Brand posts on Sina Weibo: Predicting popularity

Sina Weibo has become one of the most popular sites in China. Attracted by the huge commercial potential on Sina Weibo, many companies create their own Weibo accounts to enhance their online influence. Companies deliver information and interact with customers through placing brand posts. In this paper, some features are extracted from properties of brand posts and evaluate the roles of these features in predicting the popularity of brand posts. The reposts count of a brand post is regarded as an important indicator of online popularity because it reveals both the public attention and the spread of a post. Linear regression is used to figure out the relationship between the selected features and the reposts counts. In our experiments, it is possible to predict ranges of popularity on Sina Weibo with an overall 81% accuracy.