Skip to content

Predicting customer churn using a binary classification model based on the Bank Churn dataset. This project involves handling missing data, training a model on key customer metrics, and assessing model accuracy, specificity, and sensitivity.

Notifications You must be signed in to change notification settings

stefagnone/Bank_Churn_Prediction

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

18 Commits
 
 
 
 
 
 
 
 

Repository files navigation

Bank Churn Prediction Analysis

Project Overview

In this project, I developed a predictive model to determine customer churn for a bank, utilizing Kaggle’s Bank Churn Dataset. The model identifies key features that contribute to churn and provides insights into customer retention strategies.

The analysis focused on:

  • Handling missing values in the dataset.
  • Building and evaluating a classification model.
  • Predicting churn status for a new customer dataset.

The model helps the bank understand customer behavior and make informed decisions on retaining clients.

Technologies Used

  • R: Data processing, analysis, and modeling
  • Machine Learning: Classification algorithms for predictive modeling
  • Evaluation Metrics: Model accuracy, specificity, and sensitivity

Repository Structure

  • Data/: Contains datasets used in the analysis (BankChurnDataset.csv, NewCustomerDataset.csv).
  • Code/: R script with the data preparation, modeling, and evaluation code (classification.r).
  • Images/: Contains any visuals generated from the analysis.

Key Insights

  • The model achieved a high accuracy rate, with notable specificity and sensitivity, indicating effective churn prediction.
  • Key predictors of customer churn include account duration, balance, and transaction frequency.
  • The model provides actionable insights for improving customer retention strategies.

Instructions

  1. Clone this repository.
  2. Run the R script (classification.r) in RStudio or another compatible R environment.
  3. Ensure all necessary packages are installed.
  4. Review the model output and insights from the predictions.

Contact

Connect with me on LinkedIn for more information or to discuss this project further.

About

Predicting customer churn using a binary classification model based on the Bank Churn dataset. This project involves handling missing data, training a model on key customer metrics, and assessing model accuracy, specificity, and sensitivity.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages