Customer-Churn-Prediction

# -Customer-Churn-Prediction

Customer Churn Prediction

This project aims to predict customer churn in the telecommunications industry using machine learning techniques. Churn refers to customers who stop using the company's services. By analyzing historical customer data, we built a logistic regression model to identify the factors influencing churn and predict which customers are at risk of leaving.

Project Highlights

EDA (Exploratory Data Analysis):

Visualized churn distribution across features like contract type, tech support, internet service, and tenure.

Identified key churn indicators (short tenure, no tech support, high monthly charges).

Handled outliers and missing values with appropriate methods.

Data Preprocessing:

Encoded categorical features with LabelEncoder and OneHotEncoder.

Handled missing data using mode and median.

Detected and removed outliers using the IQR method.

Modeling:

Built a Logistic Regression model using Scikit-learn.

Achieved balanced performance with accuracy, precision, recall, and F1-score metrics.

️ Tools and Technologies

Python

Pandas, NumPy

Matplotlib, Seaborn, Plotly Express

Scikit-learn

Jupyter Notebook

Files Included

churn_analysis.ipynb: Jupyter notebook with full code

churn_report.pdf: Project report with visualizations and explanations

README.md: Project description and setup guide

How to Run

Clone the repository

Install required libraries from requirements.txt

Open and run the notebook churn_analysis.ipynb

Future Enhancements

Try other ML models like Random Forest, XGBoost

Apply feature selection or PCA

Integrate customer usage logs if available