Abstract:Identifying rumor sources in online social networks (OSNs) plays a crucial role in controlling the spread of rumors and mitigating their damage. However, the community structure of OSNs and timeliness of rumors increase the complexity of accurately characterizing rumor-spreading behavior, making it extremely challenging to identify the source of rumor in OSNs. Conventional studies on rumor source identification often overlook the community structure of OSNs and timeliness of rumors. This paper proposes a rumor sources identification framework that take these two aspects into consideration. First, we transfer the snapshots of a dynamic OSN into a community-structured dynamic OSN through 4 meticulously designed phases. Second, instead of considering a constant influence of rumors in traditional techniques, we introduce a time-varying dynamic propagation parameter to quantify the timeliness of rumor, and apply the propagation parameter to establish a microscopic rumor spreading model. This process addresses the issue of modeling the timeliness of rumor. Third, we adopt sensor-based observation techniques to collect the propagation information of rumor, and a hybrid sensor deployment strategy is designed to improve the efficiency of information collection. Fourth, we propose an algorithm for identifying single and multiple rumor sources in community-structured dynamic OSN, this algorithm initiates with the utilization of some criteria to analyze the gathered propagation information and estimate the suspicious communities that harbor the source of rumors. Subsequently, it applies maximum likelihood estimation method within each community to determine the ultimate source of the rumor. Experimental results on real-world and synthetic networks indicate that our method can accurately identify the real rumor sources. To the best of our knowledge, the proposed method is the first that can be used to identify rumor sources in community-structured dynamic OSNs.

Online Social Network Information Source Identification Algorithm Based on Multi-Attribute Topological Clustering

Comprehensive Influence of Topological Location and Neighbor Information on Identifying Influential Nodes in Complex Networks.

Community Detection Based on Density Peak Clustering Model and Multiple Attribute Decision-Making Strategy TOPSIS

Identify information sources with different start times in complex networks based on sparse observers

Identifying Traffic Clusters in Urban Networks Based on Graph Theory Using License Plate Recognition Data

Multi-source detection based on neighborhood entropy in social networks

Attribute-Enhanced De-anonymization of Online Social Networks

Locating Influence Sources in Social Network by Senders and Receivers Spaces Mapping

A fast multi-level algorithm for community detection in directed online social networks

Multi-factor information matrix: A directed weighted method to identify influential nodes in social networks

A novel method of spectral clustering in attributed networks by constructing parameter-free affinity matrix

Enhancing topology adaptation in information-sharing social networks

Node Multi-Attribute Network Community Healthcare Detection Based On Graphical Matrix Factorization

Timeliness-aware rumor sources identification in community-structured dynamic online social networks

Efficient Time Series Clustering And Its Application To Social Network Mining

Important node identification for complex networks based on improved Electre Multi-Attribute fusion

Identification of high-influential nodes based on multi-order neighbor propagation metrics and topological features

Cross Multi-Type Objects Clustering in Attributed Heterogeneous Information Network

Mutual Clustering Coefficient-based Suspicious-link Detection approach for Online Social Networks

Relation strength-aware clustering of heterogeneous information networks with incomplete attributes

Influence Identification Method of Complex Network Nodes Based on Local and Global Structure