BankChurners

A data science project that analyzes bank customer churn using logistic regression and decision trees. Uses the Kaggle “Credit Card Customers” dataset to predict which customers are likely to attrite.

Features

• Data cleaning and exploratory data analysis
• Visualization of customer demographics and behavior patterns
• Logistic regression classification model
• Decision tree classifier
• Model evaluation with confusion matrices

Dataset

The dataset contains 10,127 customers with 21 features including:

• Customer age, gender, education, marital status
• Income category, card category
• Credit limit, transaction history
• Inactivity period, contact count
• Attrition flag (target variable: churned or existing)

Technologies

• Python 3
• Pandas, NumPy
• Matplotlib, Seaborn
• Scikit-learn (LogisticRegression, DecisionTreeClassifier)
• Jupyter Notebook

How to Run

Open BankChurners.ipynb in Jupyter Notebook or Google Colab.

Project Structure

BankChurners/
├── BankChurners.ipynb  - Main analysis notebook
└── BankChurners.csv    - Dataset

Authors

Ritik Sirsikar, Jasdeep Singh Dhillon