FIFA Player Analytics (Data science Project)

Conducted an end-to-end data science project analyzing FIFA player data using R, covering data cleaning, transformation,

descriptive statistics, visualization, and advanced modeling (logistic regression and KNN). Built a logistic regression model to identify key factors influencing player ratings (e.g., position, body type, international

reputation), achieving 64% accuracy and 74% sensitivity; validated assumptions and optimized the model using VIF and

stepwise selection. Applied k-Nearest Neighbors to predict player positions based on 28 skill attributes, achieving 82% overall accuracy and

providing actionable insights on talent alignment for emerging players.