Digital Credit Scoring Dataverse is a curated open-access repository dedicated to advancing AI-driven, ethical, and explainable credit scoring frameworks, especially tailored for thin-file and underserved consumers in rural and semi-urban economies. This Dataverse hosts datasets, synthetic benchmarks, modeling workflows, code scripts, evaluation reports, and domain-specific descriptors that underpin data-driven financial inclusion research. It integrates alternative data sources such as mobile transaction records, digital utility payments, behavioral attributes, and remote sensing proxies for livelihood and risk profiling. Rooted in fairness-aware modeling, multi-task learning, and responsible financial AI, the repository aligns with global mandates such as the World Bank’s Digital Financial Inclusion agenda (2025), IMF guidelines on Responsible AI in Finance, and the UN Sustainable Development Goals. All resources emphasize transparency, reproducibility, and equity in algorithmic decision-making. The repository also includes support materials such as data dictionaries, ethical model cards, usage protocols, and evaluation metrics relevant to creditworthiness prediction and sustainability-linked assessments, including carbon scoring. Keywords: Digital Credit, Credit Scoring, Alternative Data, Thin-File Consumers, Machine Learning, Financial Inclusion, ESG, Explainable AI (XAI), Responsible AI, Fairness in AI, Multi-Task Models, Carbon-Aware Scoring.
Featured Dataverses

In order to use this feature you must have at least one published or linked dataverse.

Publish Dataverse

Are you sure you want to publish your dataverse? Once you do so it must remain published.

Publish Dataverse

This dataverse cannot be published because the dataverse it is in has not been published.

Delete Dataverse

Are you sure you want to delete your dataverse? You cannot undelete this dataverse.

Advanced Search

1 to 3 of 3 Results
Jun 29, 2025
Shukla, Deepa, 2025, "Replication Data for: Credit Scoring of Thin File Consumers", https://doi.org/10.7910/DVN/EGAIKO, Harvard Dataverse, V1, UNF:6:tIIKPwlCPuBzLVc1RbTvlQ== [fileUNF]
Abstract: This dataset contains synthetic data designed for evaluating machine learning-based credit scoring models for thin-file consumers. It includes borrower profiles with alternative data attributes such as digital payments, utility bills, and behavioral features. This dataset supports research in AI-driven credit risk assessment, fairness aud...
Jul 26, 2024 - Harvard Dataverse
Shukla, Deepa, 2024, "Replication Data for: Alternative Datasets for Credit Scoring of Thin File Consumers", https://doi.org/10.7910/DVN/TJ6RMQ, Harvard Dataverse, V1
Credit scoring is essential in financial services, allowing institutions to assess consumers' creditworthiness. Traditional credit scoring models heavily rely on extensive transaction history, which often poses a significant challenge for thin-file consumers—individuals with limited credit history. This comprehensive review aims to explore and eval...
May 12, 2024 - Harvard Dataverse
Deepa Shukla, 2024, "Replication Data for: Credit scoring of thin file consumers", https://doi.org/10.7910/DVN/6MLVVI, Harvard Dataverse, V1, UNF:6:tIIKPwlCPuBzLVc1RbTvlQ== [fileUNF]
The rapid evolution of machine learning (ML) offers transformative potential for the credit scoring industry, especially in addressing the challenges faced by "thin-file" consumers who lack substantial credit histories. Traditional credit scoring models often fail to accurately assess these consumers due to insufficient data, leading to potential e...
Add Data

Sign up or log in to create a dataverse or add a dataset.

Share Dataverse

Share this dataverse on your favorite social media networks.

Link Dataverse
Reset Modifications

Are you sure you want to reset the selected metadata fields? If you do this, any customizations (hidden, required, optional) you have done will no longer appear.