Objective:
- Prediction of the winner of an international matches Prediction results are "Win / Lose / Draw"
- Apply the model to predict the result of FIFA world cup 2022.
Data:
-Nous avons les rangs FIFA de 1993 à 2018 donnés par
https://www.kaggle.com/datasets/tadhgfitzgerald/fifa-international-soccer-mens-ranking-1993now
- Nous avons les rangs FIFA de 1992-2022
https://www.kaggle.com/datasets/cashncarry/fifaworldranking
-L’historique des matches de football depuis 1872 donné par
https://www.kaggle.com/datasets/martj42/international-football-results-from-1872-to-2017
-Les statistiques de chaque équipe depuis 2018 tirées de Wikipédia
https://en.wikipedia.org/wiki/National_team_appearances_in_the_FIFA_World_Cup#Overall_team_records
-Les statistiques des joueurs tirées de
https://www.kaggle.com/antoinekrajnc/soccer-players-statistics
-Fifa index
https://www.fifaindex.com/fr/team/1335/france/fifa23/
-Football nations Stats
https://fbref.com/en/countries/
-Football nations Stats https://fbref.com/en/countries/
-Players data to scrap https://fbref.com/en/players/e42d61c7/Achraf-Hakimi
Environment and tools
1. Jupyter Notebook
2. Numpy
3. Pandas
4. Seaborn
5. Matplotlib
6. Scikit-learn
7. xgboost
8. scipy
9. joblib
we chose XGBoost in model and got an accuracy of 78% on the training set and 63% accuracy on the test set
Lifecycle
site web
https://world-cup-2022-predictions.herokuapp.com/