Graph Sparsification and Semi-Supervised Learning: an Experimental Study

Abstract

This study focuses on comparing the various graph sparsification methods that have been devised and tests the efficiency when compared to one another. And on the latter side of the project, semi-supervised learning algorithms are trained on a combination of labeled and unlabeled data. Hence, different sparsification methods are explored and the effect of such methods in Semi Supervised Graph Based Algorithms are evaluated.