Behavior of Retweeting in Microblogging Service: A Study of Information Propagation from User Attributes
Jichao Zhao,Xun Liang,Zhiming Xu,Yi Wang,Yu Liu
DOI: https://doi.org/10.1109/icsssm.2016.7538614
2016-01-01
Abstract:The behavior of retweeting, or information propagation, in microblogging service is a primary task for many research areas, and has aroused the attention from both academy and industry. In the former research on the behavior of retweeting, the network structure and/or the user influence have been widely employed, while the basic attributes of the users are neglected. In this paper, the behavior of retweeting, especially on information propagation, were explored from a vast of user attributes, including time bucket, time vector, the number of fans, the increased number of fans in unit time, the increased number of tweets (refers to the content of the microblog) in unit time, gender and the sign of V. We employ the power of the fitting method, explore the effectiveness of the user attributes on the tweet propagation, and thus explain the complex behavior of the information propagation in microblogging services. In the experiment, we adopted different retweeting quantity limits, and increased the different attributes in turn, studied the combination of different attributes effects on the velocity of propagation. The results show that the features of time and fans play a significant role in the tweet propagation, while other attributes have much less effectiveness. Our empirical testing also conveys that information diffusion is a complex and dynamic process influenced by a variety of attributes and many other aspects.