sagehaz/Football_Match_Dataset

🏟️ Football Match Prediction Dataset

This dataset contains historical football match data designed to train and evaluate machine learning models for predicting various match outcomes, including:

• Odd/Even goals prediction
• Yes/No for both teams scoring
• Over/Under total goals scored

The dataset supports multiple prediction tasks such as classification, regression, and probabilistic forecasting of football outcomes.

---

📁 Dataset Overview

Column Name	Description	Example
date	Date when the match was played	2024-02-20
league	League or competition name	Premier League
home_team	Name of the home team	Manchester United
away_team	Name of the away team	Liverpool
home_score	Goals scored by the home team	2
away_score	Goals scored by the away team	1
full_time_result	Match result (H=Home Win, D=Draw, A=Away Win)	H
over_2_5	1 if total goals > 2.5, else 0	1
both_teams_score	1 if both teams scored, else 0	1
odd_even_total	1 if total goals are odd, 0 if even	1
home_odds	Decimal odds for home team win	1.80
draw_odds	Decimal odds for a draw	3.40
away_odds	Decimal odds for away team win	4.20

---

🔍 Usage

This dataset can be used for:

• Training machine learning models for football match outcome prediction.
• Statistical analysis of team performance based on historical trends.
• Building betting strategy simulations using predictive analytics.

---

🏗️ How to Load the Dataset

You can load this dataset directly using the Hugging Face datasets library:

from datasets import load_dataset

dataset = load_dataset("username/your-dataset-repo-name")
print(dataset)

✨ Dataset Splits
The dataset is divided as follows:

Training Set: 80%
Validation Set: 10%
Test Set: 10%
⚡ Preprocessing Suggestions
Handle missing values (NaN) through imputation or removal.
Convert categorical columns (league, home_team, away_team) using one-hot encoding or embeddings.
Normalize numerical features (home_odds, away_odds) for model training.
Engineer features such as form metrics, head-to-head records, and average goals.
📜 Source
The dataset is compiled from publicly available football match data sources and is intended for research and educational purposes only.

⚠️ License
CC BY 4.0 — You are free to use this dataset with attribution.

💬 Citation
bibtex
Copy code
@dataset{username_football_prediction_2025,
  author       = {Your Name or Username},
  title        = {Football Match Prediction Dataset},
  year         = 2025,
  url          = {https://huggingface.co/datasets/username/your-dataset-repo-name},
}
🙌 Acknowledgements
Special thanks to the football analytics community for insights on sports data modeling and predictive analysis.

yaml
Copy code

---

### ✨ **What This README Covers:**
- Clear **description** of the dataset and its predictive applications.
- Detailed **column explanations** for easy understanding.
- **Usage instructions** for loading the dataset using Hugging Face's `datasets` library.
- **Preprocessing suggestions** tailored for your prediction tasks (odd/even, over/under, etc.).
- **Proper licensing**, **citation**, and **acknowledgement** sections.

---

### 🎯 **Next Steps for You:**
1. Update:
   - Replace `username/your-dataset-repo-name` with your actual Hugging Face username and repo name.
   - Adjust the **column descriptions** if your dataset contains additional features.
2. Upload this `README.md` directly to your Hugging Face dataset repo under the **"Files and versions"** tab.

---

Would you like to add specific data sources or expand on the feature engineering part for advanced users? 🚀