Main Article Content
The rise in social network data leads to extensive information propagation where the information shared without authenticity results in massive diffusion of rumor. Rumors during the present pandemic situation of COVID-19 create fear, anxiety and a negative impact on individuals. Identification of rumor sources helps control the undesirable effects of rumor diffusion in a social network. This research targets discovering the starting point of a rumor in the social network with improved accuracy and reduced network space. The proposed algorithm of TPSD achieved this by applying the search space reduction method and reverse propagation technique together. The network is examined using a monitor-based approach and rumor is diffused by a discrete-time susceptible-infected model and using incremental propagation delay. The incremental delay helps to detect the nominee partition precisely with the help of a partition-connected graph. In this nominee partition, rumors from the monitor nodes are reproduced in the reverse direction to identify the source. The experiment is performed in synthetic and real-world data collected from a semantic web of Twitter. The previous work shows the accuracy concerning distance error 0-4 hop distance. The experimental result illustrates that the actual source is identified within 0-1 hops in a real-world social network such as Twitter and Facebook. The experimental results reveal that our approach outperforms the current methods.
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.