Link Prediction using Machine Learning Algorithms
Meng Zhang · Zenodo (CERN European Organization for Nuclear Research) · 2020
With the advent on Internet, research on social network has improved in a rapid pace. In the context of Social Network Analysis (SNA), link prediction has become an important research direction. In this paper, we use supervised learning along with social network metrics to improve the accuracy of the link prediction task. We used multiple machine learning algorithms that are commonly used for prediction task. We also improved the classifiers features by extracting and adding multiple SNA metrics such as centrality metrics. We analyzed our method on a dataset of 2000 authors that occasionally collaborated with each other to write papers. Our analysis showed that enhancing the feature list by SNA metrics increased the accuracy of the prediction task.