UIDAI Aadhaar Analysis Project

This repository contains an end‑to‑end analysis of UIDAI Aadhaar enrolment and update activity, from raw API exports to cleaned datasets, exploratory analysis, and presentation‑ready visuals.

Top-Level Files

• HACKATHON_SUBMISSION.ipynb – Main, polished notebook used for submission; loads precomputed analysis_results outputs and cleaned_data to generate all final plots, explanations, and insights.
• aadhaar_analysis.ipynb – Original exploratory notebook used to experiment with loading, cleaning, and aggregating the raw CSV splits.
• aadhaar_analysis_clean.ipynb – Refined version of the exploratory analysis with a cleaner, more streamlined pipeline.
• run_analysis.py – Scripted version of the core analysis pipeline that reads raw split CSVs, builds aggregated tables (national_daily, state_panel, enrolment_by_state), and writes them into analysis_results/.
• .gitignore – Standard ignore rules for virtual environments, raw archives, and editor/OS artefacts.

Data Folders

• api_data_aadhar_enrolment/

  • Contains raw UIDAI enrolment API exports split into multiple CSV files (e.g. api_data_aadhar_enrolment_0_500000.csv, etc.).
  • Each row represents enrolment counts by date, state, district, pincode, and age buckets (0–5, 5–17, 18+).

• api_data_aadhar_demographic/

  • Raw demographic update API exports split into CSV partitions.
  • Columns track counts of demographic updates by date, state, district, pincode, and age buckets (5–17, 17+).

• api_data_aadhar_biometric/

  • Raw biometric update API exports split into CSV partitions.
  • Columns track counts of biometric updates by date, state, district, pincode, and age buckets (5–17, 17+).

• cleaned_data/

  • enrolment_clean.csv – Concatenated and cleaned enrolment data: standardised dates (ISO format), trimmed state/district names, zero‑padded pincodes, non‑negative counts, duplicates removed.
  • demographic_clean.csv – Cleaned demographic updates with the same standardisations applied to date/state/district/pincode fields.
  • biometric_clean.csv – Cleaned biometric updates with consistent schema and quality checks.
  • These files are the preferred inputs for advanced, cross‑dataset analysis.

• analysis_results/

  • national_daily.csv – Aggregated national‑level daily time series of enrolments and updates.
  • state_panel.csv – State‑level panel with total enrolments, total demographic/biometric updates, and intensity measures (e.g. updates per 1,000 enrolments).
  • enrolment_by_state.csv – State‑level enrolment breakdowns by age group.
  • 01_national_timeseries.html – Interactive Plotly view of the national daily time series.
  • 02_top_states_enrolment.html – Top states by enrolment volume and age‑group composition.
  • 03_age_wise_distribution.html – Age‑wise enrolment distribution visualisation.
  • 04_update_intensity_scatter.html – Scatter plot comparing demographic vs biometric update intensity by state.
  • 05_anomalies.html – Anomaly detection view based on z‑scores over the enrolment time series.
  • 06_forecast.html – 14‑day enrolment demand forecast visualisation.
  • INSIGHTS_REPORT.txt – Textual summary of key findings and storylines from the analysis.

How to Use This Project

1. Open HACKATHON_SUBMISSION.ipynb to view the full analysis narrative and final visuals.
2. Use cleaned_data/ if you want to build new models or visualisations on top of a high‑quality, standardised dataset.
3. Use run_analysis.py and the raw api_data_aadhar_*/ folders if you need to fully reproduce the analysis_results/ tables from scratch.