Source code for the paper Improving the Core Resilience of Real-world Hypergraphs: Concepts, Observations, and Algorithms, where we formulate and study the problem of improving the resilience of hypergraphs through the means of augmention hyperedges. In light of several empirical patterns regarding the core resilience, we propose CoReA (Core Resilience Improvement by Hyperedge Augmentation). CoReA not only proves theoretically sound but also empirically effective in enhancing the core resilience of real-world hypergraphs. Extensive experiments on ten real-world datasets demonstrates the consistent superiority of CoReA over the baselines.

Note: if a preview of the supplementary document PDF file does not appear properly, please download the file.

The datasets can be found in the Hyperedges folder. We remove any duplicate hyperedges and isolated nodes before experimenting with the methods.

Source:

• coauth-MAG-Geology: https://www.cs.cornell.edu/~arb/data/coauth-MAG-Geology/
• coauth-MAG-History: https://www.cs.cornell.edu/~arb/data/coauth-MAG-History/
• contact-high-school: https://www.cs.cornell.edu/~arb/data/contact-high-school/
• contact-primary-school: https://www.cs.cornell.edu/~arb/data/contact-primary-school/
• email-Enron: https://www.cs.cornell.edu/~arb/data/email-Enron/
• email-Eu: https://www.cs.cornell.edu/~arb/data/email-Eu/
• NDC-classes: https://www.cs.cornell.edu/~arb/data/NDC-classes/
• NDC-substances: https://www.cs.cornell.edu/~arb/data/NDC-substances/
• threads-ask-ubuntu: https://www.cs.cornell.edu/~arb/data/threads-ask-ubuntu/
• threads-math: https://www.cs.cornell.edu/~arb/data/threads-math-sx/

The pre-computed statistics of the hypergraphs are in the Hypergraph folder:

• Core Influence: where the core influences are recorded.
• Core Strength: where the core strengths are recorded.
• Coreness: where the core numbers are recorded.
• Size: where the hyperedge size distributions are recorded.

The source code is in the src folder.

starting at the src folder, run the command:

• To compile: javac main.java

• To run: java main dataset max_node_id tie degeneracy scheme rank budget batch_size

• dataset: name of the dataset.

• max_node_id: maximum node index in the dataset.

• tie: a string represents the tie-breaking scheme in Step 1-1 of CoReA: "S/I"/"1/L"/"random" for CS/CI, 1/CI, and Random, respectively.

• degeneracy: an integer represents whether the degeneracy-node requirement is enforced in Step 1-2 of CoReA: 1 for enforced, and 0 for waived.

• scheme: a string represents the selection scheme in Step 1-2 of CoReA: "I/S"/"random" for CI/CS and Random, respectively.

• rank: a string represents the scoring method in Step 2 of CoReA: "main"/"mrkc"/"random" for the Core Influence-Strength (as of our method), MRKC, and Random, respectively.

• budget: an integer represents the budget.

• batch_size: an integer represents the batch size.

For example: java main contact-high-school 327 1/I 0 random random 10 2

This code finds 10 augmented hyperedges, batch size 2, for the 'contact-high-school' dataset: tie-breaking T in Step 1-1 is 1/CI, the degeneracy in Step 1-2 is waived, the selection step in Step 1-2 is Random, the scoring method in Step 2 is Random.

The augmented hyperedges are recorded in 'Results', with the name of the file as "contact-high-school-1I-0-random-random-10-2.txt", representing the dataset and the experiment configurations.