Hook
More breakout videos from this creator.
Let's predict the winners of the World Cup finals using machine learning. The model was able to predict the exact two-one score for Argentina against England, and it was also able to predict all the winners for the quarter-final games. But unfortunately, for the France vs Spain game, we got the Mbappe special, got cooked. So let's predict the World Cup final game between Spain and Argentina. And to actually train my model, I'm training it all for free off these Statbomb data sources and I'm filling in any gaps with the Football Data API. The type of model I use to actually make the predictions is called an XGBoost model, which is basically a bunch of decision trees, which work together to help predict a win or a loss. For the XGBoost model, I trained it on 20 distinct features from different tournaments to try to get as much data as possible. If you want to get access to my World Cup machine learning model, download Axon on the App Store, and if you're on Android or desktop, go to axonlearn.app. And since the World Cup is ending, I'm planning on doing more prediction models for other sports. So let me know what you want to see next!!! After compiling all the data, the model is really just giving a 51% confidence for Spain winning the game with an expected score of two to one.