A small-sized startup is keen on reducing its customer churn and has hired you as a Machine Learning engineer for this task. Churn rate is a marketing metric that describes the number of customers who leave a business over a specific time period. Every user is assigned a prediction value that estimates their state of churn at any given time. This value is based on:
- User demographic information
- Browsing behavior
- Historical purchase data among other information
As an expert, you are required to build a sophisticated Machine Learning model that predicts the churn score for a website based on multiple features.
The dataset folder contains the following files:
- train.csv: 36992 x 25
- test.csv: 19919 x 24
The columns provided in the dataset are as follows:
| Column Name | Description |
|---|---|
| customer_id | Represents the unique identification number of a customer |
| Name | Represents the name of a customer |
| age | Represents the age of a customer |
| security_no | Represents a unique security number that is used to identify a person |
| region_category | Represents the region that a customer belongs to |
| membership_category | Represents the category of the membership that a customer is using |
| joining_date | Represents the date when a customer became a member |
| joined_through_referral | Represents whether a customer joined using any referral code or ID |
| referral_id | Represents a referral ID |
| preferred_offer_types | Represents the type of offer that a customer prefers |
| medium_of_operation | Represents the medium of operation that a customer uses for transactions |
| internet_option | Represents the type of internet service a customer uses |
| last_visit_time | Represents the last time a customer visited the website |
| days_since_last_login | Represents the no. of days since a customer last logged into the website |
| avg_time_spent | Represents the average time spent by a customer on the website |
| avg_transaction_value | Represents the average transaction value of a customer |
| avg_frequency_login_days | Represents the no. of times a customer has logged in to the website |
| points_in_wallet | Represents the points awarded to a customer on each transaction |
| used_special_discount | Represents whether a customer uses special discounts offered |
| offer_application_preference | Represents whether a customer prefers offers |
| past_complaint | Represents whether a customer has raised any complaints |
| complaint_status | Represents whether the complaints raised by a customer were resolved |
| feedback | Represents the feedback provided by a customer |
| churn_risk_score | Represents the churn risk score that ranges from 1 to 5 |
See the document for the analysis done
The predictive model built in this project was deployed to a web app developed using Python/Flask for the backend and HTML/CSS/Bootstrap for the front end. Live site
- Front-end: HTML/CSS, Bootstrap
- Backend: Python/Flask
- Deployment: Render
- Machine Learning Algorithm: XGBoost classifier
For this project, having a Python version less than v3.12 on your system is preferable to avoid compatibility issues during dependencies installation. Python 3.10.11 was used in the project.
Check your Python version by inputting python --version in the command-line prompt.
Follow these steps to set up the development environment for the churn prediction project:
- Create a Project Directory:
- Open your terminal and create a new directory on your local machine called churn_prediction:
mkdir churn_prediction
- Open your terminal and create a new directory on your local machine called churn_prediction:
- Navigate to this new Project Directory:
cd churn_prediction - Create a Python virtual environment inside the churn_prediction directory:
python -m venv .venv - Activate the Virtual Environment:
.venv\Scripts\activate - While still in the churn_prediction directory, clone this project repository into your local:
git clone https://github.com/aayomide/churn-risk-prediction.git - Navigate to the Cloned Repository:
cd churn-risk-prediction - Install all required dependencies using pip:
pip install -r requirements.txt - Start the Web Application. Run the following command to start the Flask web application:
flask run