Chinese Text Feature Extraction and Classification Based on Deep Learning
Ruishuang Wang, Zhao Li, Jian Cao, Tong Chen · Proceedings of the 3rd International Conference on Computer Science and Application Engineering · 2019
With the rapid development of deep learning, neural networks have been widely used in natural language processing tasks and achieved good results. Since convolutional neural networks can acquire high-level features that can better represent textual semantic information, convolutional neural networks (CNN) and convolutional recurrent neural networks (CRNN) are used to establish feature extraction models to extract text features. At the same time, tf-idf and word2vec methods are used to represent text features, and then feed them into SVM and Random forest classifier to classify Chinese academic papers dataset. Experimental results show that the classification results obtained by using the CNN and CRNN feature extraction model are better than using the TF-IDF and Word2vec feature extraction methods. In addition, the classification results obtained by using SVM and Random forest classifier are better than that of the original neural network.