No abstract

No abstract

No abstract

No abstract

In this paper, we introduce the Auto-Annotation LDA models (aaLDA), a statistical model of non-labeled documents. This model generates the annotation of LDA automatically. We derive the annotation of LDA using a k-means methods combined with a pre-processing of the corpus. In this paper, we use aaLDA models to categorize "zhongwenshilei" corpus, which is a famous Chinese corpus. Then we make a compare with the traditional LDA methods.