University of Leicester
Browse
A Social Sensing Model for Event Detection and User Influence Discovering in Social Media Data Streams.pdf (544.16 kB)

A Social Sensing Model for Event Detection and User Influence Discovering in Social Media Data Streams

Download (544.16 kB)
journal contribution
posted on 2019-10-14, 16:40 authored by Lei-Lei Shi, Lu Liu, Yan Wu, Liang Jiang, John Panneerselvam, Roy Crole
Online social networks (OSNs) have emerged as a major platform for sharing information through social relationships and are one of the major sources of big data. Social networks can even accommodate sharing of live streaming data among the connected users. However, social information on social networks is often locally exploited rather than capturing the changes in the entire network over time. Obtaining user's influence statistics is limited only in their local vicinity, which may not facilitate capturing the changes in the user and post influences across the entire network, thereby resulting in lower accuracy while measuring user's topical influence. Moreover, low-influence users always exist in the network publishing low-quality posts. With the objectives of accurately capturing highly influential users and posts, this article proposes a novel dynamic social sensing model, named dynamic PageRank (DPRank) model, to evaluate the dynamic topical influence of the users of social information on social networks during the social information evolution. We deploy our proposed model to real-world Twitter data sets, which demonstrates the effectiveness of our proposed model against notable existing methods while identifying the true influence of users and posts in a dynamically evolving social network.

Funding

This work was partially supported by the Natural Science Foundation of Jiangsu Province under Grant BK20170069, UKJiangsu 20-20 World Class University Initiative programme, and UK-Jiangsu 20-20 Initiative Pump Priming Grant. Lu Liu is the corresponding author.

History

Citation

IEEE Transactions on Computational Social Systems, 2019

Author affiliation

/Organisation/COLLEGE OF SCIENCE AND ENGINEERING/Department of Informatics

Version

  • VoR (Version of Record)

Published in

IEEE Transactions on Computational Social Systems

Publisher

Institute of Electrical and Electronics Engineers (IEEE)

eissn

2329-924X

Acceptance date

2019-08-20

Copyright date

2019

Available date

2019-10-14

Publisher version

https://ieeexplore.ieee.org/document/8859629

Language

en

Usage metrics

    University of Leicester Publications

    Categories

    Exports

    RefWorks
    BibTeX
    Ref. manager
    Endnote
    DataCite
    NLM
    DC