galafis/Customer-Behavior-Analytics
Analise de comportamento de clientes com segmentacao RFM (KMeans), previsao de churn (Random Forest) e dashboard Plotly. Python + R + Flask API.
What's novel
Analise de comportamento de clientes com segmentacao RFM (KMeans), previsao de churn (Random Forest) e dashboard Plotly. Python + R + Flask API.
Code Analysis
9 files read · 4 roundsPerforms RFM-based customer segmentation with KMeans, churn prediction with Random Forest, and generates Plotly dashboards from a CSV or synthetic data, with a thin Flask API for data queries and an R script offering hierarchical clustering and RFM scoring.
Strengths
The R script demonstrates more sophisticated analysis (hierarchical clustering, named RFM segments, PCA visualization) than the Python code, and the Jupyter notebook shows good defensive coding with schema validation and clear synthetic-data labeling. The project is complete and runnable end-to-end with a coherent pipeline.
Weaknesses
The Python implementation is a straightforward sklearn wrapper with a crude CLV formula, no data validation beyond file existence, and the Flask server is disconnected from the analytics engine. Tests only verify that methods return non-None objects without checking correctness, and the Dockerfile references a non-existent entry point (src/main.py).
Score Breakdown
Signal breakdown
Innovation
Craft
Traction
Scope
Evidence
Commits
30
Contributors
1
Files
12
Active weeks
8
Repository
Language
Python
Stars
1
Forks
0
License
MIT