https://github.com/pegah-ardehkhani/customer-churn-prediction-and-analysis

Analysis and Prediction of the Customer Churn Using Machine Learning Models (Highest Accuracy) and Plotly Library
https://github.com/pegah-ardehkhani/customer-churn-prediction-and-analysis

accuracy churn-prediction classification classification-algorithm cross-validation customer-churn customer-churn-analysis customer-churn-prediction data-science feature-engineering feature-importance gridsearchcv imbalanced-data machine-learning machine-learning-algorithms plotly python roc-auc sklearn telco

Last synced: 5 months ago
JSON representation

Analysis and Prediction of the Customer Churn Using Machine Learning Models (Highest Accuracy) and Plotly Library

• Host: GitHub
• URL: https://github.com/pegah-ardehkhani/customer-churn-prediction-and-analysis
• Owner: Pegah-Ardehkhani
• License: mit
• Created: 2022-06-14T10:01:32.000Z (over 3 years ago)
• Default Branch: main
• Last Pushed: 2023-01-05T17:04:08.000Z (almost 3 years ago)
• Last Synced: 2025-04-24T13:41:11.384Z (7 months ago)
• Topics: accuracy, churn-prediction, classification, classification-algorithm, cross-validation, customer-churn, customer-churn-analysis, customer-churn-prediction, data-science, feature-engineering, feature-importance, gridsearchcv, imbalanced-data, machine-learning, machine-learning-algorithms, plotly, python, roc-auc, sklearn, telco
• Language: Jupyter Notebook
• Homepage:
• Size: 1.99 MB
• Stars: 10
• Watchers: 1
• Forks: 4
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# Customer Churn Prediction and Analysis 😁😡  ![license](https://img.shields.io/github/license/Pegah-Ardehkhani/Customer-Churn-Prediction.svg) ![releases](https://img.shields.io/github/release/Pegah-Ardehkhani/Customer-Churn-Prediction.svg)  [![nbviewer](https://img.shields.io/badge/render-nbviewer-orange.svg)](https://nbviewer.org/github/Pegah-Ardehkhani/Customer-Churn-Prediction/blob/main/Telco%20Customer%20Churn%20Prediction%20and%20Analysis.ipynb)

> **`Note`**: Use [**nbviewer**](https://nbviewer.org/github/Pegah-Ardehkhani/Customer-Churn-Prediction/blob/main/Telco%20Customer%20Churn%20Prediction%20and%20Analysis.ipynb) (recommended) or google colab in order to view interactive plotly graphs. You can see all the codes and the outputs in nbviwer without running the whole code again.



  



## Dataset 📔

[Kaggle link: Telco Customer Churn](https://www.kaggle.com/datasets/blastchar/telco-customer-churn)

[Github link: Telco Customer Churn](https://github.com/IBM/telco-customer-churn-on-icp4d/tree/master/data)

## Objectives 🏆

In this project, these questions will be answered:

* [x] What's the % of Customers Churn and customers that keep in with the active services?

* [x] Is there any patterns in Customers Churn based on the gender?

* [x] Is there any patterns/preference in Customers Churn based on the type of service provided?

* [x] What's the most profitable service types?

* [x] Which features and services are most profitable?

* [x] Which features have the most impact on predicting customers churn?

* [x] Which model is the best for predicting churn?

## Project's Table of Contents ✍️

  Click to expand!

1. Problem statement

2. Import Libraries and Data

3. Handling Missing Values

4. Data Analysis and Visualization

5. Outlier Detection

6. Check for Rare Categories

7. Categorical Variables Encoding

8. Balance Data

9. Dataset Splitting

10. Feature Scaling

11. Modeling and Parameter Optimization

12. Feature Importance

13. Results

## Libraries 📚

**application** | **libraries**

--- | --- 

handle table-like data and matrices | pandas, numpy 

visualisation | plotly, seaborn, missingno 

classification models | sklearn, xgboost, mlens 

balance data | imblearn