How hoaihue04/rfm-kmeans-clustering works
This repository documents an Online Retail customer-segmentation workflow: a Python notebook cleans transactions, calculates RFM features and clusters customers; PostgreSQL transforms staging data into a star-schema warehouse; and a Power BI report presents business and segment insights. The README also describes campaign recommendations and seasonal/product analysis. The graph reflects documented relationships only; source excerpts were unavailable, so artifact-level details beyond the README and filenames are omitted.
Main entry points: rfm_kmeans_clustering.ipynb RFM and K-Means · 01_create_dw_tables.sql Create DW Tables · 02_load_dw_from_staging.sql Load Warehouse
Components 9
Customer Analysis
- RFM and K-Means
rfm_kmeans_clustering.ipynb - K-Means Model
models/kmeans_rfm_k4-20111210.joblib
Data Warehouse
- Create DW Tables
sql/01_create_dw_tables.sql - Load Warehouse
sql/02_load_dw_from_staging.sql - Validate Data
sql/03_validate.sql - PostgreSQL Staging and DW
Business Intelligence
- Power BI Dashboard
Other components
- Online Retail II
- Business User
Connections 6
- Online Retail II → RFM and K-Means: provides transactions
README.md - RFM and K-Means → PostgreSQL Staging and DW: writes staging
README.md - Create DW Tables → PostgreSQL Staging and DW: creates tables
README.md - Load Warehouse → PostgreSQL Staging and DW: loads star schema
README.md - PostgreSQL Staging and DW → Power BI Dashboard: imports data
README.md - Power BI Dashboard → Business User: shows insights
README.md