multi-type pooling. Experiments show that the CNN with multi-convolution and multi-type pooling (CNN-MCMP) obtains better performance on text classification compared with both the shallow machine-learning models and other CNN architectures.