Skip to main content

Table 2 The significance of multiple observables across the enron and stack-overflow data sets

From: Annotated hypergraphs: models and applications

   Non-preserving Preserving
DataFeatureOrig. ValAvg. Val.z-score Avg. Val.z-score
enronConnected components1.001.000.00 1.000.00
 Local role MI0.140.03red15.68 0.06red9.78
 Local role Entropy1.241.36blue-12.47 1.34blue-8.24
 Node role Entropy0.991.33blue-19.55 0.990.00
 Weight. degree entropy5.875.83red3.00 5.87-0.39
 Weight. eigenvector entropy4.025.84blue-154.46 5.88blue-55.09
 Weight. pagerank entropy6.366.11red26.47 6.14red13.74
stack-overflowConnected components1168.001166.710.06 1041.01red6.49
 Local role MI0.220.23blue-7.69 0.22blue-3.88
 Local role Entropy0.870.88blue-4.73 0.88blue-6.32
 Node role Entropy0.050.07blue-19.78 0.050.00
 Weight. degree Entropy13.8913.81red14.97 13.84red8.71
 Weight. eigenvector entropy7.958.19-0.38 7.281.24
 Weight. pagerank entropy14.2814.23red18.13 14.24red13.88
  1. Significance scores coloured red are two standard deviations larger in the original data than the null model. In contrast, those coloured blue are two standard deviations smaller. The results for all data sets are presented in Fig. 6