This repository contains the code for the paper Bavarian: Betweenness Centrality Approximation with Variance-Aware Rademacher Averages (PDF), by Cyrus Cousins, Chloe Wohlgemuth, and Matteo Riondato, appearing in the proceedings of KDD'21, and in ACM Transactions on Knowledge Discovery from Data.
An Amherst College Data* Mammoths project. This work was funded, in part, by NSF award IIS-2006765.
- For the build process: CMake, at least 3.8.0;
- To compile Bavarian and its dependencies: a C++ compiler that supports the main features of C++20 and OpenMP; recent clang and g++ satisfy these requirements;
- To run the experiments, analyze their results, and generate the figures:
- Python 3.8 or successive, with the NumPy, Matplotlib, and pandas libraries;
- LaTeX (Alternatively you can comment out the relative lines in
staticres.py).
We did not test the code on Windows, but we did test it on macOS and various *NIX flavors (GNU/Linux and FreeBSD).
- Run the
run_everything.shscript in this directory. It will first checkout the necessary submodules (NetworKit and FindTBB), copy the necessary code in the correct locations, and then compile the code, run the exact algorithm on the graphs, and finally run Bavarian on each graph with the same set of parameters used for the experiments reported in the paper, and finally it will generate the figures. - Wait :-). Running everything will take a while (potentially a few days, depending on your machine). The reasons are that (1) the exact experiments take a long time as some graphs are large; and (2) we run every Bavarian experiment multiple times for each combination of parameters, and there are many combinations of the parameters (sample_size, mc-trials, epsilons, …).
- Once the script has completed, you can find the figures in
../res/.
Copyright 2021-2022 Cyrus Cousins and Matteo Riondato
Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.