FIFA 2018 World Cup Match Outcome Prediction Using Machine Learning

Project description.

Built a machine learning-based predictive system to forecast match outcomes for the FIFA World Cup by analyzing and modeling historical data spanning from 1930 to 2018. The project involved comprehensive feature engineering, model selection, and ensemble learning techniques to increase the accuracy of match result predictions. By integrating various data sources and testing multiple algorithms, the final model achieved improved reliability and performance, making it suitable for sports analytics and fan engagement platforms.

Goal

The goal is to use Machine Learning to predict who is going to win the FIFA World Cup 2018.

Predict the outcome of individual matches for the entire competition.

Run simulation of the next matches i.e quarter finals, semi finals and finals.

These goals present a unique real-world Machine Learning prediction problem and involve solving various Machine Learning tasks: data integration, feature modelling and outcome prediction.

Data

I used two data sets from Kaggle - Results of the matches since 1930 and the World Cup 2018 Dataset. I used results of historical matches since the beginning of the championship (1930) for all participating teams.

Environment and tools

Jupyter Notebook

Numpy

Pandas

Seaborn

Matplotlib

Scikit-learn

I chose Logistic Regression in my model and got an accuracy of 57% on the training set and 55% accuracy on the test set. I also used the FIFA ranking as of April 2018 dataset and a dataset containing the fixture of the group stages of the tournament.

According to this model Brazil is likely to win this World Cup.

Areas of further Research/ Improvement

Dataset - to improve dataset you could use FIFA, the game not the organisation, to assess the quality of each team player.

A confusion matrix would be great to analyse which games the model got wrong.

We could ensemble that is, we could try stacking more models together to improve the accuracy.