Football Player Market Value Prediction

This project focuses on predicting the market values of football players. It is based on data scraped from the SOFIFA website and involves multiple stages including data collection, cleaning, feature engineering, and model creation.

Project Overview

Data Collection: Scraped player data from the SOFIFA website using Beautiful Soup.

Data Cleaning & EDA: Used Pandas and NumPy for data manipulation and performed exploratory data analysis (EDA) with Matplotlib and Seaborn.

Feature Engineering: Applied feature scaling and selection to prepare the data for the model.

Model Creation: Built a Multiple Linear Regression model to predict football player market values.

1. Data Collection

SOFIFA Website Data:

Scraped Data:

2. Data Cleaning

Value: Converted player market values to millions.

Wage: Standardized wages to thousands.

Release Clause: Standardized release clauses to millions.

Height and Weight: Converted player height to centimeters and weight to kilograms.

Team & Contract: Split the combined column into two separate columns for better clarity.

Feature Datatypes: Corrected the datatype for each feature.

Missing Values: Addressed any missing values by applying appropriate techniques such as imputation or removing incomplete entries.

2. EDA

Distribution of Continuous Features

Relation b/w Continuous Features and Value_Mill

Distribution of Discrete Features

Relation b/w Discrete Features and Value_Mill

Strip Plot

Model Prediction: