FIFA 19 Player's Rating Repository!

About

Dataset: FIFA 19 Dataset from Kaggle

The order of our notebook is:

1. Introduction
2. Data Cleaning
3. Exploratory Data Analysis
4. Machine Learning
5. Conclusion & Takeaways

Problem Definition

• For a GoalKeeper, which Goalkeeping attribute contributes most to his overall rating? Predict the overall rating of goalkeepers with that attribute
• For each Non-Goalkeeping Position, which non-physical attribute contributes most to each of their overall rating? Predict the overall ratings of each non-goalkeeping position players with just their respective highest correlated attribute.

Models Used

1. Linear Regression
2. Decision Tree Regressor
3. Random Forest

Conclusion

• GKReflexes is the buggest predictor of GoalKeeper overall rating
• Positioning is the biggest predictor for Attacker overall rating
• Composure and vision are the biggest predictor for midfielder overall rating
• Interceptions is the biggest predictor for defender overall rating
• Random Forest model is best at predicting overall rating

Takeaways

• Random Forest from sklearn
• Linear Regression from sklearn
• Decision Tree Regressor from sklearn
• Data Visualisation from plotly
• Github collaboration
• Learn to compare and utilise different models to suit our needs for prediction

References

https://plotly.com/python/
http://scikit-learn.org/stable/modules/generated/sklearn.ensemble.RandomForestClassifier.html
https://en.wikipedia.org/wiki/Random_forest
https://en.wikipedia.org/wiki/Linear_regression
https://gdcoder.com/decision-tree-regressor-explained-in-depth/

Contributors

• @qtxbryan Bryan Tan Shao Xuan
• @rphoen Phoen Yu Heng Ryan
• @haqimthehipster Norhaqim Azalee