 Link prediction is a technique used to identify potential connections between entities in a given dataset. Existing methods rely on network properties such as node centrality, edge density, and weight distribution, but these methods can be inaccurate due to their limited scope. Our approach uses PageRank, a measure of importance for webpages, to identify influential nodes within a network. We then use global clustering coefficients to form groups of similar nodes and finally use maximum likelihood estimation to predict optimal links between them. Experiments with real-world datasets show that this approach outperforms other existing methods. This article was authored by Lakshmi S. Nair, Swaminathan J. Rahman, and Saipavankrishnan Agam.