Identifying Influential Spreaders On a Weighted Network Using HookeRank Method

Influence maximization is a significant research problem that requires the selection of influential users who are capable of spreading information in the network such that it can reach to a large number of people. Many real-world networks like Road Network, Email Networks are weighted networks. Influence maximization on weighted networks is more challenging than an unweighted network. Many methods, such as weighted-degree rank, weighted-voteRank, weighted-eigenvalue rank, and weighted-betweenness rank methods, have been used to rank the nodes in weighted networks with certain limitations. In this manuscript, we propose a Hooke’s law-based approach named HookeRank method to identify spreaders in a weighted network. We model edge weights as spring constants. The edges present in the network are modeled as springs, which are connected in series and parallel. They elongate by a distance under the effect of a given constant force following Hooke’s law of elasticity, and this is the equivalent propagation distance between nodes in the network. The proposed the model finds relevant influential nodes, that can propagate the information to other nodes. A higher HookeRank score implies the greater influential capability of the node in the network. We compared our proposed algorithm with state-of-the-art models and found that it performs reasonably well on real-life data-sets using epidemic spreading Susceptible-Infected-Recovered model.