VandinLab / PRESTO

Code for PRESTO (published at SDM2021)

PRESTO: Simple and Scalable Sampling Techniques for the Rigorous Approximation of Temporal Motif Counts

This is the reporistory for the code and the dataset presented in: PRESTO: Simple and Scalable Sampling Techniques for the Rigorous Approximation of Temporal Motif Counts published at SDM2021. If you use our code or dataset please cite our work.

For Linux

Here is an example of usage

# Build the code
make
cd examples/

# Run the experimental procedure algorithms
bash run_demo.sh

This will run the algoritms on the datasets in the folder test_graphs/ and the motifs in the folder motifs/; concatenating the whole tests in a unique file for each dataset.

Known Issues and notes

There are some compilation issues due to the library that are required from the OS version where you compile the whole package, in particular:

• If you run the make command on an Ubuntu OS with a version lower than 18.04 (i.e., 14.04, 16.04) then you need to comment the struct __exception in lines 17-23 of the file glib-core/bd.h
• Depending on the updates of your OS, some compilation flags may not be supported, in particular the debug ones, thus avoid to enable them in such situation.

EquinixChicago dataset

In this folder you can find the link to the EquinixChicago dataset used in our publication, please read carefully the README file!