https://github.com/susmnty/client-iq
The project focuses on predicting customer status using machine learning. It classifies customers as active, inactive, or at risk. Python and ML libraries are used for data analysis and modeling. The goal is to help businesses reduce churn. It supports better decision-making through predictive insights.
https://github.com/susmnty/client-iq
business-analytics churn-prediction customer-segmentation data-preprocessing feature-engineering machine-learning-models model-evaluation
Last synced: 12 months ago
JSON representation
The project focuses on predicting customer status using machine learning. It classifies customers as active, inactive, or at risk. Python and ML libraries are used for data analysis and modeling. The goal is to help businesses reduce churn. It supports better decision-making through predictive insights.
- Host: GitHub
- URL: https://github.com/susmnty/client-iq
- Owner: susmnty
- Created: 2025-05-10T17:01:40.000Z (about 1 year ago)
- Default Branch: main
- Last Pushed: 2025-05-10T17:05:44.000Z (about 1 year ago)
- Last Synced: 2025-05-21T05:13:11.663Z (about 1 year ago)
- Topics: business-analytics, churn-prediction, customer-segmentation, data-preprocessing, feature-engineering, machine-learning-models, model-evaluation
- Language: Jupyter Notebook
- Homepage:
- Size: 66.4 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# Customer Account Status Prediction: ML Model Comparison & Deployment
This project leverages a comprehensive machine learning pipeline to predict the current status of bank customer accounts. Using a dataset enriched with demographic, behavioral, and transactional features, multiple classification algorithms are compared to identify the most accurate and robust model. The project includes full preprocessing, outlier handling, feature encoding, model training, evaluation, and deployment with Streamlit.
## 📊 Features
- Cleans and preprocesses real-world financial customer data
- Compares 12+ models including Random Forest, XGBoost, Neural Networks, and Ensemble methods
- Visual comparison of model performance
- Saves the best-performing model using `joblib`
- Deploys with a user-friendly Streamlit web interface for live predictions
## 🚀 How to Run
### 1. Clone this repo
```bash
git clone https://github.com/yourusername/ML-Customer-Status-Prediction.git
cd ML-Customer-Status-Prediction
```
### 2. Install dependencies
```bash
pip install -r requirements.txt
```
### 3. Run Streamlit app
```bash
streamlit run app.py
```
## 📁 Files
- `Data_TarImp_CleanOutlier.csv`: Cleaned dataset
- `model_training.py`: Script for data processing, model training & comparison
- `best_model.pkl`: Saved trained model
- `app.py`: Streamlit app
- `requirements.txt`: All dependencies
---
📄 requirements.txt
pandas
numpy
matplotlib
seaborn
scikit-learn
xgboost
joblib
streamlit
---
# Road map
