Journal of Engineering Sciences Vol 13 Issue 08,2022,ISSN:0377-9254

PREDICTION OF LOAN APPROVAL USING MACHINE LEARNING

ALGORITHM
1
M.Tech Student, Department of Computer Science Engineering, Jogaiah Institutes of Technology and Sciences, National
Highway 214, Kalagampudi, Dist, Palakollu, Andhra Pradesh 534268
2
Professor,Principal, Department of Computer Science Engineering, Jogaiah Institutes of Technology and Sciences, National
Highway 214, Kalagampudi, Dist, Palakollu, Andhra Pradesh 534268
1
KONDETI DURGA BHAVANI, 2A.VEERABADHRA RAO

ABSTRACT
Now-a-days obtaining loans from banks have become a very common phenomenon. The banks gain profits from the loans
lent to their customers in the form of interest. While approving a loan, the banks should consider many factors such as
credit history and score, reputation of the person, the location of the property and the relationship with the bank. Many
people apply for loans in the name of home loan, car loan and many more. Everyone cannot be approved based on above
mentioned conditions. There are so many cases where applicant’s applications for loans are not approved by various
finance companies. The right predictions whether to give a loan to a customer or not is very important for the banks to
maximize the profits. The idea behind this project is to use NAIVE BAYES algorithm to predict whether a customer can get
a loan from a bank or not.
Significant. However, the benefits can only be reaped if
the bank has a robust model to accurately predict which
1.1 INTRODUCTION customer's loan it should approve and which to reject, in
order to minimize the risk of loan default.
Finance companies deal with all kinds of loans such as
house loans, vehicle loans, educational loans, personal 1.3 EXISTING SYSTEM:
loans etc... And has a presence across areas such as Existing system uses machine learning algorithms like
cities, towns and village areas. A Customer- first decision tree and random forest. Although we can get
requests for a loan and after that Finance Company good efficiency using the Random Forest classifier and
validates the customer eligibility for the loan and of Decision Tree classifier, we can get great results by using
approve. Details like marital status, gender, NAIVE BAYES classifier. The lender must manually
education, and number of dependents, Income, Loan review each application, supported the main points
Amount, credit history, and others are given in the form provided by the applicant like gender, legal status,
to fill up by the applicants. Therefore, a robust model is education, number of dependents, income, loan amount
built taking those details as input to verify whether an credit.
applicant is eligible to apply for loan or not. The target
variable here is Applicants "Loan Status" and the other 1.4 PROPOSED SYSTEM:
variables are predictors. After building the Machine
The proposed system automates the method of
Learning model a Web Application is to be developed for
determining the applicant’s loan approval status. It
a user interface that allows the user to see instantly if
concentrates on a single or multiple datasets containing
he/she is eligible to get a loan by entering the given
the details of the loan applicants. In the proposed
details.
system, NAIVE BAYES Classification Machine
1.2 MOTIVATION Learning model is been used. NAIVE BAYES
Loan prediction is a very common real-life problem that algorithm is one in all the supervised learning
every finance company faces in their lending operations. algorithms, which is dependent on BAYES theorem
If the loan approval process is automated, it can save a which is applicable for solving classification problems.
lot of man hours and improve the speed of service to the It is good for foretelling the right result in the current
customers. The increase in customer satisfaction and real-world case scenarios and also help the finance
savings in operational costs are agency to offer the finances in the right hands and also
help the applicants in obtaining loan in a much faster
way. The key advantage of this model is, it gives more

www.jespublication.com Page 516

Journal of Engineering Sciences Vol 13 Issue 08,2022,ISSN:0377-9254

accuracy when compared to the existing system.  Evaluating model: Evaluating the best fit model from
1.5 ADVANTAGES: the models used and choosing the model right fitted
for the data.
 The result obtained by the NAIVE BAYES
Algorithm is best compared to any other 3.2 PROCESS MODE USED:
Algorithms.

