A method based on combined-convolutional neural network (Combined-CNN) for Chinese news text classification is proposed. First of all, in order to solve the problem of a lack of special term set for Chi-nese news classification, a vocabulary suitable for Chinese long text classification is made by construct-ing a data index method. The Word2Vec pre-trained model was used to embed the text features word vectors. Second, by optimizing the structure of the classical convolutional neural network (CNN) model, a new idea of Combined-CNN model is proposed, which solves the problem of incomplete feature ex-traction of local text blocks and improves the accuracy rate of Chinese news text classification. Effective model regularization and RAdam optimization algorithm are designed in the model to enhance the model training effect. The experimental results show that the precision of the Combined-CNN model for Chi-nese news text classification reaches 93.69%. Compared with traditional machine learning methods and deep learning algorithms, the accuracy rate is improved by a maximum of 11.82% and 1.9%, respectively, and it is better than the comparison model in Recall and F-Measure. Finally, the Chinese news classifica-tion algorithm of the Combined-CNN is applied to realize a personalized recommendation system.
本站僅提供期刊文獻檢索。
