Influence maximization algorithm based on cross propagation in location-based social networks

The problem of influence maximization is one of the key issues in social networks. Most of the current studies focus on online social networks while ignoring offline interpersonal relationship networks. Fortunately, the cross propagation considers the characteristics of both the online social networks and offline interpersonal relationship networks, which is more suitable for the real scenarios. In this paper, we design a cross propagation model based on location-based social networks to establish a connection between online social networks and offline interpersonal relationship networks. Where the offline interpersonal relationships are mined by the similarity of POIs, which are based on the encounter characteristics. Then, an influence maximization algorithm based on cross propagation model is provided. The simulation results indicate that the propagation effect of influence in cross propagation networks is better than that only in online social networks, and the proposed algorithm has higher performances in terms of the running time and the sphere of influence.