Centrality measure based approach for detection of malicious nodes in twitter social network

  • Authors

    • Krishna Das
    • Smriti Kumar Sinha
    2018-09-22
    https://doi.org/10.14419/ijet.v7i4.5.21147
  • Malicious Nodes, Centrality Measures, Clustering Coefficient, Anomaly Behavior, Information Diffusion

  • In this short paper, network structural measure called centrality measure based mathematical approach is used for detection of malicious nodes in twitter social network. One of the objectives in analysing social networks is to detect malicious nodes which show anomaly behaviours in social networks. There are different approaches for anomaly detection in social networks such as opinion mining methods, behavioural methods, network structural approach etc. Centrality measure, a graph theoretical method related to social network structure, can be used to categorize a node either as popular and influential or as non-influential and anomalous node. Using this approach, we have analyzed twitter social network to remove anomalous nodes from the nodes-edges twitter data set. Thus removal of these kinds of nodes which are not important for information diffusion in the social network, makes the social network clean & speedy in fast information propagation.

     

     

     

  • References

    1. [1] Barnett V. and Lewis T.,â€Outliers in Statistical Dataâ€, third ed. John Wiley & Sons, 1994, Chichester, UK.

      [2] Beauchamp MA.,â€An improved index of centralityâ€, Behav Sci, 10(2), 161–163, 1965.

      [3] Bimal Viswanath, M. Ahmad Bashir, Mark Crovella and Saikat Guha, “Towards Detecting Anomalous User Behavior in Online Social Networksâ€, 23rd USENIX Security Symposium, August 20–22, 2014, San Diego, CA.

      [4] Bonacich P. and Lloyd P.,â€Eigenvector-like measures of centrality for asymmetric relationsâ€, textitSocial Networks, 23(3), 191–201, 2001.

      [5] Chandola V., Banerjee A. and Kumar V.,â€Anomaly detection for discrete sequences: a surveyâ€, IEEE Trans. Knowl. Data Eng., 24 (5), 823–839, 2012.

      [6] Chandola V., Banerjee A. and Kumar V.,â€Anomaly detection: A surveyâ€, ACM Computing Survey, (CSUR) 41 (3), 15, 2009.

      [7] David Savage, Xiuzhen Zhang, Xinghuo Yu, Pauline Chou and Qingmai Wang, “Anomaly detection in online social networksâ€, Social Networks, 39, 62–70, 2014.

      [8] Freeman L. C.,â€Centrality in social networks: conceptual clarifica- tionâ€, Social Networks, 1(3), 215–239, 1979.

      [9] Hodge V.J. and Austin J.,â€A survey of outlier detection methodolo- giesâ€, Artif. Intell. Rev., 22(2), 85 –126, 2004.

      [10] Katz L.,â€A new status index derived from sociometric analysisâ€, Psychometrika, 18(1), 39 –43, 1953.

      [11] Landherr A., Friedl B. and Heidemann J., (2010),â€A Critical Review of Centrality Measures in Social Networksâ€, Business & Information Systems Engineering, 2(6), 371–385, 2010.

      [12] Li H., Cui J.T. and Ma J.F.,â€Social influence study in online networks: a three-level reviewâ€, J. Comput Sci Technol., 30(1), 184 –99, 2015.

      [13] Li K-L, Huang H-K, Tian S-F, and Xu W., â€Improving one-class SVM for anomaly detectionâ€, Int Conf Mach Learn Cybernetics,30(5), 77 81, 2003.

      [14] Nieminen J.,â€On the centrality in a graphâ€, Scand J Psychol, 15(1), 332 –336, 1974.

      [15] Ravneet Kaur, Mankirat Kaur and Sarbjeet Singh,â€A Novel Graph Centrality Based Approach to Analyze Anomalous Nodes with Negative Behaviorâ€, International Conference on Information Security & Privacy (ICISP2015), 11 –12 December 2015, Nagpur, INDIA.

      [16] Sabidussi G.,â€The centrality index of a graphâ€, Psychometrika, 31(4), 581–603, 1966.

      [17] WIREs Comput Stat 2015, 7:223–247. doi: 10.1002/wics.1347.

      [18] Nicholas A. Heard ET. al., “Bayesian anomaly detection methods for social networksâ€, The Annals of Applied Statistics, 2010, Vol. 4, No. 2, 645 – 662.

  • Downloads

  • How to Cite

    Das, K., & Kumar Sinha, S. (2018). Centrality measure based approach for detection of malicious nodes in twitter social network. International Journal of Engineering & Technology, 7(4.5), 518-521. https://doi.org/10.14419/ijet.v7i4.5.21147