Analyzing the Influence Spread in Geo-Social Networks

Abstract

The emergence of geo-social networks as a social dynamic system with the potential to transmit locally relevant advertisements far and wide is opening up new avenues for viral marketing. The geo-social network setting presents a unique challenge to the influence spread problem because of the need to account for the spatial heterogeneity of nodes and nodes' connections. Additionally, from the standpoint of company managers, it is essential to strike a balance between the objectives of maximising the distribution of influence and minimising the expense of promotion. So, it's important to optimise for each of these goals simultaneously in a way that is seamless. In this research, we design a multio-bjective optimization-based influence spread framework for geo-social networks, providing decision-makers with a comprehensive picture of Pareto-optimal options. To accommodate a wide range of users, we first change our original problem into a weighted coverage problem, which is based on the reverse influence sampling (RIS) model. To this end, we present a greedy-based gradually approximation strategy and a heuristic-based particle swarm optimization approach to resolving this issue. The efficacy and efficiency of our suggested methods have been empirically validated through extensive tests on two real-world geo-social networks.