A theoretical review on multiplex influence maximization models: Theories, methods, challenges, and future directions

• Published in: Expert systems with applications 2025-03, Vol.266, Article 125990
• Main Authors: Achour, Oumaima, Romdhane, Lotfi Ben
• Format: Article
• Language: English
• Subjects: Influence maximization; Multiplex social networks; Propagation models; Seed users
• ISSN: 0957-4174
• DOI: 10.1016/j.eswa.2024.125990

Online social networks (OSNs) have become an integral part of our daily lives, shaping the way social relationships evolve. Influence maximization (IM) in OSNs has been widely studied by various researchers during the last two decades due to its wide range of applications including viral marketing, public health, recommendation systems, disease spread and prevention, etc. It is known as the problem of selecting a small set of users with size ≤k known as seed users from a social network that can maximize the spread of information, influence, or behavior within the network. Recently, online users have joined multiple OSNs simultaneously such as Facebook, Twitter, and Instagram, thereby creating a new complex environment for the IM problem called multiplex network. Therefore, studying the IM problem in multiplex social networks, where the same set of users engage in various social networks simultaneously represents a new research direction that researchers have started to explore. While numerous surveys have explored IM algorithms from different perspectives, they primarily focus on traditional methods designed for single networks, overlooking multiplex networks. To fill this gap, in this paper, we present a theoretical review of recent IM algorithms and solutions proposed to solve the IM problem in multilayer networks, with a particular focus on multiplex networks. To build the foundation of the IM problem in multiplex networks, we start by presenting the IM problem in its basic form and reviewing the well-used propagation models and their extensions in single-layer networks to investigate then how they are adapted to cope with the diffusion process within multiplex networks. Next, we present the multiplex IM problem and provide a comprehensive taxonomy of the different existing algorithms proposed to solve it. Afterward, a comparative analysis illustrated by a comprehensive table outlining the strengths and weaknesses of each model is presented. Finally, diverse applications of multiplex IM, highlighting emerging trends and future directions that researchers can consider within this topic are provided. •The multiplex influence maximization problem is presented.•Taxonomy of existing IM models in multiplex social networks is provided.•A comparison of existing multiplex IM methods according to a clear categorization is presented.•New trends for detecting influential nodes are also presented.