Professional Documents
Culture Documents
VISION
To produce Competent Engineering Graduates & Managers with a strong base of Technical &
Managerial Knowledge and the Complementary Skills needed to be Successful Professional
Engineers & Managers.
MISSION
To fulfill the vision by imparting Quality Technical & Management Education to the Aspiring
Students, by creating Effective Teaching/Learning Environment and providing State – of the – Art
Infrastructure and Resources.
CORE VALUES
Commitment to Continual Improvement
Inclusiveness
Enhancement of Student Learning through Innovative Instructional Methods
Collaboration with Industry and other Institutions for Mutual Benefits Pursuit of Excellence in
all Activities.
Approved by AICTE, New Delhi, permanently affiliated to JNTUA, Ananthapuramu
Accredited by NBA, Ayyalurumetta, Nandyal, Kurnool (DT), AP-518502
Department of Computer Science and Engineering
VISION
st
To produce Industry ready Software Engineers to meet the challenges of 21 Century.
MISSION
Impart core knowledge and necessary skills in Computer Science and Engineering through
innovative teaching and learning methodology.
Inculcate critical thinking, ethics, lifelong learning and creativity needed for industry and
society
Cultivate the students with all-round competencies, for career, higher education and self-
employability
Program Outcomes (PO's):
PO Graduate PO Statements
Number Attributes
9 Individual and team Function effectively as an individual, and as a member or leader in di verse teams,
work and in multidisciplinary settings.
TITLE PO1 PO2 PO3 PO4 PO5 PO6 PO7 PO8 PO9 PO10 PO11 PO12
Evaluation of
Machine Learning
Algorithms for the
Detection of Fake 2 2 2 2 3 3 3 2
Bank Currency
A PROJECT REPORT ON
“EVALUATION OF MACHINE LEARNING ALGORITHMS FOR
THE DETECTION OF FAKE BANK CURRENCY”
A Project report submitted in partial fulfillment of the requirement for the award
of degree of
BACHELOR OF TECHNOLOGY
IN
By
G.INDU 20AM1A0517
D.ABDUL BASHA 20AM1A0501
D.SIVA KISHORE 20AM1A0549
Under the Esteemed guidance of
ESTD-2007
AYYALURU (V), NANDYAL-518503 Kurnool Dist. A.P.,
Batch: 2020-2024
SVR ENGINEERING COLLEGE
Approved by A.I.C.T.E., New Delhi& Affiliated to JNTU, Anantapur.
CERTIFICATE
This is certified that the technical seminar entitled, “Evaluation of Machine
Learning Algorithms for the Detection of Fake Bank Currency” is being
submitted by G.INDU(20AM1A0517), D.ABDUL BASHA(20AM1A0501),
D.SIVA KISHORE(20AM1A0549) in partial fulfillment of the requirements
for the award of Degree of BACHELOR OF TECHNOLOGY in
COMPUTER SCIENCE & ENGINEERING to SVR ENGINEERING
COLLEGE is a record of bonafide work carried out by them during the academic
year 2023-2024 under our guidance and supervision.
The results presented in this technical seminar have been verified and found to be
satisfactory.
External Examiner
ACKNOWLEDGEMENT
We wish to regard a deep sense of gratitude and sincere thanks to our guide
Mrs.M.HIMA BINDU, M.Tech Assistant professor, Department of Computer science
and Engineering, SVR ENGINEERING COLLEGE who has guided and motivated
with scholar advice, meticulous care and deep interest throughout the project
duration. She has been a constant source and encouragement and offered valuable
guidance and suggestions from time to time and helped us at various stages of the
project work. Her motivation has been a source of inspiration and model for us to
follow .As stated in the past, “THERE IS NO WAY WITHOUT A GUIDE IN LIFE
“It would not have been possible for us to complete this wonderful and challenging
project without our beloved guide, Mrs.M.HIMA BINDU, M.Tech
G.INDU 20AM1A0517
D.ABDUL BASHA 20AM1A0501
D.SIVA KISHORE 20AM1A0549
ABSTRACT
The one important asset of our country is Bank currency and to create
discrepancies of money miscreants introduce the fake notes which resembles to
original note in the financial market. During demonetization time it is seen that so
much of fake currency is floating in market. In general by a human being it is very
difficult to identify forged note from the genuine not instead of various parameters
designed for identification as many features of forged note are similar to original one.
