/player-position-prediction

This project aims to predict the playing positions of football players using a K-Clustering model implemented on Azure. The prediction is based on the FIFA 23 player statistics dataset, which contains detailed attributes of players.

Primary LanguageJupyter Notebook

Football Player Position Prediction using K-Clustering Model on Azure

This project aims to predict the playing positions of football players using a K-Clustering model implemented on Azure. The prediction is based on the FIFA 23 player statistics dataset, which contains detailed attributes of players.

Dataset

The FIFA 23 player stats dataset provides comprehensive attributes for football players, including ratings, skills, physical attributes, and performance metrics.

Methodology

The project follows these steps:

  1. Data Preparation: Clean and transform the dataset for clustering analysis.
  2. Feature Selection: Choose relevant attributes to represent player characteristics.
  3. Model Training: Use K-Clustering to group players into positions.
  4. Model Evaluation: Assess cluster quality using evaluation metrics.
  5. Prediction: Assign new players to positions based on their attributes.

Azure Implementation

Azure is used for scalability, reliability, and machine learning tools and services. Azure services utilized may include Azure Machine Learning, Azure Databricks, Azure Blob Storage, and Azure Kubernetes Service (AKS)