Pipeline and modeling sample for customer churn
Find a file
2026-07-27 19:57:50 -04:00
db initial load to landing 2026-07-27 12:38:27 -04:00
dbt/kkbox_churn first vertical slice 2026-07-27 19:57:50 -04:00
scripts docstrings and cleanup 2026-07-27 12:53:22 -04:00
.gitignore initial load to landing 2026-07-27 12:38:27 -04:00
readme.md update readme 2026-07-27 19:41:41 -04:00

KKBox's Churn Prediction

(Kaggle)

In this project, customer churn prediction is calculated using data from Kaggle. The primary personal goal is to build on and demonstrate full-stack data modeling, analytics engineering, and machine learning capabilities. While Kaggle is generally focused on machine learning techniques, my implementation focuses on a mixture of data engineering and data science.

Source: https://www.kaggle.com/competitions/kkbox-churn-prediction-challenge

Project pipeline:

  1. Data Extract/Load into PostgreSQL landing tables
  2. Data Cleaning and Feature Engineering using dbt