Case Study BFSI

Uploaded by

borsepratik30

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

106 views3 pages

Case Study BFSI

Uploaded by

borsepratik30

We take content rights seriously. If you suspect this is your content, claim it here.

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 3

Case Study: BFSI

To develop a solution that can evaluate credit-worthiness of Mohan(client) without credit histories by
leveraging data from various centrally governed social security numbers or farming turnover, and other
such features using predictive models for the Howrah District Cooperative Bank Ltd.

1 Data acquisition
Gathered data from a variety of sources, including
government implementing agencies, the banking
Develop Algorithms 2 system, and farmer.
Developed algorithms using AI tools to categorize
the data into 5 specific labels.
3 Constructive Scorecard
Created a scorecard to interpret parameters such
Statistical Tools and Machine
Learning
4 as the ability to pay back, credit history soundness
and susceptibility to default.

Employed statistical methods to gauge the

reliability of the score card. This helps to suspect
the chance of a default.
Solutions and Recommendations
Numeric Data capture Non-Numeric Data capture
Loan Requirement details: Future production plans
Loan amount Feedback about farmers from local authorities like
Interest rate Gram Panchayat
Loan tenure Type of irrigation used
Collateral value (if applicable) Background of co-applicants and guarantors
Loan Purpose: Monsoon patterns
Purchase of "subsidized farm machinery and schemes", More Additional Numeric Information:
The use of the phrase "subsidized" indicates that Mohan is Years of experience in farming.
qualified for government schemes, which in turn confirms
that he has turned in his necessary paperwork, including Ownership of livestock (if any), with net worth.
his AADHAR and PAN documents. Annual turnover of crop yield.
> Eligibility for the Scheme reveals: Value of existing machinery.
Loan amount obtained from informal sources
Income level, land ownership, area under cultivation, age (Ex: moneylenders).
group
Previous and direct benefit transfers- Details of previous
loans (amount, interest rate, tenure, repayment status,
promised collateral).
Sources of information may include government
organizations like SFAC, FPO and NABARD.

Data aggregated through AI power Plotting conclusion using Statistical

techniques tools
Label the data as excellent/good/average/bad/very Perform Kolmogorov - Smirnov Test
bad using tools such as decision tree, neural networks,
clustering. It measures the maximum vertical separation between two
Develop an algorithm to integrate multiple labels to cumulative distribution (Ex. Good and Bad) in a credit
form one parameter head. scorecard.
Develop a score card. Plotting "Confusion Matrix"
False Negative is Misclassifying a Good Loan as Bad loan
and Vice Versa
Evaluates Performance of an algorithm for fraud detection
Personalized financial recommendations
Outsource Generative AI based software such as the TransUnion CIBIL Score that gauges the
creditworthiness of clients when their history is missing.
Collaborate with other banks and develop AI based Financial Literacy products.

Key Takeaways
This case study highlights the importance of data analysis, strategic thinking, and effective implementation
of structured problem solving. In conclusion, to address Mohan's loan requirements and enhance financial
inclusion for farmers, implementing generative AI software like TransUnion CIBIL Score, collaborating with
other banks for AI-driven financial literacy, and creating user-friendly products is essential. These solutions
not only assist individuals like Mohan but also enhance loan processing efficiency and security within the
BFSI sector, ensuring a more robust and inclusive financial landscape.