SMART INDIA HACKATHON 2025
Ministry/Organization/Student Innovation: Student Innovation
Problem Statement ID: SIH25126
Problem Statement Title: Smart Loan Eligibility Prediction for Rural Borrowers with Alternate data using
AI/ML
Theme: Agriculture, FoodTech & Rural Development
Problem Statement Category: Software
Institute Name: Dr. Vishwanath Karad MIT World Peace University, Pune, Bharat
Institute Code: U-0938
Team ID: SIH250040
Team Name: Data Dons
Data Dons IDEA KEY
PROPOSED SOLUTION FEATURES
METHODOLOGY
Our proposed solution is an AI/ML-powered loan eligibility AI/ML loan eligibility prediction:
system designed specifically for rural borrowers such as leverages advanced models to
accurately
farmers, artisans, traders, and micro-enterprises, to predict
forecast borrower eligibility.
loan eligibility and repayment capacity with higher accuracy.
Bias-free and inclusive credit
By delivering reliable eligibility predictions and risk assessment: ensures fair evaluation
assessments, the solution empowers financial institutions to of applicants from diverse
extend credit responsibly, reduces default risks, and backgrounds without discrimination.
significantly improves financial inclusion for underserved rural
communities. Faster loan processing: accelerates loan
approval through AI-driven automation,
NEED reducing manual delays and ensuring
timely access to credit for rural borrowers.
To bridge the gap between underserved rural borrowers and
formal financial institutions by using AI/ML to analyze
alternative data, enabling fair, transparent, and inclusive credit Explainable and transparent
access while reducing default risk. decisions: provides clear reasoning
behind approvals or rejections to build
USP trust.
Lending made simple, trust made stronger! Reducing default for lenders:
Our AI/ML-powered platform delivers fair, transparent, and minimizes financial risks by identifying
and preventing high-risk loans.
accurate loan decisions.
It safeguards banks while opening doors for rural borrowers.
Faster, safer, and smarter approvals drive true financial
Easy to use: designed with simplicity to
inclusion. ensure smooth adoption by both lenders
Turning credit into confidence, and opportunity into growth. and borrowers.
2
@SIH Idea submission- Template
Data Dons TECHNICAL APPROACH
TECHNOLOGY STACK ALGORITHMS USED
WEBSITE WORKFLOW
SDG XG
ClassifierRandom forest Boost Logistic
Classifier Regression
UI/UX
Credi
t
@SIH Idea submission- Template 3
FEASIBILITY AND VIABILITY
Data Dons
FEASIBILITY SWOT ANALYSIS
• Technical: AI/ML can use alternative data where formal credit history is STRENGTHS WEAKNESSES
missing; cloud and mobile platforms make it scalable and accessible. • Financial Inclusion: Helps rural • Data Availability & Quality:
• Operational: Easy to integrate with banks/MFIs and requires minimal farmers, artisans, and micro- Rural borrowers may lack digital
entreprises access formal credit. payment history or accurate
training. • Explainable records.
AI: Transparent
• Financial: Cuts manual assessment costs, reduces defaults, and ensures scoring builds trust among low- • Initial Adoption Resistance:
long-term savings for lenders. literacy borrowers Micro financial institutes and small
banks may distrust AI-driven
VIABILITY • Scalability: Can be adopted by
microfinance institutions, rural decisions initially.
banks, and cooperatives at large • Infrastructure Limitations: Poor
scale. internet connectivity and
• Social: Promotes financial inclusion and builds borrower trust in formal smartphone penetration in some
regions.
banking.
• Economic: Sustainable model with banks/NBFCs, creating value for both
lenders and borrowers. OPPORTUNITIES THREATS
• Scalability: Can expand across regions and sectors, with AI models • Government Push for Financial • Bias & Fairness Risks: If not
improving over time. Inclusion: Schemes like PMMY monitored, AI may unintentionally
(MUDRA), Jan Dhan Yojana, and discriminate against certain
Digital India align with this solution.
OVERCOME WEAKNESSES AND • Partnership Potential: Can
communities.
• Data Privacy & Security: Risk of
collaborate with NBFCs, MFIs, Agri-
• THREATS
Data Availability & Quality: Use alternative data sources (mobile usage, utility tech startups, and cooperatives.
misuse of sensitive personal and
financial data.
• Expansion Beyond Loans: Model • Regulatory Challenges:
bills, transaction history) and apply data cleaning techniques to improve can extend to insurance, savings, Compliance with RBI, data
reliability. and financial literacy platforms. protection, and KYC norms.
