9th IEEE International Conference on Collaborative Computing: Networking, Applications and Worksharing

Research Article

Silence Behavior Mining on Online Social Networks

Download22 downloads
  • @INPROCEEDINGS{10.4108/icst.collaboratecom.2013.254085,
        author={Qingbo Hu and Guan Wang and Shuyang Lin and Philip Yu},
        title={Silence Behavior Mining on Online Social Networks},
        proceedings={9th IEEE International Conference on Collaborative Computing: Networking, Applications and Worksharing},
        publisher={ICST},
        proceedings_a={COLLABORATECOM},
        year={2013},
        month={11},
        keywords={social network user behavior information propagation},
        doi={10.4108/icst.collaboratecom.2013.254085}
    }
    
  • Qingbo Hu
    Guan Wang
    Shuyang Lin
    Philip Yu
    Year: 2013
    Silence Behavior Mining on Online Social Networks
    COLLABORATECOM
    IEEE
    DOI: 10.4108/icst.collaboratecom.2013.254085
Qingbo Hu1, Guan Wang1, Shuyang Lin1, Philip Yu1,*
  • 1: UIC
*Contact email: psyu@uic.edu

Abstract

Keeping silence widely exists in human society and has been studied in social science for a long time. Similar to real social networks, in online social networks, after observing an event from their friends, users may not decide whether to share it at once due to different reasons. In influence propagation process, we observe that there are three states regarding to one's reaction on an event: activated state (shared), inactivated state (not shared) and silent state (take longer time to make decisions). Silent state is an intermediate status before turning into inactivated or activated state. In this paper, we provide a mathematical definition of “silence" based on the length of hesitating time before a user makes decisions. But, during the hesitation period, silent users behave exactly like the users who have entered the inactivated state. In order to differentiate them in this case, we develop a Similarity Interest (SI) model to identify silent users by quantifying the interest of users toward the event. Furthermore, comparing to real social networks, we reveal unique behavior of silent users in online social networks and use the Transient Influence Principle to explain it. At last, based on experimental results, we design a Diffusion with Silence (DS) model incorporating Similarity Interest model and two traditional diffusion models, in order to capture the silence behavior. Our experiment shows that the DS model can better depict the process of information propagation.