To discriminate between fake bank currency and original note is a challenging task.
So, there must be an automated system that will be available in banks or in ATM
machines. To design such an automated system there is need to design an efficient
algorithm which is able to predict whether the banknote is genuine or forged bank
currency as fake notes are designed with high precision. In this project six
supervised machine learning algorithms are applied on dataset available on UCI
machine learning repository for detection of Bank currency authentication. To
implement this we have applied Support Vector machine, Random Forest, Logistic
Regression, Naïve Bayes, Decision Tree, K- Nearest Neighbor by considering three
train test ratio 80:20, 70:30 and 60:40 and measured their performance on the basis
various quantitative analysis parameter like Precision, Accuracy, Recall, MCC, F1-
Score and others. And some of SML algorithm is giving 100 % accuracy for
particular train testratio.
LIST OF CONTENTS
CHAPTER 1:INTRODUCTION 1-2
3.3 SDLC 10
3.3.7 Maintenance 14
5.1 Python 24
6.1.1 Implementation 35
6.1.2 Testing 35
CHAPTER 8:CONCLUSION 45
3.3.1 SDLC 10
1 Dataset 39
2 Run.bat File 40
3 Upload Training Dataset 40
4 Note Lengths 41
5 Data Preprocessing 41
6 Run Algorithms 42
7 Table&Graph 43
8 Upload TestData 44
9 Check Genunie or Fake 44
Evaluation of Machine Learning Algorithms for the Detection of Fake
Bank Currency
1. INTRODUCTION
Financial activities are carrying out in every second by many persons in which
one most important asset of our country is Banknotes. Fake notes are introduced the
market to create discrepancies in the financial market, even they resemble to the original
note. Basically they are illegally created to complete various task. In 1990 forgery issue is
not much of concern but as in late 19th century forgery has been increasing drastically. In
20th century technology is increasing very vastly that will help the frauds to generate fake
note whose resemblance is like genuine not and it is very difficult to discriminate them
This will lead to financial market to its lowest level. Government have designed banknote
with some features by which we can identify genuine. But frauds are creating fake note
with almost same features with nice accuracy that make it very difficult to identify
genuine note. So, now a day it is required that bank or ATM machines must have some
system that can identify the forged note from the genuine note. To determine the
legitimacy of the banknote artificial intelligence and Machine learning (ML) can play a
vital role to design such a system that ca identify forged note from the genuine bank
currency Now days, supervised machine learning (SML) approaches for classification
problem is widely used. For medical disease its shows even promising results. Few
authors have only applied SML algorithms on bank currency authentication. To identify
weather a note is genuine or fake we have to develop an automation system. Initially, the
input is an image of note and from different image processing techniques we can extract
the features of note. Further these images are given as an input to the SML algorithms to
predict whether note is original or fake. In review we can see that not much of work is
done on this side.
Contribution of the project: First we have visualized the dataset taken from UCI
ML repository using different types of plotting, pre-processed the data. Further, SML
algorithms Logistic regression (LR), Naive Bayes (NB), Decision tree (DT), Random
tree (RT), KNN, SVM are applied on the data set which contains the features extracted
from the bank currency to classify them as whether it is original or not. For analysis of
their result, we have applied SML algorithms on dataset with three different train test
ratio and their results are compared on the basis of different SML algorithms standard
evaluation parameters like MCC, F1 Score, NPV, NDR, accuracy and others.