• Adoption Resistance: Conduct awareness workshops with banks, MFIs, and
rural borrowers; offer easy-to-use UI/UX to build trust.
• Data Privacy & Security: Use end-to-end encryption, anonymization, and
secure cloud storage; comply with ISO/PCI standards.
• Regulatory Challenges: Align with RBI, Data Protection Bill, and KYC norms;
maintain audit logs for compliance.
• Market & Technology Risks: Ensure scalable architecture with regular updates,
backups, and disaster recovery planning.
@SIH Idea submission- Template 4
Data Dons
BENEFIT
IMPACT S
Improved Financial Inclusion:
Rural borrowers like farmers, artisans, and traders get fair access to
formal credit despite lacking traditional credit history.
Social Economic Environmental
Reduction in Exploitation by Moneylenders:
Cuts down the long manual approval process into near-instant
eligibility checks, reducing waiting time from weeks to minutes.
Economic Growth in Rural Areas:
• Enables rural
Borrowers get clear insights into why they are eligible (or not), • Promotes • Encourages
borrowers to
building trust in the banking system. financial adoption of eco-
invest in
inclusion for friendly farming
productivity
underserved and green
Economic Growth in Rural Areas: improvements like
communities. enterprises.
AI/ML ensures more accurate risk prediction, reducing non- better seeds,
performing assets (NPAs) for banks and financial institutions. tools, and
• Reduces defaults • Lowers
• Builds trust
inventory.
and NPAs, paperwork and
and
ensuring carbon footprint
Enhanced Living Standards: transparency
sustainable and via digital
Access to affordable credit enables households to invest in in lending.
profitable rural processes..
education, healthcare, housing, and livelihood improvements.
• Improves quality banking rural
• Stimulates
operations.by
economies • Promotes
of life through
Stronger Rural Communities: creating jobs, sustainable
better access to
With financial stability, communities experience collective supporting resource use in
credit for
growth, reduced migration to cities, and better resilience to entrepreneurship, rural
education,
economic shocks. and increasing communities..
health, and
livelihoods. household income.
@SIH Idea submission- Template 5
Data Dons
RESEARCH AND REFERENCES
RESEARCH
• AI-driven credit scoring has improved lending accuracy by 85% and significantly reduced non-performing loans, strengthening overall financial stability.
• Modern AI systems now analyze 30+ borrower features, including digital footprints like social media and mobile usage, improving prediction accuracy and financ
ial inclusion.
• Decision Tree models provide 88% accuracy with high interpretability, ensuring transparency and regulatory compliance in credit risk assessment.
• Deployment & Adoption Path – Target partners include Small Finance Banks & NBFC-MFIs, aligning with RBI’s Priority Sector Lending (PSL) framework and SFB O
perating Guidelines to drive nationwide financial inclusion.
• RBI-aligned Lending Rules – Our scoring system will integrate RBI’s RFML (2022) guidelines on borrower protection, repayment-to-income caps, and transparent
pricing, ensuring regulatory compliance.
REFERENCES
https://rrjournals.com/index.php/rrijm/article/view/2130?utm
https://www.rbi.org.in/commonman/english/scripts/FAQs.aspx
https://ijsra.net/sites/default/files/IJSRA-2024-2258.pdf
5 YEARS PLAN
YEAR YEAR YEAR YEAR YEAR
Prototype Integrate advanced Expand pilot to Fraud Nationwide
1 2
features like an AI
3
more institutions, 4
detection,Credit 5
development, pilot rollout and
with a small Chatbot for borrower gather more data, bureau integration Global scale
microfinance support and KYC and improve the and automation expansion
institution. Verification. model.
@SIH Idea submission- Template 6
IMPORTANT INSTRUCTIONS
Please ensure below pointers are met while submitting the Idea PPT:
1.Kindly keep the maximum slides limit up to six (6). ( Including the title slide)
2.Try to avoid paragraphs and post your idea in points /diagrams / Infographics /pictures
3.Keep your explanation precise and easy to understand
4.Idea should be unique and novel.
5.You can only use provided template for making the PPT without changing the idea
details pointers (mentioned in previous slides).
6.You need to save the file in PDF and upload the same on portal. No PPT, Word Doc or
any other format will be supported.
Note - You can delete this slide (Important Pointers) when you upload the details of
your idea on SIH portal.
@SIH Idea submission- Template 7
