https://github.com/mohamed-samy26/service-cancellation-predictor

Machine learning based system that can predict service cancellation for a business based on its dataset using 5 models
https://github.com/mohamed-samy26/service-cancellation-predictor

cart-algorithm data-cleaning decision-trees id3 jupyter-notebook knn logistic-regression machine-learning python svm tkinter

Last synced: about 1 month ago
JSON representation

Machine learning based system that can predict service cancellation for a business based on its dataset using 5 models

• Host: GitHub
• URL: https://github.com/mohamed-samy26/service-cancellation-predictor
• Owner: Mohamed-Samy26
• Created: 2022-04-28T22:46:07.000Z (over 3 years ago)
• Default Branch: main
• Last Pushed: 2022-06-25T04:35:56.000Z (over 3 years ago)
• Last Synced: 2025-03-01T20:32:06.404Z (7 months ago)
• Topics: cart-algorithm, data-cleaning, decision-trees, id3, jupyter-notebook, knn, logistic-regression, machine-learning, python, svm, tkinter
• Language: Jupyter Notebook
• Homepage:
• Size: 3.5 MB
• Stars: 2
• Watchers: 1
• Forks: 1
• Open Issues: 0
• Metadata Files:
  • Readme: README.md

Awesome Lists containing this project

README

          
# Service Cancellation Predictor

Machine learning based system that can predict service cancellation for a business based on its dataset using 5 different machine learning models







## About Project

 Service cancellation is when a customer decides to end thier businesses or unsubscribe from a service with a specific company. 

This system is oriented to determine the possibility of customers cancelling services. 

For most businesses, the ability to predict that a particular customer is at a high likelihood of cancelling service could actually result in a better way

for handling such problems.

Foreseeing business-related actions is the core of this project, and therefore, this system was developed to meet the business related requirements 

to predict user churn with an average accuracy of 78% with an attractive user-friendly UI.

Dataset Link: [Service Cancellation Dataset](https://drive.google.com/file/d/1PrE7kY9h8yTawg0Ul0Ij5RFjiSO0hdYk/view)

  

## Problem and Algorithms

  

Problem stipulated that based on 20 attributes and 7043 record we should make several models that predict whether a user will cancel his service or not 

we applied 4 models :

* `Decision Tree (ID3 / CART)` 

* `Logistic Regression `

* `SVM `

* `KNN`

  

Models consumed cleaned data and produced the following accuracies:   

| Model | Accuracy |

| --- | --- |

| Decision Tree(ID3) | 77.69% |

| Decision Tree(CART) | 74.2784% |

| Logistic Regression | 80.7109% |

| SVM | 79.18% |  

| KNN | 74.48% |  


## GUI

  

### Splash Screen  

![Splash Screen](https://github.com/NourAyman10/ServiceCancellationPredictor/blob/master/splashScreen.gif)  

### Main Screen

#### Empty with no data

![empty inputs](https://github.com/NourAyman10/ServiceCancellationPredictor/blob/master/noData.png) 

![inputs validation](https://github.com/NourAyman10/ServiceCancellationPredictor/blob/master/dataValidation.png)

![methodology validation](https://github.com/NourAyman10/ServiceCancellationPredictor/blob/master/noMethodology.png)

#### All data filled

![filled inputs](https://github.com/NourAyman10/ServiceCancellationPredictor/blob/master/inputsWithAllData.png)

![how to fill inputs](https://github.com/NourAyman10/ServiceCancellationPredictor/blob/master/toFillData.png)  

#### When customer cancels the service

![customer cancel service](https://github.com/NourAyman10/ServiceCancellationPredictor/blob/master/coustomerWillCancelService.png)  

#### When customer keeps the service

![customer accept service](https://github.com/NourAyman10/ServiceCancellationPredictor/blob/master/customerWillKeepTheServices.png)

#### Training model on input data

![train model](https://github.com/NourAyman10/ServiceCancellationPredictor/blob/master/trainPhoto.png)  

#### Testing models accuracy

![test model](https://github.com/NourAyman10/ServiceCancellationPredictor/blob/master/test_accuracy.png)

![test logistic regression model](https://github.com/NourAyman10/ServiceCancellationPredictor/blob/master/test_accuracy_logisticReg.png)

  * Visualising results

  

![show graphs](https://github.com/NourAyman10/ServiceCancellationPredictor/blob/master/showGraphs.png)  

 

## Contributors

- [Mohamed-Samy26](https://github.com/Mohamed-Samy26) 

- [RedBiscuits](https://github.com/RedBiscuits)

- [KareemHussen](https://github.com/KareemHussen)

- [NourAyman10](https://github.com/NourAyman10)

- [HaneenIbrahim2](https://github.com/HaneenIbrahim2)

- [OmarSherif2](https://github.com/OmarSherif2)

- [Abdelrhman-Sayed70](https://github.com/Abdelrhman-Sayed70)