Rumor correction maximization problem in social networks

Admittedly, innovations can spread rapidly in online social networks, while the of malicious rumors lead to a series negative consequences. Therefore, it is necessary take effective measures limit influence information. In reality, people will become an adopter after being influenced by their friends. Meanwhile, they be more likely follower if have received relevant information advance. Motivated these observations, we study rumor correction maximization problem using both seed and boost nodes. We first focus on nodes propose Boosting Rumor Correction Maximization (BRCM) under Independent Cascade model. prove that BRCM NP-hard, objective function non-submodular. To handle it, devise efficient algorithm with data-dependent approximation ratio. explore nodes, Seed Selection Minimum are proposed, respectively. Accordingly, design two algorithms. Finally, extensive empirical results three networks manifest efficiency our approaches show superiority over other baselines.