# souhrid-dey/telecom-customer-churn-prediction architecture

> Architecture diagram of the GitHub repository souhrid-dey/telecom-customer-churn-prediction, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-10-05). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/souhrid-dey/telecom-customer-churn-prediction
- Repository: https://github.com/souhrid-dey/telecom-customer-churn-prediction

## Overview

This is a notebook-based telecom churn analysis and retention-strategy project, not a deployed service. Its main workflow takes customer records through cleaning, exploratory analysis, leakage-safe feature preparation, model comparison and tuning, then threshold and cost-benefit analysis to inform retention actions. The README also describes driver analysis and a six-initiative playbook. The notebooks were not sampled, so the graph shows only documented workflow relationships; it does not infer internal notebook calls or execution order beyond those claims.

Main entry points: [`notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb`](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb) (Final analysis notebook), [`data/customer_churn.csv`](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/data/customer_churn.csv) (Customer data), [`notebooks/final/phase2_master_comparison.csv`](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/phase2_master_comparison.csv) (Comparison results), [`reports/final/SouhridDey_CustomerChurnPrediction_FinalReport.pdf`](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/reports/final/SouhridDey_CustomerChurnPrediction_FinalReport.pdf) (Final report)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_data["Data and exploration"]
  node_customer_data[("Customer data<br/>[customer_churn.csv]")]
  node_analysis_notebook["Final analysis notebook"]
  node_data_hygiene["Data hygiene"]
  node_standard_eda["Standard EDA"]
  node_segmented_eda["Segmented EDA"]
  node_cluster_eda["Clustered EDA"]
end

subgraph group_modeling["Model development"]
  node_feature_prep["Feature preparation"]
  node_model_benchmarks["Model benchmarks"]
  node_model_tuning["Model tuning"]
end

subgraph group_decisions["Business decisions"]
  node_threshold_analysis["Threshold analysis"]
  node_cost_benefit["Cost-benefit analysis"]
  node_driver_analysis["Churn drivers"]
  node_retention_playbook["Retention playbook"]
end

subgraph group_deliverables["Project deliverables"]
  node_comparison_results["Comparison results"]
  node_final_report["Final report"]
end

node_analyst(("Analyst"))

node_analyst -->|"runs analysis"| node_analysis_notebook
node_customer_data -->|"is cleaned"| node_data_hygiene
node_analysis_notebook -->|"implements cleaning"| node_data_hygiene
node_analysis_notebook -->|"performs exploration"| node_standard_eda
node_analysis_notebook -->|"segments analysis"| node_segmented_eda
node_analysis_notebook -->|"clusters analysis"| node_cluster_eda
node_analysis_notebook -->|"prepares features"| node_feature_prep
node_analysis_notebook -->|"benchmarks models"| node_model_benchmarks
node_analysis_notebook -->|"tunes models"| node_model_tuning
node_analysis_notebook -->|"calibrates thresholds"| node_threshold_analysis
node_analysis_notebook -->|"models tradeoffs"| node_cost_benefit
node_analysis_notebook -->|"analyzes drivers"| node_driver_analysis
node_analysis_notebook -->|"informs actions"| node_retention_playbook
node_model_benchmarks -.->|"summarized in"| node_comparison_results
node_analysis_notebook -.->|"documents findings"| node_final_report

click node_customer_data "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/data/customer_churn.csv"
click node_analysis_notebook "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb"
click node_data_hygiene "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb"
click node_standard_eda "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb"
click node_segmented_eda "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb"
click node_cluster_eda "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb"
click node_feature_prep "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb"
click node_model_benchmarks "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb"
click node_model_tuning "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb"
click node_threshold_analysis "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb"
click node_cost_benefit "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb"
click node_driver_analysis "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb"
click node_retention_playbook "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb"
click node_comparison_results "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/phase2_master_comparison.csv"
click node_final_report "https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/reports/final/SouhridDey_CustomerChurnPrediction_FinalReport.pdf"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_customer_data,node_analysis_notebook,node_data_hygiene,node_standard_eda,node_segmented_eda,node_cluster_eda toneBlue
class node_feature_prep,node_model_benchmarks,node_model_tuning toneAmber
class node_threshold_analysis,node_cost_benefit,node_driver_analysis,node_retention_playbook toneMint
class node_comparison_results,node_final_report toneRose
class node_analyst toneIndigo
```

## Components

### Data and exploration

- [Customer data](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/data/customer_churn.csv): `data/customer_churn.csv`
- [Final analysis notebook](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb): `notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb`
- [Data hygiene](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb): `notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb`
- [Standard EDA](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb): `notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb`
- [Segmented EDA](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb): `notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb`
- [Clustered EDA](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb): `notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb`

### Model development

- [Feature preparation](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb): `notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb`
- [Model benchmarks](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb): `notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb`
- [Model tuning](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb): `notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb`

### Business decisions

- [Threshold analysis](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb): `notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb`
- [Cost-benefit analysis](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb): `notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb`
- [Churn drivers](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb): `notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb`
- [Retention playbook](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb): `notebooks/final/SouhridDey_CustomerChurnPrediction_Notebook.ipynb`

### Project deliverables

- [Comparison results](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/notebooks/final/phase2_master_comparison.csv): `notebooks/final/phase2_master_comparison.csv`
- [Final report](https://github.com/souhrid-dey/telecom-customer-churn-prediction/blob/main/reports/final/SouhridDey_CustomerChurnPrediction_FinalReport.pdf): `reports/final/SouhridDey_CustomerChurnPrediction_FinalReport.pdf`

### Other components

- **Analyst**

## Connections

- Analyst → Final analysis notebook: runs analysis (evidence: `README.md`)
- Customer data → Data hygiene: is cleaned (evidence: `README.md`)
- Final analysis notebook → Data hygiene: implements cleaning (evidence: `README.md`)
- Final analysis notebook → Standard EDA: performs exploration (evidence: `README.md`)
- Final analysis notebook → Segmented EDA: segments analysis (evidence: `README.md`)
- Final analysis notebook → Clustered EDA: clusters analysis (evidence: `README.md`)
- Final analysis notebook → Feature preparation: prepares features (evidence: `README.md`)
- Final analysis notebook → Model benchmarks: benchmarks models (evidence: `README.md`)
- Final analysis notebook → Model tuning: tunes models (evidence: `README.md`)
- Final analysis notebook → Threshold analysis: calibrates thresholds (evidence: `README.md`)
- Final analysis notebook → Cost-benefit analysis: models tradeoffs (evidence: `README.md`)
- Final analysis notebook → Churn drivers: analyzes drivers (evidence: `README.md`)
- Final analysis notebook → Retention playbook: informs actions (evidence: `README.md`)
- Model benchmarks → Comparison results: summarized in
- Final analysis notebook → Final report: documents findings
