Modeling information diffusion in online social networks using a modified forest-fire model.

Information dissemination has changed rapidly in recent years with the emergence of social media which provides online platforms for people worldwide to share their thoughts, activities, emotions, and build social relationships. Hence, modeling information diffusion has become an important area of research in the field of network analysis. It involves the mathematical modeling of the movement of information and study the information spread pattern. In this paper, we attempt to model information propagation in online social networks using a nature-inspired approach based on a modified forest-fire model. A slight spark can start a wildfire in a forest, and the spread of this fire depends on vegetation, weather, and topography, which may act as fuel. On similar lines, we labeled users who haven't joined the network yet as , existing users as , and information as . The spread of information across online social networks depends upon users-followers relationships, the significance of the topic, and other such features. We introduce a novel state to the traditional forest-fire model to represent non-spreaders in the network. We validate our method on six real-world data-sets extracted from Twitter and conclude that the proposed model performs reasonably well in predicting information diffusion.