The one important asset of our country is Bank currency and to create
discrepancies of money miscreants introduce the fake notes which resembles to original
note in the financial market. In general by a human being it is very difficult to identify
forged note from the genuine not instead of various parameters designed for
identification as many features of forged note are similar to original one.. So, there must
be an automated system that will be available in banks or in ATM machines. To design
such an automated system there is need to design an efficient algorithm which is able to
predict whether the banknote is genuine or forged bank currency as fake notes are
designed with high precision. In this project six supervised machine learning algorithms
are applied on dataset available on UCI machine learning repository for detection of
Bank currency authentication. To implement this we have applied Support Vector
machine, Random Forest, Logistic Regression, Naïve Bayes, Decision Tree, K- Nearest
Neighbor by considering three train test ratio 80:20, 70:30 and 60:40 and measured their
performance on the basis various quantitative analysis parameter like Precision,
Accuracy, Recall, MCC, F1-Score and others. And some of SML algorithm is giving
100 % accuracy for particular train test ratio.
2. LITERATURE SURVEY
With the advance of digital imaging technologies, color scanners and laser
printers make it increasingly easier to produce counterfeit banknotes with high
resolution. Almost every country in the world face the problem of counterfeit currency
notes.Even receiving Fake notes from ATM counters, vending machines and during
elections have also been reported at some places. There is a need to design a system that
is helpful in recognition of counterfeit notes. In this project, we propose a system based
on multiple-kernel support vector machines for counterfeit banknote recognition.
The ROC curve is one of the most common statistical tools useful to
assess classifier performance. The selection of the best classifier when ROC curves
intersect is quite challenging. A novel approach for model comparisons when ROC
curves show intersections is proposed. In particular, the relationship between ROC
orderings and stochastic dominance is investigated in a theoretical framework and a
general class of indicators is proposed which is coherent with dominance criteria also
when ROC curves cross. Furthermore, a simulation study and a real application to credit
risk data are proposed to illustrate the use of the new methodological approach.
All banknotes have security features which are intended to help determine
whether they are false or genuine. Typically, however, the general public has limited
knowledge of where on a banknote these security features can be found. Here, we tested
whether counterfeit detection can be improved with the help of salient elements,
designed to guide bottom-up visuospatial attention.
3. SYSTEM ANALYSIS
Financial activities are carrying out in every second by many persons in which
one most important asset of our country is Banknotes. Fake notes are introduced in the
market to
Create discrepancies in the financial market, even they resemble to the original
note. Basically they are illegally created to complete various task. In 1990 forgery issue
is not much of concern but as in late 19th century forgery has been increasing
drastically. In 20th century technology is increasing very vastly that will help the frauds
to generate fake note whose resemblance is like genuine not and it is very difficult to
discriminate them. This will lead to financial market to its lowest level. To stop this and
to conduct smooth transaction circulation forged bank currency must be conserved.
Less Accuracy
Limited effectiveness
Time-consuming
Costly errors
More Accuracy
Faster Detection Speed
3.2.1 Modules
To implement this project we have designed following modules
1) Upload Fake Currency Dataset: : using this module we will upload Upload Fake
Currency Dataset
2) Dataset Preprocessing: : using this module we will Dataset Preprocessing
3) Run KNN Algorithm : using this module we will Run KNN Algorithm
4) Run Naïve Bayes: : using this module we will Run Naïve Bayes
5) Run Decision Tree: using this module we will upload Run Decision Tree
6) Run SVM: using this module we will Run SVM
7) Run Random Forest : using this module we will Run Random Forest
8) Run Logistic Regression: using this module we will Run Logistic Regression
9) Run Extension LightGBM Algorithm : using this module we will Run Extension
LightGBM Algorithm Comparison Graph
10) Fake Currency Detection from Test Data: using this module we will Fake
Currency Detection from Test Data
FUNCTIONAL REQUIREMENTS
SOFTWARE REQIREMENTS:
System Attributes:
1) filename
2) X,Y
3) dataset
4) main
5) text
No need
USECASE
Use cases - Use cases describe the interaction between the system and external users that
leads to achieving particular goals.
