This project is a study of Bias in Premier League Officials using data sourced from http://www.football-data.co.uk/ in tandem with machine learning techniques.
The study looked at 10 years of data beginning with the 2010-2011 season and analyzed the allocation of fouls and cards (both yellow and red) to understand if patterns of bias were present.
In order to accomplish this task, a multi-layer perceptron model was built to conduct binary classification. The full process included: exploratory data analysis, data normalization, creating an equation to quantify bias based on the presented dataset, partitioning the data into train/test sets, building a model, and model accuracy validation.