Learning-based link prediction analysis for Facebook100 network

In social network science, Facebook is one of the most interesting and widely used networks media platforms. Its data has significantly contributed to evolution research link prediction techniques, which are important tools in mining analysis. This paper gives first comprehensive analysis on Facebook100 network. We stu- dy performance evaluate multiple machine learning algorithms different feature sets. To derive features, we use embeddings topology-based techniques such as node2vec vectors similarity metrics. addition, also employ node- -based available for network, though rarely found other datasets. The adopted approaches discussed results clearly presented. Lastly, compare review applied models, where overall classification rates