Combining Machine Learning and Human Experts to Predict Match Outcomes in Football: A Baseline Model

In this paper, we present a new application-focused benchmark dataset and results from set of baseline Natural Language Processing Machine Learning models for prediction match outcomes games football (soccer). By doing so give the accuracy that can be achieved exploiting both statistical data contextual articles human sports journalists. Our is focuses on representative time-period over 6 seasons English Premier League, includes newspaper previews The Guardian. presented in paper achieve an 63.18% showing 6.9% boost traditional methods.