Research on User Gender Prediction of Chinese Microblog Based on Short Text Analysis

Abstract

Gender information of Weibo users is taken as the research object. After pretreatment of user's text information, the linguistic features and topic features are extracted. For each user, building space vector model based on linguistic features, topic features and both. Then using SVM machine learning algorithm to construct classifier for gender prediction. Experiments show that the linguistic features and topic features can predict the sex of the users accurately, and the effect is superior to other features used in gender prediction.