Tibetan Weibo User Group Division Based on Semantic Information in the Era of Big Data

Algorithm 3

Improved LDA topic model based on Weibo users.

Input:News user text data: User_Document1

Output:Weibo users’ topic distribution: User_Topics_Matrix

1: get LDA_Model by Algorithm 2

2: for alld, do

3: predict topics of d by LDA_Model

4: combine short texts of microblogs belonging to the same topic to long texts Document2.

5: end for

6: for alld, do

7:  predict topics of d by LDA_Model

8:  update of LDA_Model

9: end for

10: for alluser, do

11: predict topics of d by LDA_Model

12: get Document_Topics_Matrix of d

13: end for

14: returnUser_Topics_Matrix