How shraddhadangat/predictive-churn-analysis works
This is a notebook-based customer churn prediction project. Its documented workflow loads the CSV, prepares and splits the data, fits logistic regression, evaluates the model, and identifies high-risk customers from churn probabilities. The README establishes these stages and the reported metrics; no notebook source was available to verify implementation details or additional wiring.
Main entry points: Predictive_Churn_Analysis.ipynb Logistic Regression · customer_churn_model_v2.csv Churn Dataset
Components 11
Data Preparation
- Data Loading
Predictive_Churn_Analysis.ipynb - Preprocessing
Predictive_Churn_Analysis.ipynb - One-Hot Encoding
Predictive_Churn_Analysis.ipynb - MinMax Scaling
Predictive_Churn_Analysis.ipynb - Train-Test Split
Predictive_Churn_Analysis.ipynb
Modeling
- Logistic Regression
Predictive_Churn_Analysis.ipynb
Prediction & Analysis
- Model Evaluation
Predictive_Churn_Analysis.ipynb - Churn Probability
Predictive_Churn_Analysis.ipynb - High-Risk Customers
Predictive_Churn_Analysis.ipynb
Other components
- Analyst
- Churn Dataset
customer_churn_model_v2.csv
Connections 10
- Analyst → Data Loading: starts analysis
- Churn Dataset → Data Loading: loads data
- Data Loading → Preprocessing: prepares data
- Preprocessing → One-Hot Encoding: encodes features
- One-Hot Encoding → MinMax Scaling: scales features
- MinMax Scaling → Train-Test Split: splits data
- Train-Test Split → Logistic Regression: fits model
- Logistic Regression → Model Evaluation: evaluates model
- Logistic Regression → Churn Probability: predicts probability
- Churn Probability → High-Risk Customers: identifies risk