A Comparative Study of Chinese Patent Literature Automatic Classification Based on Deep Learning

摘要

Patent literature automatic classification is of great significance to literature retrieval and management. In this study, deep learning technique is used for Chinese patent literature classification, and automatic classification accuracy rates of one existing model and six deep learning models are compared. Compared with "TFIDF+Logistic Regression" model, the deep learning model has better effect of patent literature automatic classification. Furthermore, the "Word2Vec+GRU+TextCNN" model in seven models has the highest classification accuracy rate, and attention mechanism has little effect on classification results.