Skip to content

Latest commit

 

History

History
28 lines (21 loc) · 1.48 KB

README.md

File metadata and controls

28 lines (21 loc) · 1.48 KB
Raw

USF MSDS 699: Machine Learning Laboratory

Final Project: Soccer Player Transfer Value Prediction

Team Members

Data Source

Research Questions

  • Why are some soccer players worth more than others?
  • Can we predict the value of a soccer player in the transfer market from their ratings and characteristics?

Abstract

We fitted a Random Forest model to predict the value of a FIFA 19 player based on his ratings and characteristics in the game. Using the model, we concluded that reaction time and age are the most important attributes of a player. Meanwhile, attacking player attributes are more predictive than defensive player attributes. Other important attributes include ball control and goalkeeper reflexes.

Order of Notebook Execution

  1. Data Cleaning
  2. Model Selection and Training

Data Cleaning

The data cleaning and feature engineering process are recored in the Clean Notebook.

Model Selection and Training

The modeling selection and training process are recorded in the Final Consolidation Notebook.