yallashoot/football-commentary-dataset

⚽ YallaShoot Football Commentary Dataset

A curated dataset of football (soccer) match commentary, player mentions, and match events — built to power NLP models for the Arabic and global football community.

📌 Dataset Description

This dataset contains structured football match commentary collected from live match feeds, covering top leagues including:

• 🏴󠁧󠁢󠁥󠁮󠁧󠁿 English Premier League (EPL)
• 🇪🇸 La Liga
• 🏆 UEFA Champions League
• 🌍 Arab World Leagues (Saudi Pro League, Egyptian Premier League)

It is designed for tasks such as:

• Sentiment Analysis — Is the commentary positive (goal!) or negative (miss/foul)?
• Named Entity Recognition (NER) — Detecting player names, club names, stadiums
• Event Classification — Goal, Yellow Card, Red Card, Substitution, etc.
• Text Generation — Auto-generating match commentary

---

📂 Dataset Structure

yallashoot-football-commentary/
├── train/
│   └── commentary_train.csv
├── test/
│   └── commentary_test.csv
└── validation/
    └── commentary_val.csv

Data Fields

Column	Type	Description
id	int	Unique row ID
match_id	string	Match identifier
minute	int	Match minute (0–90+)
commentary	string	Raw commentary text
event_type	string	goal / foul / yellow_card / red_card / substitution / other
sentiment	string	positive / negative / neutral
player_mentioned	string	Player name if mentioned, else null
team	string	Home or Away team name
league	string	League name
language	string	en / ar

Example Row

{
  "id": 1042,
  "match_id": "EPL_2024_MCI_ARS",
  "minute": 73,
  "commentary": "GOAL! Erling Haaland taps it in from close range! Manchester City lead 2-1!",
  "event_type": "goal",
  "sentiment": "positive",
  "player_mentioned": "Erling Haaland",
  "team": "Manchester City",
  "league": "Premier League",
  "language": "en"
}

---

🚀 Usage

from datasets import load_dataset

ds = load_dataset("yallashoot/football-commentary-dataset")

# Access training split
train_data = ds["train"]

# Filter for goals only
goals = train_data.filter(lambda x: x["event_type"] == "goal")

# Filter by league
epl_data = train_data.filter(lambda x: x["league"] == "Premier League")

---

🎯 Intended Use Cases

Use Case	Task Type
Football sentiment analysis	Text Classification
Player name extraction	NER / Token Classification
Live commentary generation	Text Generation
Match event detection	Multi-label Classification
Arabic football NLP	Multilingual NLP

---

📊 Dataset Statistics

Split	Rows	Languages	Leagues Covered
Train	40,000	EN, AR	EPL, La Liga, UCL, SPL
Test	5,000	EN, AR	EPL, La Liga, UCL
Validation	5,000	EN, AR	EPL, La Liga, UCL

---

🏷️ Labels

Event Types

goal · foul · yellow_card · red_card · substitution · corner · penalty · offside · other

Sentiment

positive · negative · neutral

---

📜 License

This dataset is released under Creative Commons Attribution 4.0 (CC BY 4.0).
You are free to use, share, and adapt — with attribution to YallaShoot.

---

🔗 About YallaShoot

YallaShoot is a leading football live scores and match streaming platform serving millions of Arabic-speaking football fans worldwide.

• 🌐 Website: yallashoot.com
• 📱 Platform: Live scores, match highlights, standings
• 🤖 HuggingFace: huggingface.co/yallashoot

This dataset is part of YallaShoot's effort to bring AI and NLP to the football world.

---

📝 Citation

If you use this dataset, please cite:

@dataset{yallashoot2024football,
  author    = {YallaShoot AI Team},
  title     = {YallaShoot Football Commentary Dataset},
  year      = {2024},
  publisher = {HuggingFace},
  url       = {https://huggingface.co/datasets/yallashoot/football-commentary-dataset}
}

---

🤝 Contributing

We welcome contributions! If you have football commentary data you'd like to add:

1. Open an issue describing your data
2. Follow the dataset schema above
3. Submit a pull request

---

Built with ❤️ for the global football community by YallaShoot