2. LITERATURE SURVEY

 RAJ, J. S., & ANANTHI, J. V, “Recurrent neural

networks and nonlinear prediction in support vector
Machine” Journal of Soft Computing Paradigm
(JSCP), 1(01), 33-40, 2019.
 X.FRENCIS JENSY, V. P. SUMATHIJANANI
SHIVA SHRI, “An exploratory Data Analysis for
Loan Prediction based on nature of clients”,
International Journal of Recent Technology and
Engineering (IJRTE), Volume-7 Issue-4S, November
2018.
 SHIYANG LIAO, JUNBO WANG, RUIYUN YU,
KOICHI SATO, AND ZIXUE CHENG, “CNN for
situations understanding based on sentiment analysis
of twitter data,” PROCEDIA computer science,
111:376-381, 2017.CrossRef.
 SHIYANG LIAO, JUNBO WANG, RUIYUN YU,
KOICHI SATO, AND ZIXUE CHENG, “CNN for
situations understanding based on sentiment analysis
Architecture of the proposed loan prediction model
of twitter.

3.1 PREDICTIVE MODELLING

Simple NAIVE BAYES classifier algorithm. Banks are in a
Predictive modeling is used to analyze the data and very competitive environment; therefore, the service quality
predict the outcome. Predictive modeling used to during credit risk assessment is very important. When a
predict the unknown event which may occur in the customer demands credit from a bank, the bank should
future. In this process, we are going to create, test and evaluate the credit demand as soon as possible to gain
validate the model. There are different methods in competitive advantage. Additionally, for each credit
predictive modeling. They are learning, artificial demand, the same process is repeated and constitutes a cost
intelligence and statistics. Once we create a model, we for the bank.
can use many times, to determine the probability of
outcomes. So, predict model is reusable. Historical data  Load the data.
is used to train an algorithm. The predictive modeling  Determine the training and testing data.
process is an iterative process and often involves  Data cleaning and preprocessing.
training the model, using multiple models on the same  Fill the missing values with mean values regarding
dataset. numerical values.
 Creating the model: To create a model to run one or  Fill the missing values with mode values regarding
more algorithms on the data set. categorical variables.
 Testing a model: The testing is done on past data to  Outlier treatment.
see how the best model predicts  Apply the modeling for prediction.
 Removing the load identifier.
 Validating a model: Using visualization tools to
 Create the target variable (based on the
validate the model.

www.jespublication.com Page 517

Journal of Engineering Sciences Vol 13 Issue 08,2022,ISSN:0377-9254

requirement). In this approach,  They require special consideration during the software
 Target variable is loan-status. architecture/high-level design phase which increases
 Create a dummy variable for categorical variable costs.
(if required) and split the  Their implementation does not usually map to the
 Training and testing data for validation. specific software sub- system,
 Apply the model  It is tough to modify non-functional once you pass the
 LR method architecture phase.
 RF method
 SVM method 4.5 KEYLEARNING
 Determine the accuracy followed by confusion
matrix The character of the time period, the length of road, the
4.1 FUNCTIONALREQUIREMENTS weather, the bus speed and the rate of road usage are
adopted as input vectors in Support Vector Machine.
 Data Collection
 Data Preprocessing 4.6 SYSTEM REQUIREMENTS
 Training and Testing
OS: Windows or Linux
 Modeling
 Predicting PYTHON IDE: python 2.7.x and above

PYCHARM IDE Required,

4. 2 NON-FUNCTIONAL REQUIREMENTS
JUPYTER NOTEBOOK
 Usability requirement
 Serviceability requirement Setup tools and pip to be installed for 3.6 and above
 Manageability requirement
 Recoverability requirement LANGUAGE: Python Scripting
 Security requirement
4.7 HARDWARE REQUIREMENTS
 Data Integrity requirement
 Capacity requirement RAM: 4GB and Higher
 Availability requirement
 Scalability requirement PROCESSOR: Intel i3 and above

