This project present machine learning based solution for churn identification using past data.
The goal here is to provide an accurate model to predict if a customer will churn, given the ~170 columns containing customer behavior, usage patterns, payment patterns, and other features that might be relevant.
The dataset is provided by kaggle: https://www.kaggle.com/competitions/telecom-churn-case-study-hackathon-c54/data
In order to use that dataset on collab, you will need to follow the steps below:
- Create a Kaggle account on https://www.kaggle.com.
- In the "Account" tab of your user profile (https://www.kaggle.com/ <username>/account), select "Create API Token". This will trigger a download of kaggle.json.
- Open the json file and you will find the username and the key.
- Copy the key and paste it into the notebook cell when prompted.
To run locally, clone the repository, go to the diretory and install the requirements.
pip install -r requirements.txt