Improved link prediction technique using community detection algorithm

Abstract

As social technology has connected substantial number of users together to interact and share of millions of information across the network. It is essential to foresee the frequent future links of the users which are connecting together and has the future possibility to connect. The social hub network is dynamic as it changes the structure at different timestamps. The network obtained at time t is varied at time t+1. In order to predict the ongoing changes on network, graph embedded techniques are used to obtain an unsupervised graph with different parameters of nodes and edges which can be used in machine learning methods. In this paper, we device a community detection algorithm with edge betweenness, closeness, betweenness, degree, hubs and authority parameters to predict the efficiency of the model with the dataset and visualize a graph network to determine the centrality of the network model.