NESA SecGraph

Demo-Group1-1:

The demo took a plain text social network graph dataset and anonymized it with ID removal which took out the names and made the directed graph undirected.

Demo-Group1-2:

The demo took a graph dataset and anonymized it by AddDel Edge Editing that changed 20% of the edges. Then it used reconciliation attack from the paper An efficient reconciliation algorithm for social networks with some randomly picked seeds and a minimum matching score of 50 to try to de-anonymize the anonymized graph. At the end of the attack, we found out that 246/270, ~91%, of the nodes were correctly matched.

Demo-Group1-3:

The demo took the origional graph and the anonymized graph from Demo 2 and calculated the Degree utility remains after the anonymization.