Leveraging Uncertainty in Collective Opinion Dynamics with Heterogeneity

Code for the experiments and plots in our paper.

Repository Structure

The code base consists of four main folders

• models: containing code to build all Bayes and Naive variations in the experiments
• experiments: containing code to run experiments on these models
• plotting_and_analysis: contains code to analyse and plot executed experiments
• hpc_scripts: contains scripts to run experiments and do some prelimenary analysis on HPC clusters

Setup

To setup generate a conda or venv with the env.yml file.

Run an Experiment

To simply run an experiment try out src/experiments/local_test.py There you can also easily play around with some parameters of the models or scenario.

Run lots of Experiments (on HPC)

To run los of experiments try out src/experiments/experiment_runner.py or to run this on an HPC cluster src/hpc_scripts/hpc_slurm.bash

Data & Plotting

Plotting tools can be found in plotting_and_analysis. To analyse our original data, please find the data via this link.