4.3 ADVANTAGES OF NON-FUNCTIONAL HARD DISK: 500GB: Minimum

REQUIREMENT
5.1 DESIGN PHASE
 The nonfunctional requirements ensure the software Design is a multi step process that focuses on data structure,
system follow legal and compliance rules. Software architecture, procedural details and interface
 They ensure the reliability, availability, and between modules. The design process also translates the
performance of the software system requirements into the presentation of software that can be
 They ensure good user experience and ease of operating accessed for quality before coding begins. Computer
the software. software design changes continuously as new methods;
 They help in formulating security policy of the software better analysis and broader understanding evolved.
system. Software design at a relatively early stage in its revolution.
Therefore, software design methodology lacks the depth,
4.4 DISADVANTAGES OF NON-FUNCTIONAL flexibility and quantitative nature that are normally
REQUIREMENT associated with more classical engineering disciplines.
However, the techniques for software design do exist,
 None functional requirement may affect the various criteria for design qualities are available and design
high-level software subsystem notation can be applied. The purpose of the design phase is
to plan a solution of the problem specified by the

www.jespublication.com Page 518

Journal of Engineering Sciences Vol 13 Issue 08,2022,ISSN:0377-9254

requirements document. The design of a system is perhaps 6.2 CODING (SOURCE CODE)
the most critical factor affecting the quality of the software.
It has a major impact on the project during later phases,
particularly during testing and maintenance.

5.2 CONCEPTUAL DESIGN:

Conceptual Design is an early phase of the design process,
in which the broad outlines of function and form of
something are articulated. It includes the design of
interactions, experiences, processes and strategies. It
involves an understanding of people's needs - and how to
meet them with products, services, & processes. The
unified modeling language allows the software engineer to
express an analysis model using the modeling notation that
is governed by a set of syntactic, semantic and pragmatic
rules. A UML system is represented using five different
views that describe the system from a distinctly different
perspective. Each view can be defined by a set of diagrams.

UML is specifically constructed through two different

domains. They are:

 UML analysis modeling, this focuses on the user

model and structural model views of the system.

6.1 IMPLEMENTATION
This chapter includes the implementation of the design and
source code. In this phase the design is translated into code.
Computer programs are written using a conventional
programming language or an application generator.
Programming tools like Compilers, Interpreters, and
Debuggers are used to generate the code. Different high-
level programming languages like C, C++, Pascal, Java,
.Net are used for coding. With respect to the type of
application, the right programming language is chosen.

www.jespublication.com Page 519

Journal of Engineering Sciences Vol 13 Issue 08,2022,ISSN:0377-9254

6.4 USE CASE DIAGRAM

Use case diagram at its simplest is a representation of a

user's interaction with the system that shows the
relationship between the user and the different use cases in
which the user is involved. A use case diagram can identify
the different types of users of a system and the different use
cases and will often be accompanied by other types of
diagrams as well. Actors are the external entities that
interact with the system. The use cases are represented by
either circles or ellipses.

6.3 CLASS DIAGRAM

Class diagrams give an overview of a system by showing

its classes and the relationships among them. Class
diagrams are static - displays what interacts but not what
happens when they do interact. In general a class diagram
consists of some set of attributes and operations. Operations
will be performed on the data values of attributes.
6.5 SEQUENCE DIAGRAM

A Sequence diagram shows object interactions arranged in

time sequence. It depicts the objects and classes involved
in the scenario and the sequence of messages exchanged
between the objects needed to carry out the functionality of
the scenario. Sequence diagrams are typically associated
with use case realizations in the Logical View of the system
under development.

www.jespublication.com Page 520

Journal of Engineering Sciences Vol 13 Issue 08,2022,ISSN:0377-9254

7.1 INTRODUCTION TO TESTING

The purpose of testing is to discover errors. Testing is the

6.6 COMPONENT DIAGRAM process of trying to discover every conceivable fault or
weakness in a work product. It provides a way to check the
Component diagram is a special kind of diagram in UML. functionality of components, sub-assemblies, assemblies
The purpose is also different from all other diagrams and/or a finished product. It is the process of exercising
discussed so far. It does not describe the functionality of the software with the intent of ensuring that the Software
system but it describes the components used to make those system meets its requirements and user expectations and
functionalities. does not fail in an unacceptable manner. There are various
types of tests. Each test type addresses a specific testing
requirement.

