INFLUENTIAL NODES PREDICTION USING LINKS INFORMATION IN SOCIAL NETWORKS
Keywords:
social systems, influential nodes, statistical mechanicsAbstract
Identifying influential nodes and measure the influence of nodes in social networks, has been inspired by analogies between social behavior and statistical mechanics. Social interactions among humans create complex networks, and despite an increase of online communication, the interaction between physical proximity remains a fundamental way for people to connect. Here we can initiate a research on the foundations of ranking nodes, a fundamental ingredient of analyzing social systems. In order to understand the essence and the exact rationale of node ranking algorithms we suggest the axiomatic approach of agent based model taken in the formal theory of social choice. Based on essential factors of influence propagation (such as the location and neighborhood of source node, propagation rate) and network invulnerability, we propose a novel strategy to search the influential nodes in terms of outgoing and ingoing links to the node. The aim of this work is to identify the influential nodes as they affect the hierarchical structures of the network. By analyzing the data and describing how these nodes affect the network structure, we aim to obtain new tools and methodology which will help us to describe how networks grow and fall apart in smaller structures, which have similar features with the large network, but different dynamics. In order to characterize this phenomenon and explore the correlation between collective behaviors and locally interacting elements, we use statistical methods and visualization software as a combined approach to understand the behavior of the network for a given behavior of the influential nodes that we use to recreate our network. The results of our research on real-world networks’ dataset show that the proposed method outperforms state-of-the-art influence algorithms.
References
Arnoux, T., Tabourier, L., and M. Latapy. (2018). Predicting interactions between individuals with structural and dynamical information. CoRR
Arnoux, T., Tabourier, L. and M. Latapy. (2017). Combining structural and dynamic information to predict activity in link streams. In ASONAM
Albert, R. & Baraba´si, A.-L. (2002). Statistical mechanics of complex networks. Rev. Mod. Phys. 74, 47–97
Backstrom, L., & Leskovec, J. (2011). "Supervised Random Walks: Predicting and Recommending Links in Social Networks", ACM Internation Conference on Web Search and Data Mining (WSDM)
Barrat, A., Barthelemy, M. & Vespignani, A. (2008). Dynamical Processes on Complex Networks (Cambridge University Press, 2008).
Brandes, U. (2001). A Faster Algorithm for Betweenness Centrality. Journal of Mathematical Sociology 25(2):163-177, https://doi.org/10.1080/0022250X.2001.9990249
Brandes, U. (2008). On Variants of Shortest-Path Betweenness Centrality and their Generic Computation. Social Networks 30(2):136-145,.https://doi.org/10.1016/j.socnet.2007.11.001
Baytas, I. M., Xiao, C., Zhang, X., Wang, F., Jain, A. K., & Zhou. J. (2017). Patient subtyping via time-aware lstm networks. In KDD,
Cheng, J., Bernstein, M., Danescu-Niculescu-Mizil, C., & Leskovec, J. (2017). Anyone can become a troll: Causes of trolling behavior in online discussions. In CSCW, [10] J. Cheng, C
Goyal, A., F. Bonchi, L.V.S. (2010). Lakshmanan. Learning influence probabilities in social networks. In Proc. WSDM
Leskovec, J., & Krevl, A. (n.d.). http://snap.stanford.edu.data
Liben-Nowell D., & J. Kleinberg. (2003). The link prediction problem for social networks. In IKM ’03, pages 556–559
Newman, M. E. J. (2010). Networks: An Introduction (Oxford University Press)
