基于标签主题模型的互联网文本分类

utomatic text classification in data mining is a very important task.With the rapid development of Internet, research in this area will help people to mine more useful information. Probabilistic topic model in the application of text classification is mainly focused on feature extraction now, but this paper applies improved topic model for text classification,which is an effective solution for the traditional LDA unsupervised features and incorporates the label information into a unified topic model, consequently enhancing the effectiveness of the model interpretability. Experimental results show that the classification performance of this method compared with mainstream discriminant classification algorithm has improved.