Exploring Inter-connectivity Link Prediction: An Insight from Social Network Science

Abstract

Physical and computational science communities are becoming interested in link prediction for complicated networks. Thus, various algorithms can be used to extract missing data, detect erroneous interactions, assess network evolution mechanisms, and so on. The contributions from computational science views and approaches, such as arbitrary methods and maximum likelihood methods are highlighted in this paper, which summarizes current advances in link prediction algorithms. One of the frequently discussed research topics in the area of social network analysis is link prediction. Numerous practical applications of this issue may arise in the future, including modelling recommender systems, fraud detection, stock prediction, and more. One can comprehend the dynamics of network evolution if the absent links or the links that will appear in the future are foreseen. One may undoubtedly develop superior choice models for the business strategy based on the observed patterns of structural changes, which can have high performance value and lower market risks. In this paper, the author has presented overview of background of link prediction issue in Social Network Science and also presented the link prediction methods for modern research in network science