Influence Maximisation in Social Networks Using Neighbourhood Degree Based Vote-Rank Approach

Group of people sharing their interests and opinions among each other form a social network. Nowadays, this network is enlarging at a rapid rate. Therefore, spreading information in these large networks is a non-trivial task. Influence Maximisation steps up as a possible solution to this complex problem. In Influence Maximisation, we find the $k$ most influential nodes from a set of $n$ nodes such that by converging the information reach on these $k$ nodes, the propagation of information is maximised. There have been several classical centrality measures introduced in the past few years, but there is still scope for improvement. In this paper, we propose ‘neighbourhood degree based vote-rank’ (NDVR) method where each node votes for its neighbour nodes. We initialise each node with a vote value, which is calculated as the average degree of the neighbour of that node. The value of received votes sums up as election score. The node with the highest election score wins the round and is declared influential node. This is followed by an updation phase where the voting value of the neighbour nodes which voted for the influential node in the previous round is decreased depending on the initial vote value and the average degree of the network. This technique of influence maximisation ensures that the regions covered by the previously selected nodes have minimum intersection with the region covered by the current selected node. The proposed method performs favourably against state-of-the-art techniques on several benchmark datasets.