Information influence measurement based on user quality and information attribute in microblogging

Abstract

In microblogging, information influence is very hard to be measured. Because there are a lot of spam messages, and these messages were retweeted and commented by a lot of spam users in microblogging. Early methods for a message influence measurement mainly use message retweet and comment times. But they don't consider the quality of user, who takes part in the information diffusion. In this paper, we propose a novel measurement method based on user quality and information attribute. Experimental results show that our method can measure the influence more accurately and reduce the spam message influence in the dataset more effectively.