Multiplayer Knowledge Based Subset Construction Fork

Authors: Gustaf Bergmark & Gustaf Johansson

This forked version adds compatability with our MAGIIAN game generator and statistical analysis tools.

An extensive description of the mkbsc implementation can be found in the original repo developed by August Jacobsson & Helmer Nylén

Requirements

This library uses the following external programs, which need to be installed for the library to work.

• NetworkX, which can be installed via pip3 install networkx
• pydot, which can be installed via pip3 install pydot
• Graphviz, which can be downloaded from their website

Note: the library was written for Python 3.5, NetworkX 2.1 and pydot 1.2.4. To increase the chance that everything works as intended you may wish to use those versions.

Our additions

A few but important functions were added in order to make this implementation of the MKBSC usable with our random MAGIIAN game generator.

main.py

In generatedgames.txt you will find the same set of 5 randomly generated MAGIIAN games we provided here, the idea is that you can generate a set of games with our generator and plug that file into main.py which will apply the MKBSC and determine whether they are stable or divergent. It will then update the stable value and write to analysedgames.txt. This will in turn make it possible to make use of the statistical analysis tools provided here https://github.com/gustafbergmark/MKBSC-Statistical-Analysis.

reduce.py

This function takes a set of games and reduces them, meaning a component from the game graph is removed until all components in the graph are necessary for divergence. This means that reduced games are divergent and generally quite small.

firstIteration.py

During our research project we felt that it might be interesting to analyse only the first expansion of MAGIIAN games, so this function takes a set of games and prints the first iteration of each game to another file. The firstiterations.txt file is the result when the 5 generated games provided in generatedgames.txt is used with the function.

print.py

This function looks at each game from a set of games and generates a folder with pictures of the game itself and the first 4 expansions. The generated folders are labeled in such a way that it tells you if the game is divergent or stable. OBS! it's very important that you have the "pictures" folder downloaded otherwise this will not work.

parse.py

This is a helper function which reformats the data from a generated MAGIIAN in such a way that the implemented MKBSC can be ran on it.