Man-of-the-Match-FIFA-2018

Recently I found this dataset on Kaggle and I am an avid football fan so I decide to create a simple project which predicts the man of the match award.

Feature description

The dataset is not big, and it contains the following features (typical stats of a football match):

1. Date

2. Team

3. Opponent

4. Goal Scored

5. Ball possesion %

6. Attempts_

7. On-Target

8. Off-Target

9. Blocked

10. Corners

11. Offside

12. Pass Accuracy %

13. Saves

14. Passes

15. Distance Covered (Km)

16. Foul Committed

17. Yellow Card

18. Yellow & Red Card

19. Red

20. 1st Goal

21. Round

22. PSO

23. Goals in PSO

24. Own Goals

25. Own Goal time

Target/Goal of the project

The Target is the Man of the Match award.

Since the given dataset is not large, this project is created to practise feature importance/data visualization tools I learned recently.

Feature importance using SHAP values

1. SHAP summary plot

alt text

2. SHAP importance plot

alt text

3. SHAP dependence contribution plot

alt text