This project simulates the growth of a slack community using reinforcement learning. It is comprised of a simulator and acoompanying graph visualization

*** The libraries for the Web App can be viewed in requirements.txt ***

The code for my project is in simulator.py. You can run the code there using: python simulator.py Running this shows the network/community visualization as it changes over time.

If you would like to run the experiments that lead to the data you can uncomment

Line 595----> # experiment()

in simulator.py. You can also specify the number of repetitions, the growth rate, and the match rate you would like for the model.

The Project PDF can be viewed in /Figures

A completed graphStats.csv and graphStatsFinal.xls have been uploaded. Running the model rewrites graphStats.csv. These files contain the results from running an "experiment()".

You can view the web visualization for the current network at percey.herokuapp.com/demo. (click on a user and view their relationships)

--pip install from project root

You can run the Percey web app by entering python app.py in the command line. Running the App:

Percey is a Slack bot + Web App

Percey works to understand your slack group's communication, create a graph model of it, and suggest connections to improve the quality of your network.

Hope 2016 :)