The use of neural networks (NNs) to predict soccer matches has yielded great results, and succeeds where traditional statistical models fail. The use of diverse data and advanced machine learning algorithms can predict the results of matches with 80% accuracy. In this project, we created two different models using the European Soccer database. The first model was able to achieve ~70% accuracy and the second model reached ~68%. Both of these models take in different soccer stats. It is important that this model, or any model that attempts to predict the outcome of a match, should be used responsibily. An individual may believe that this model will help them make a more informed decision when gambling, but it should be noted that the prediction is not guarenteed. We can limit the risks of this tool by placing a disclaimer to potential users, informing them not to use this tool for the purposes of gambling.

https://github.com/ytaa2021/MLSoccerPredictions/blob/main/README.md