Identifying Possible Winners in Ranked Choice Voting Elections with Outstanding Ballots

Paper

Jelvani, Alborz, and Amélie Marian. "Identifying Possible Winners in Ranked Choice Voting Elections with Outstanding Ballots." Proceedings of the AAAI Conference on Human Computation and Crowdsourcing. Vol. 10. No. 1. 2022.

Data

Testing is performed using the dataset below:

New York City Democratic Mayor Citywide

The data folder which should be placed in the root directory can be found here

Installing Dependecies

The Python dependecies can be installed through the requirements.txt file provided. Run:

pip install -r requirements.txt

Additionally, Graphviz should nanually be installed and added to the system PATH.

What to run

src/autorun.py

This script will run each election in the src/elections.csv file. The output contains the data we present in Table 1 along with the minimum bound ballots for each possible winnner.

src/gen_graphs.py - Generate elimination order DAGs

Be default, we have this set to produce the compressed DAG for the New York Democratic Member of the City Council 16th Council District election.

Set the election file path variable file as desired. Set the number of candidates in the election num_cands and number of missing ballots missing as desired. Choose if you want to use DAG compression in compress_graph. If you want to remove certain candidates from the election, add their candidate id's to the remove list. Running the script will produce the graph in output.eps and also opens a window to see the graph. Note: You almost always have to play around with the node sizing and font parameters to achieve a good fit in the window.

plots/plot_data.py

This script will generate the plot in Figure 2.

Other files

src/make_tree.py

This contains the core of our algorithms. Algorithm 1 from the paper is implemented in _verify(). Algorithm 2 is implemented in _make_tree(). Creating an elimination DAG with compression is implemented in tree2graph().