About

This code produces the decision-theoretic bootstrapping experiments from the following paper:

Shih-Wen Huang*, Jonathan Bragg*, Isaac Cowhey, Oren Etzioni, Daniel S. Weld. Toward Automatic Bootstrapping of Online Communities Using Decision-theoretic Optimization. In Proceedings of the 19th ACM Conference on Computer-Supported Cooperative Work & Social Computing (CSCW '16). 2016.

Installation

To create a Conda environment with name {ENV_NAME}, run conda env create -n {ENV_NAME} -f environment.yml

Running

To test a very basic example, run python exp.py -c config/config_std_tiny.json -p sample_policies.json.

This should create config_std_tiny.json_sample_policies.json.csv in the main directory.

Footnotes

• denotes equal contribution