FIFA 2018 Winner Predication by Mithilesh thakkar

This is a python implementation of the fifa(football worldcup) 2018 each match's winner and final winner.

The purpose of this is to try and predict the top 3 teams for World Cup 2018 using classification models coupled with poisson distribution to predict the exact results of the semi-finals, third place playoff and final.

Final Predictions based on this notebook:

Winner: Germany

2nd Place: Spain

3rd Place: France

And they are same as the results of the FIFA 2018.

Contents:

1. Import Necessary Packages/Datasets

• Numpy (for implementation of basic linear algebra operation)
• Pandas (for data analysis and manipulation)
• Matplotlib (for ploting the graph)
• Seaborn (for statistical data visualization)
• Sklearn (for implemanting different classifiers)

2. Data Cleaning

• This is done for making the dataset useable for our prediction

3. Classification Models to predict match results (Win/Draw/Lose)

• classifier used:
  • First I have used the basic logistic regression method to predict the winner.

Results

• Currently I am getting the winner with the accuracy of 57% using the logistic regression..

Future Work

• And I will try to use the XGBoosting,Random Forest, Decision Tree, K-Nearest Neighbour and SVM (Linear Kernel). classifier to predict the winner which can increase the accuracy.

Acknoweldgement

• I acknowelde ahmedelnaggar for providing the dataset for the preication.