Influence Maximizing in a Diffusion Model based on Propagating Negative Ideas in Social Networks

One of the interesting fields in social networks is Influence Maximization (IM). Influence maximization is the problem of finding k seed nodes that maximizes the influence in a network under a diffusion model. IM is a NP-hard problem which is proven that a greedy based approach, finds a solution 63% near the optimal solution under some circumstances [1]. Many algorithms and heuristics were proposed to either optimize the seed selection process timeor introduce new diffusion models and study real behaviors in a social network. Time constraints and Trust are two important factors in today’s human lives. Diffusion models having time constraint consider the time a node or individual takes to propagate and infect its friend. Models based on trust relations, study the effects these behavior has on a network. Despite good and successful studies on these two factors, only a few of them studied both Time constraint and Trust together in a diffusion model. Recent proposed models tried to simulate the real world communication behaviors, however the propagation of negative ideas through a network which its entities trust or distrust each other is not considered. In this paper we propose a new diffusion model based on both trust and time and the propagation of negative ideas. We used a greedy based framework to select seed nodes and compared our results with other related methods and the results were promising

Keywords:

Social Network , Influence Maximization , Diffusion Models , Trust , Negative ideas , Time