7.2 TYPES OF TESTING:

 Unit Testing
 Integration Testing
 Functional Testing
 White Box Testing
 Black Box Testing

7.3 BLACK BOX TESTING

When applied to machine learning models, black box

testing would mean testing machine learning models
6.7 DEPLOYMENT DIAGRAM: without knowing the internal details such as features of the
machine learning model, the algorithm used to create the
The main purpose of the deployment diagram is to
model etc. The challenge, however, is to verify the test
represent how software is installed on the hardware
outcome against the expected values that are known
component. It depicts in what manner a software interacts
beforehand.
with hardware to perform its execution. Both the
deployment diagram and the component diagram are
closely interrelated to each other as they focus on software
and hardware components.

www.jespublication.com Page 521

Journal of Engineering Sciences Vol 13 Issue 08,2022,ISSN:0377-9254

The model gives out the correct output when different

inputs are given which are mentioned. Therefore the
Program is said to be executed as expected or correct
program

 INTEREST RATE

 PURPOSE OF LOAN

8. OUTPUT

8.1 UNIVARATE ANALYSIS

 LOAN AMOUNT

www.jespublication.com Page 522

Journal of Engineering Sciences Vol 13 Issue 08,2022,ISSN:0377-9254

 HOME OWNERSHIP WISE LOAN 8.3RESULT

8.2 YEAR WISE LOAN:

www.jespublication.com Page 523

Journal of Engineering Sciences Vol 13 Issue 08,2022,ISSN:0377-9254

9. CONCLUSION 8 GURLOVE SINGH, AMIT KUMAR GOEL ,”Face

Detection and Recognition System using Digital Image
Therefore, the developed model automates the method Processing” , 2nd International conference on Innovative
of determining the applicant’s creditworthiness. It Mechanism for Industry Application ICMIA 2020, 5-7
focuses on information containing the main points of March 2020, IEEE Publisher.
the loan applicants. In this system NAIVE BAYES
9 AMIT KUMAR GOEL, KALPANA BATRA, POONAM
Classification model is used. In Machine Learning,
PHOGAT,” Manage big data using optical networks”,
NAIVE BAYES classification analysis is one of the
Journal of Statistics and Management Systems “Volume 23,
supervised learning algorithms, which is dependent on
2020, Issue 2, Taylors & Francis.
BAYES theorem and used to solve classification
problems. Hence, it is good for predicting the right
result in the current world scenario and also help the
bank to give the money in the right hands and also help
the people in getting loan in a much faster way. The
main advantage of this system is, it gives more
accuracy.

REFERENCES

1 TOBY SEGARAN, “Programming Collective Intelligence:

Building Smart Web 2.0 Applications.” O’REILLY
MEDIA.

2 DREW CONWAY AND JOHN MYLES WHITE,” Machine

Learning for Hackers: Case Studies and Algorithms to
Get you Started,” O’REILLY MEDIA.

3 TREVOR HASTIE, ROBERT TIBSHIRANI, AND

JEROME FRIEDMAN, “The Elements of Statistical
Learning: Data Mining, Inference, and Prediction,”
Springer, Kindle

4 PHILHYO JIN DO ,HO-JIN CHOI, “Sentiment analysis

of real-life situations using location, people and time as
contextual features,” International Conference on Big Data
and Smart Computing (BIGCOMP), pp. 39-42. IEEE, 2015.

5 BING LIU, “Sentiment Analysis and Opinion Mining,”

Morgan & Claypool Publishers, May 2012.

6 SHIYANG LIAO, JUNBO WANG, RUIYUN YU, KOICHI

SATO, AND ZIXUE CHENG, “CNN for situations
understanding based on sentiment analysis of twitter data,”
PROCEDIA computer science, 111:376-381,
2017.CrossRef.

7 K I RAHMANI, M.A. ANSARI, AMIT KUMAR GOEL,

“An Efficient Indexing Algorithm for CBIR,”IEEE-
International Conference on Computational Intelligence &
Communication Technology, 13-14 Feb 2015.

www.jespublication.com Page 524