Epidemiological Reverse Influence Maximization in Social Networks with Negative Influencing

Influence Maximization (IM)-based profit maximization estimates a seed set of users that maximizes the profit in social networks. On the other hand, Reverse Influence Maximization (RIM) finds the seeding cost of the same process. The profit is measured by the maximum number of users that can be activated by the seed users and the seeding cost is given by the minimum number of users which should be activated in order to activate the seed or target users. The major drawback of most of the existing IM models is that the seed users are assumed to be activated initially. However, seed users may also be influenced by some other icon users they follow. Moreover, none of the state-of-the-art models has considered negative influencing. Thus, in this paper, we propose an epidemic model-based RIM model with negative influencing (EN-RIM) for seeding cost optimization. The simulation shows that the EN-RIM model outperforms many existing models.