This mini-project aims to analyze, using a Bayesian network, how different stats taken at the 10 minutes mark of a League of Legends game influence the final win.
We used python and pgmpy to create the Bayesian network and a dataset available on Kaggle to compute the CPDs of the model.
We found that, as we expected, there are statistics that affect the final outcome more, such as the difference in gold and experience, while other values have almost no importance, such as the towers difference.
All the information about the work can be found in the python notebook.