Statistical Evaluation on Subgraphs

[sebastian9991]

Include a statistics folder that includes tests between credible and non-credible subnetworks that we have defined. Also, include a full experiment that plots and saves results.

• Clustering
• PageRank
• Indegree/Outdegree distribution
• Similar statistics
• Other graph-level statistics
• average degree, max degree, min degree
• etc.

Auxilary:

• Include an experiment to show the correlation between PageRank scores and our PC1 gold-standard scores.

[sebastian9991]

I'll attach a snippet of code in this image to give an idea of the usage of Jacob and Andy's TGM code as it relates to us.

[ekmpa]

I feel like the 'auxiliary' point is quite important. Not exclusively for pagerank, but we'll probably want to explore a few algorithms on the whole network and evaluate with regards to our labels