Such example implementations are particularly applicable to big data analytics in which there is a large volume of data and the data includes real world data that may be noisy. For example, for data utilized in determining gene expression, genes are related to different conditions, and the bipartite network involves a first type of node (genes) and a second type of node (conditions/diseases that can occur). In practice, conducting experiments for every type of gene combination is impractical because the conditions and genes are too numerous. Through example implementations, analysis can be done on such a bipartite network to identify which combination of genes are likely to cause which conditions through missing link detection, and then the users can focus on those particular gene/condition experiments.

In another example implementation involving drug discovery, the bipartite network can involve different types of molecules and different types of conditions (e.g., side effects, disease treatment efficacy). Drug discovery can involve an inordinate amount of experimentation as there can be too many different types of molecules and conditions that the user may be interested in. By applying the algorithms as described herein, the causation between drug molecule combinations and conditions can be more accurately determined than link prediction algorithms of the related art, and thus the user can focus the drug experiments to test such conditions accordingly.