Customer-Churn-Prediction-Nishi

Customer Churn Prediction Project

About the Project

In this project, I have created a machine learning model to predict whether a customer will churn or not. Customer churn means when a customer stops using a service. This is an important problem for companies because it helps them understand customer behavior and improve their services.

Objective

The main objective of this project is to:

Analyze customer data
Identify important features affecting churn
Build a prediction model
Deploy the model as a simple web application

Dataset

The dataset used in this project contains information about customers such as:

Personal and account details
Services used by customers
Monthly and total charges
Churn status (Yes/No)

Technologies Used

I used the following tools and technologies:

Python
Pandas and NumPy
Matplotlib and Seaborn
Scikit-learn
Streamlit

Model Used

I have used the Random Forest Classifier for this project. It is a powerful algorithm that works well for classification problems and gives good accuracy. The model is trained using the processed dataset.

Model Evaluation

To check the performance of the model, I used:

Accuracy Score
Confusion Matrix
Classification Report
ROC Curve

Project Files

app.py → used for the Streamlit web application
model.pkl → saved trained model
columns.pkl → feature columns used in prediction
requirements.txt → list of required libraries

What I Learned

From this project, I learned:

Data preprocessing and handling
Machine learning model training
Model evaluation techniques
Basic deployment using Streamlit

Conclusion

This project helped me understand how machine learning can be applied to real-world problems like customer churn prediction. It also gave me experience in building and deploying a complete ML project.

Live Application

You can try the project here: https://customer-churn-prediction-nishi-kgcy3sx9bbrkpp9ksj6npx.streamlit.app/

Author

Name: Nishi Patel