Diversified Budgeted Influence Maximization in Dynamic Social Networks

Influence maximization is a fundamental problem in network analysis, which attempts to identify a subset of nodes that maximizes the spread of influence/information. This problem has applications in various fields such as social networks, viral marketing, advertisements, and political campaigns, where understanding and exploiting network dynamics lead to effective strategies to promote behaviors, products, or ideas. The goal is to strategically select seed nodes to maximize the overall impact or adoption of an idea, behavior, or product in the network. Most of the existing IM algorithms give equal cost to selecting nodes and maximizing the active nodes, which overlook the number of influenced communities. To make it applicable to real-world applications, this article presents a diversified budgeted influence maximization (DBIM) algorithm for dynamic social networks. The DBIM algorithm considers the different costs of the nodes and maximizes the number of communities. This work proposes an objective function for diversification and presents an algorithm that finds the seed set utilizing the suggested strategy. Further, we show the monotone, submodular, and NP-hardness properties of the objective function. The proposed work experimentally shows the results on eight datasets and concludes that the proposed algorithm outperforms the activated nodes and the number of communities on all the datasets.