How najlaalhusaini/telchochurn-datamining works
This is a notebook-based telecom churn data-mining project, not a deployed application. Its documented workflow moves from data collection through preprocessing and feature analysis to classification, clustering, evaluation, and performance comparison. The README does not map those stages to individual notebooks, so workflow nodes use the existing Source Code directory rather than assigning responsibilities to particular phases. Relationships shown are documented workflow order; notebook internals were not available for inspection.
Main entry points: Raw_dataset.csv Raw customer data · Source Code Data preprocessing
Components 12
Data preparation
- Raw customer data
DataSet/Raw_dataset.csv - Data preprocessing
Source Code - Prepared customer data
DataSet/Preprocessed_dataset (1).csv - Phase 1 notebook
Source Code/Phase1.ipynb
Modeling and analysis
- Feature analysis
Source Code - Churn classification
Source Code - Customer clustering
Source Code - Phase 2 notebook
Source Code/Phase2.ipynb
Evaluation
- Model evaluation
Source Code - Performance comparison
Source Code - Phase 3 notebook
Source Code/Phase3.ipynb
Other components
- Project analyst
Connections 7
- Project analyst → Raw customer data: provides data
README.md - Raw customer data → Data preprocessing: collects then prepares
README.md - Data preprocessing → Feature analysis: precedes
README.md - Feature analysis → Churn classification: precedes
README.md - Churn classification → Customer clustering: precedes
README.md - Customer clustering → Model evaluation: precedes
README.md - Model evaluation → Performance comparison: precedes
README.md