I am delighted to share that our Digital Data Literacy Program (AIWC/Ujjawal Women Association) has now been extended into a machine learning–ready GitHub repository, integrating reproducible pipelines, fairness audits, and explainability reports.

🌐 Repository link: Digital Data Literacy ML

This development builds upon my prior research in credit scoring for thin-file consumers, where I introduced fairness-aware models and contributed open datasets to the Harvard Dataverse:

  • 📊 Credit Scoring of Thin-File Consumers — DOI: 10.7910/DVN/EGAIKO

The new repository represents a natural progression: applying the same rigor of Responsible AI, transparency, and open data to the domain of digital & financial literacy, impacting 5,000+ women beneficiaries across India.

🔧 Repository Contents

  • Model Card → design, intended use, and limitations
  • Fairness Report → bias and group-level audits
  • SHAP Explainability → feature-level transparency
  • Versioned Releases → ML artifacts with GitHub–Dataverse cross-links

This work aligns with a broader vision: connecting educational outcomes, credit inclusion, and sustainability through responsible AI pipelines.

💡 I warmly invite researchers, practitioners, and open-science collaborators to explore, cite, and extend this work. Let us together advance the intersection of data, models, and societal impact through transparent and ethical machine learning.

#ResponsibleAI #OpenScience #MachineLearning #DigitalDataLiteracy

More Deepa Shukla's questions See All
Similar questions and discussions