HanineAttia/Churn-Prediction-of-ooredoo-clients
Domaine:
digital infrastructureType de record:
software
Créateur:
Han
Hôte:
Telecom customer churn prediction using ML (Random Forest, XGBoost, MLP) — Streamlit app + Power BI dashboard. Internship project at Ooredoo Tunisia.
# 🎯 Customer Churn Prediction — Ooredoo Tunisia
> Predicting customer churn using Machine Learning on a 500K-row telecom dataset.
---
## 📌 Project Overview
This project builds and deploys a machine learning system to predict customer churn
for a telecom company. It includes data preprocessing, model training, performance
evaluation, and an interactive Streamlit application for real-time predictions.
---
## 🚀 Demo
### Streamlit Application
### Power BI Dashboard
---
## 📊 Results
| Model | Accuracy | ROC AUC | F1-Score (Churn) |
|-------|----------|---------|------------------|
| XGBoost | **90%** | **0.937** | **0.80** |
| Random Forest | 88% | 0.918 | 0.76 |
| MLP | 87% | 0.910 | 0.75 |
✅ **Best model : XGBoost** with 90% accuracy and 0.937 ROC AUC
---
## 🛠️ Tech Stack
- **Language :** Python
- **ML Models :** XGBoost, Random Forest, MLP (Scikit-learn)
- **Data Processing :** Pandas, NumPy
- **Visualization :** Matplotlib, Seaborn, Power BI
- **Deployment :** Streamlit
---
## ⚙️ Installation
```bash
# Clone the repository
git clone github.com
# Install dependencies
pip install -r requirements.txt
```
1. Open `Churn_Prediction.ipynb` in Google Colab or Jupyter Notebook
2. Run all cells sequentially
3. Once the Streamlit cell is executed, a **local URL** will appear (e.g. `localhost`)
4. Click the link or open it in your browser to launch the app
---
## 👩💻 Author
**Hanine Attia**
Data Science Engineering Student at ESPRIT, Tunisia