基于半监督LDA的文本分类应用研究

Nowadays, it's a time of information explosion and exponential growth of data.Most of the data is unstructured, which bears a large number of important knowledge awaits us to find out. How to work out a convenient and quick way to process text classification is also a very important issue. LDA model is an unsupervised model, which can discover latent topics under unlabeled data.In order to more effectively find latent topics, a model based on semi-supervised LDA is proposed.By finding a topic set to discover more relevant topics. In addition, the LDA model and the semi-supervisedLDA model are applied to the text of feature extraction, comparing with other models.The experiments show that semi-supervised LDA model performance slightly better.