An improved LDA algorithm for text classification

Latent Dirichlet Allocation is a classic topic model which can extract latent topic from large data corpus. This model assumes that if a document is relevant to a topic, then all tokens in the document are relevant to that topic. In this paper, we present an algorithm called gLDA for topic text classification by adding topic-category distribution parameter to LDA, which can make the document generated from the most relevant category. Gibbs sampling is employed to conduct approximate inference, and experiment results in two datasets show the effectiveness of this method.