Graph Sampling Results

Boxplots showing the quality of the sampled subgraphs. The quality/relevance is based on the recall of a given set of queries for this dataset.

Whenever we can claim a statistically significant better recall than one of our baselines, we show this using + or * signs. Our significance calculations are public as well on github: https://github.com/Data2Semantics/GraphSampling/blob/master/bin/significance/getSignificance.R