SecGraph is designed for both academia and industry. On one hand SecGraph provide an uniform platform to test new graph data algorithms. On the other hand, SecGraph provides a plat form for data owners to test the privacy of their data before release.

SecGraph is created in Java thus theoretically could support any platform that one can run a JVM on. However while implementing the graphic user interface(GUI), we used an open sourced library “SWT (The Standard Widget Toolkit)” thus to use the GUI you must have a platform with SWT support.

SecGraph currently contains the 15 Structured Data de-anonymization techniques discussed in the paper which includes:

1. The Walk-Based active attack from Wherefore art thou r3579x? anonymized social networks, hidden patterns, and structural steganography
2. The Cut-Based active attack from Wherefore art thou r3579x? anonymized social networks, hidden patterns, and structural steganography
3. The passive attack from Wherefore art thou r3579x? anonymized social networks, hidden patterns, and structural steganography.
4. The attack from De-anonymizing social networks
5. The attack from Link prediction by de-anonymization: How we won the kaggle social network challenge.
6. The attack from Community-enhanced de-anonymization of online social networks
7. The distance vector attack from Deanonymizing mobility traces: Using social networks as a side-channel
8. The random spanning tree attack from Deanonymizing mobility traces: Using social networks as a side-channel
9. The Recursive Subgraph Matching attack from Deanonymizing mobility traces: Using social networks as a side-channel.
10. The attack from A bayesian method for matching two similar graphs without seeds
11. The attack from On the performance of percolation graph matching
12. The De-Anonymization attack from Structure based data de-anonymization of social networks and mobility traces
13. The Adaptive De-Anonymization attack from Structure based data de-anonymization of social networks and mobility traces
14. The attack from An efficient reconciliation algorithm for social networks
15. The attack from Structural data de-anonymization: Quantification, practice, and implications

SecGraph current contains 12 graph utility metrics and 7 application utility metrics as detailed in the paper.
Graph Utility Metrics:

1. Degree
2. Join Degree
3. Effective Diameter
4. Path Length
5. Local Clustering Coefficient
6. Global Clustering Coefficient
7. Closeness Centrality
8. Betweenness Centrality
9. Eigen Vector
10. Network Constraint
11. Network Resilience
12. Infectiousness

Application Utility Metrics:

1. Role eXtraction
2. Reliable Email
3. Influence Maximization
4. Minimum-sized Influential Node Set
5. Community Detection
6. Secure Routing
7. Sybil Detection

SecGraph current contains 11 graph anonymization techniques as detailed in the paper.

1. naive ID removal
2. Add/Del Edge Editing
3. Switch Edge Editing
4. k-DA
5. k-iso
6. bounded t-means clustering
7. union-split clustering
8. Sala et al.’s DP based algorithm
9. Proserpio et al.’s DP based algorithm
10. Xiao et al.’s DP based algorithm
11. Random Walk based algorithm