wzhut / cliquecounting

Code for the algorithms and experiments in "A Fast and Provable Method for Estimating Clique Counts Using Turán's Theorem" by S. Jain and C. Seshadhri

CliqueCounting

Code for the algorithms and experiments in "A Fast and Provable Method for Estimating Clique Counts Using Turán's Theorem" by S. Jain and C. Seshadhri

Steps to run the code:

1. Store the graph to be analyzed as a list of edges in a .txt file in the graphs folder. cd into the python folder and run:

python sanitize.py ../graphs amazon0601.txt

This will create a file 'amazon0601.edges' in the graphs folder.

2. To run the TuránShadow algorithm, run:

./test_cliques_turan_shadow -i amazon0601.edges -m 5 -M 5 -r 10 -n 50000

This will save the tree stats in '/amazon0601_5_50000_params' csv file in results/cliques/ and the estimates of the 10 runs in 'amazon0601_5_50000_10_data'

3. To run the Edge Sampling algorithm, run:

./test_cliques_edge_sampling -i amazon0601.edges -m 5 -M 5 -r 10 -p 0.75

This will save the estimates from 10 runs of the algorithm in amazon0601_5_0.75_10_es in results/cliques/

4. To run the GRAFT algorithm, run:

./test_cliques_graft -i amazon0601.edges -m 5 -M 5 -r 10 -n 100000

This will save the estimates from 10 runs of the algorithm in amazon0601_5_100000_10_graft in results/cliques/

5. To run the Color Coding algorithm, run:

./test_cliques_color_coding -i amazon0601.edges -m 5 -M 5 -r 10

This will save the estimates from 10 runs of the algorithm in amazon0601_5_10_cc in results/cliques/

6. To run the brute force clique counting algorithm, run:

./test_cliques_brute_force -i amazon0601.edges -m 5 -M 5 -r 10

This will save the estimates from 10 runs of the algorithm in amazon0601_5_10_bf in results/cliques/

For questions, contact Shweta Jain at [email protected].