This project aims to analyze revolut's customer churn dataset and develop models to predict customer attrition. The analysis starts with data exploration, visualization, and preprocessing steps, followed by model selection and evaluation.
The dataset contains information about bank customers, including their demographics, account details, and churn status. The analysis begins with an overview of the data distribution through various visualizations:
Customer age distribution Gender distribution Dependent count distribution Education level distribution Marital status distribution Income category distribution Card category distribution Months on book distribution Total relationship count distribution Months inactive in the last 12 months distribution Credit limit distribution Total transaction amount distribution Churn vs. non-churn customer proportion Data Preprocessing
The following preprocessing steps are performed:
One-hot encoding of categorical features Upsampling the data using SMOTE to handle the imbalanced class distribution Dimensionality reduction of one-hot encoded features using Principal Component Analysis (PCA) Model Selection and Evaluation Two classification models, Random Forest and Support Vector Machine (SVM), are evaluated using 5-fold cross-validation. Their performance is assessed using the F1 score.
License This project is licensed under the MIT License. See the LICENSE file for details.
Authors: Bohdan Boiprav & Catalin Bondari