README.md

# 📰 Fake or Real News Classifier Using FastText

[![Python](https://img.shields.io/badge/Python-3.10%2B-blue?logo=python)](https://www.python.org/)  
[![FastText](https://img.shields.io/badge/Model-FastText-orange?logo=fasttext)](https://fasttext.cc/)  
[![Hugging Face](https://img.shields.io/badge/HF-Repository-purple?logo=huggingface)](https://huggingface.co/vikasgautam2003/Text_Classification_Using_FastText)

> ⚡ A **text classification project** that classifies news as **Fake or Real** using **FastText**.  

---

## 🚀 Features
✅ Train and test a FastText classifier  
✅ Support for CSV datasets  
✅ Preprocessing and simple text tokenization  
✅ Save/load trained FastText model  
✅ Ready for deployment or further experimentation  

---

## 🛠️ Tech Stack
- **Model** → FastText  
- **Language** → Python 3.10+  
- **Libraries** → pandas, fasttext  
- **Dataset** → Custom CSV dataset (`ecommerce_dataset.csv`)  

---

## 📂 Project Structure
```bash
├── .ipynb_checkpoints/           # Jupyter notebook checkpoints
├── TextClassification.ipynb       # Main notebook
├── TextClassificationPart2.ipynb  # Extended analysis notebook
├── requirements.txt               # Python dependencies
├── .gitattributes                 # Git LFS configuration
├── ecommerce.train                # Training dataset (LFS)
├── ecommerce.test                 # Test dataset (LFS)
├── ecommerce_dataset.csv          # Full dataset (LFS)
└── README.md                      # Project documentation


⚙️ Setup Instructions
1️⃣ Install Dependencies
pip install -r requirements.txt

2️⃣ Run Training / Evaluation

Open the Jupyter notebooks:

jupyter notebook TextClassification.ipynb
jupyter notebook TextClassificationPart2.ipynb


Make sure to have datasets available locally (ecommerce.train, ecommerce.test, ecommerce_dataset.csv) or tracked via Git LFS.

🎯 Use Cases

Detect fake news in real-time applications

Build NLP portfolio projects

Experiment with FastText embeddings and text classification

💡 Future Improvements

Add real-time API for classification

Integrate with Hugging Face Datasets

Hyperparameter tuning for better accuracy

Add visualization dashboards for analysis

👨‍💻 Author

Vikas Gautam