Genetic algorithm based rumor mitigation in online social networks through counter-rumors: A multi-objective optimization

摘要

Immense use of social media platforms results in quicker and wider dissemination of not only legitimate information, but also rumors which may cause mental stress and reduce the productivity of the society. One of the important ways to counteract the rumors in online social networks (OSNs) is to spread counter-rumors. In this work, we propose a Precedence based Competitive Cascade (PCC) model for the propagation of competing rumor and counter-rumor cascades. We have presented a model to compute a belief based precedence value by which a user chooses to believe the rumor or counter-rumor received during information propagation. The influence of the rumor and the counter-rumor cascades in the OSNs is analyzed by considering a neighborhood based propagation approach. Another challenging issue which has been addressed in this work is to select the minimal seed set of users for the initiation of the counter-rumor so as to reduce the message overhead in the application. This has been formulated as a multi-objective optimization problem to select the minimal set of seed users for counter-rumor to minimize the effect of the rumor. We design a Decomposition based Multi-objective Genetic (DMOG) algorithm to solve the problem. Experiments are conducted on real-world data sets to evaluate the efficacy of the proposed PCC model and DMOG algorithm by considering important parameters such as the precedence, budget and time delay.