3. PROJECT 0VERVIEW :
Credit risk is the risk of a borrower defaulting on a loan, affecting
financial institutions' stability and profitability. Credit risk
assessment uses income and credit history variables. Historically,
manual auditing and statistical techniques were costly and error-
prone. Machine learning (ML) can enhance credit risk assessment
by learning from data, identifying hidden patterns, and producing
accurate results.
4. PROBLEM STATEMENT
Despite the growing demand for credit services in
Tanzania, traditional credit assessment methods exhibit
limitations in accurately evaluating borrower risk profiles,
leading to suboptimal lending decisions and increased
financial risks for lending institutions. Manual
assessment processes often lack comprehensive insights
into individual creditworthiness, resulting in higher
instances of defaults and inefficiencies in credit
provisioning. Moreover, the absence of tailored credit
scoring mechanisms that account for regional nuances
and evolving economic conditions poses a significant
problem in promoting inclusive and responsible lending
practices
5. OBJECTVIES
MAIN OBJECTIVE :
The main objective of this project is to
develop a machine learning-based credit risk
evaluation system that can accurately and
efficiently predict the default probabilities of
borrowers and provide actionable insights
for lenders.
6. SPECIFIC OBJECTIVE
To create a dataset of transactions, historical information, and
other needed data.
To process the dataset for implementation
To create a model which will identify patterns and relationships
between collected financial data
To integrate the system and the model .
7. LITERATURE REVIEW
THEORETICAL LITERATURE/FRAMEWORK
The development and implementation of Credit risk assessment system address
various challenges and opportunities within the financial sector. The theoretical
literature and frameworks surrounding this problem highlight several key areas that
the system seeks to improve, including manual assesment, data processing, and
overall credit risk assesment efficiency. This should be used to improve credit risk
assessment by analysis of large and complex datasets, discovering hidden patterns,
and providing accurate and reliable outcomes.
8. KEY TERMS
- Credit risk: This refers to the potential of financial loss resulting from a borrower's
failure to repay a loan or meet their financial obligations.
- Credit scoring: This is a method used by lenders to evaluate the creditworthiness
of a borrower based on various factors, such as credit history, income, and other
relevant data, to determine the likelihood of repayment.
- - Deep learning: Refers to a subset of machine learning that uses neural
networks with multiple layers to understand and represent complex patterns in data,
often used in tasks like image and speech recognition.
9. METHODOLOGY
RESEARCH APPROACH
Mixed research approach
Because ,Utilizing a mixed research approach in our CREDIT RISK ASSESMENT USING
MACHINE LEARNING can offer valuable insights into user needs, preferences, and the overall
effectiveness of system application.
10. RESEARCH METHOD
Agile Model.
Pros: Iterative, adaptable, encourages user
feedback and continuous improvement
Cons: Can be less structured, requires strong
project management and communication, may
not be suitable for all research phases
11. SYSTEM IMPLEMENTATION
Prototyping :to Develop a functional prototype to visualize and test
key features
Development: Implement the system according to design
specifications
Testing: Ensure the reliability and correctness of the implemented
system
12. SYSTEM IMPLEMENTATION
CODING
Front-End Development :include
languages like HTML, CSS,
JavaScript,Frameworks/Libraries
Back-End Development :
Languages: Node.js (JavaScript),
Python, Java, Ruby, PHP ,
Frameworks
Database Management: database
includes MySQL.
TESTING
Train-Test Split
Because:Testing the model in a real or
simulated deployment environment
to ensure its compatibility,
scalability, and reliability within the
proposed system architecture.
15. SYSTEM REQUIREMENT
HARDWARE REQUIREMENTS
Processor: Multi-core processors core
i5
RAM: 8 GB
Storage: SSD for fast data access
Network: High-speed internet
connection
SOFTWARE TOOLS REQUIREMENTS
Operating Systems: Windows.
Backend Technologies(Python,
mySQLdatabase and apache server)
Frontend Technologies: includes
JavaScript Frameworks(React,
Angular, or Vue.js for frontend
development)
16. PROJECT TIMELINE & BUGET
Gathering requirements from 3-5 week
Quick design form from 6-10 week
Building prototype from 10-12 week
Initial user evaluation from 12-13 week
Refining prototype from 13-15 week
Implementing system and maintenance 15 onward
Budget :