Identifying Influential Nodes for Information Propagation in Social Networks

Milad Ebrahimkhani¹ Zainabolhoda Heshmati²

1) Msc Student, Faculty of New Sciences and Technology, University of Tehran, Tehran, Iran.
2) Assistant Professor, Faculty of New Sciences and Technology, University of Tehran, Tehran, Iran.

Publication : 5th. International Congress On Engineering, Technology and Innovation(eticong.com)
Abstract :
Today, widespread information propagation is one of the significant points in most social networks. To this end, an important challenge in these networks is to identify the influential individuals in information propagation. But identifying is targeted when it is based on different topics. In this study, a novel method based on topic and time is introduced for identifying the influencers in Twitter. In this method, the contents of the tweets are used in addition to the topological features. Using semantic features in analyzing the contents of the tweets increases the accuracy of calculating the influence of the network nodes. In the proposed method, after preprocessing and developing the graph structure of the tweets, explicit and implicit influence are calculated; finally, their combination is used. Since the proposed method represents influence based on topic at different times, a proper dataset is collected and evaluated. The evaluation results based on standard measures show excellent performance of the proposed method.
Keywords : social networks influencers topological features explicit influence implicit influence