Skip to content

Latest commit

 

History

History
58 lines (48 loc) · 2.61 KB

README.md

File metadata and controls

58 lines (48 loc) · 2.61 KB
Raw

Football Match Winner Prediction

About

This project is about learning and implementing machine learning models to predict the outcome of a football match and identify the winning team. We have extracted and built our own features that calculate and provides the stats per match. Features are the main essence of our project that highly impacts are end results.

Technology Stack

  1. Java
  2. Python
  3. Matlab
  4. Linux
  5. Machine Learning
  6. Netbeans
  7. Pycharm
  8. Matlab

Data Selection and Extraction

We selected Barclays Premier League website as our main source of data. To extract the data we wrote a script that goes through each year's data and for each year it extracts fixture table after the results of each match day. You can find the data here.

Feature Extraction

  1. Current Form
    • Home / Away Wins
    • Relative Team Position
  2. Attack Quotient
    • Shots on Target
    • Goals Scored
  3. Goals Conceded
  4. Clean Sheets

Machine Learning Models

  1. Logistic Regression
  2. Vote Algorithm
    • Naive Bayes Classifier and,
    • Random Forest

Graphical User Interface

  1. We created a simple graphical user interface that takes in the already built model for a particular year.
  2. Based on the given model it provides prediction results for all the fixtures suggested for that particular year.
  3. It also provides a way to look at stats per match defending the idea behind the predicted results.
Home Page Team Selection to View Stats
alt text alt text
Predicitons Per Match Stats Per Match
alt text alt text

Technical Research Paper

Our paper is published at International Journal of Computer Applications. You can access it using this link