Customer churn prediction

Developed a comprehensive Machine Learning project to predict customer churn in the telecommunications industry and identify customers who are likely to stop using the company's services.

The project involved analyzing 7,000+ customer records with over 20 features, exploring customer behavior, preprocessing data, and building multiple classification models to predict customer churn.

Key Features

✅ Exploratory Data Analysis (EDA): Analyzed customer behavior, monthly charges, contract types, and churn patterns using statistical analysis and data visualization.

✅ Data Preprocessing: Handled missing values, encoded categorical variables, and applied feature scaling to prepare data for machine learning models.

✅ Machine Learning Models: Implemented and compared 9+ classification algorithms, including Random Forest, Logistic Regression, Gradient Boosting, KNN, and Neural Networks.

✅ Feature Engineering: Applied dimensionality reduction techniques, including PCA, LDA, and Kernel PCA.

✅ Model Evaluation: Evaluated model performance using Accuracy, Precision, Recall, F1-Score, ROC-AUC, Confusion Matrix, and Cross-Validation.

✅ Data Visualization: Created visualizations to understand customer churn patterns and compare model performance.

Technologies Used

Python

Pandas & NumPy

Scikit-learn

Matplotlib & Seaborn

Machine Learning

Data Analysis & Visualization

Project Outcome

Built and evaluated multiple machine learning models to identify customers at risk of churn.

The project demonstrates a complete end-to-end machine learning workflow, from data preprocessing and exploratory analysis to model training, evaluation, and comparison.

Business Value: Helps telecommunications companies understand customer behavior, identify customers at risk of leaving, and support data-driven customer retention strategies.