1) Upload Fake Currency Dataset
2) Dataset Preprocessing
3) Run KNN Algorithm
4) Run Naïve Bayes
5) Run Decision Tree
6) Run SVM
7) Run Random Forest
8) Run Logistic Regression
9) Run Extension LightGBM Algorithm
10) Comparison Graph
11) Fake Currency Detection from Test Data
User Stories:
In propose project author has used many traditional algorithms such as KNN,
Decision Tree, SVM, Random Forest, Logistic Regression and Naïve Bayes but author
has not used any advance machine learning algorithms such as ELM,XGBOOST, MLP
or LightGBM so as extension we have added LightGBM algorithm and compare its
performance with existing algorithms.
opencv-python==4.5.1.48
keras==2.3.1
tensorflow==1.14.0
protobuf==3.16.0
h5py==2.10.0
sklearn-extensions==0.0.2
scikit-learn==0.22.2.post1
Numpy
Pandas
NON-FUNCTIONAL REQUIREMENT
Usability: Usability is a quality attribute that assesses how easy user interfaces are to
use. The word "usability" also refers to methods for improving ease-of-use during the
design process.(how it was handle entire project easy)
Security: the quality or state of being secure: such as. a : freedom from danger : safety.
b : freedom from fear or anxiety. c : freedom from the prospect of being laid off job
security.
Readability: Readability is the ease with which a reader can understand a written text.
Availability: the quality or state of being available trying to improve the availability of
affordable housing. 2 : an available person or thing.
Umbrella
DOCUMENT CONTROL
Umbrella
Business Requirement
Documentation
• Feasibility Study
• TEAM FORMATION
ANALYSIS &
Requirement • Project
Specification DESIGN CODE UNIT TEST
s Gathering ASSESSMEN
INTEGRATIO ACCEPTANCE
N & SYSTEM TEST
TESTING DELIVERY/IN
Umbrella
TRAINING
SDLC is nothing but Software Development Life Cycle. It is a standard which is used by
software industry to develop good software.
Stages in SDLC:
Requirement Gathering
Analysis
Designing
Coding
Testing
Maintenance
The requirements gathering process takes as its input the goals identified in the
high-level requirements section of the project plan. Each goal will be refined into a set
of one or more requirements. These requirements define the major functions of the
intended application, define operational data areas and reference data areas, and define
the initial data entities
Analysis Stage:
The planning stage establishes a bird's eye view of the intended software
product, and uses this to establish the basic project structure, evaluate feasibility and
risks associated with the project, and describe appropriate management and technical
approaches.
The design stage takes as its initial input the requirements identified in the
approved requirements document. For each requirement, a set of one or more design
elements will be produced as a result of interviews, workshops, and/or prototype efforts.
Design elements describe the desired software features in detail, and generally include
functional hierarchy diagram.
The development stage takes as its primary input the design elements described
in the approved design document. For each design element, a set of one or more
software artifacts will be produced. and functions.
During the integration and test stage, the software artifacts, online help, and test
data are migrated from the development environment to a separate test environment. At
this point, all test cases are run to verify the correctness and completeness of the
software.
During the installation and acceptance stage, the software artifacts, online help, and
initial production data are loaded onto the production server. At this point, all test cases
are run to verify the correctness and completeness of the software.
Maintenance:
ECONOMIC FEASIBILITY
A system can be developed technically and that will be used if installed must still be
a good investment for the organization. In the economical feasibility, the development
cost in creating the system is evaluated against the ultimate benefit derived from the new
systems.
OPERATIONAL FEASIBILITY
Proposed projects are beneficial only if they can be turned out into information
system. That will meet the organization’s operating requirements. Operational feasibility
aspects of the project are to be taken as an important part of the project implementation.
This system is targeted to be in accordance with the above-mentioned issues.
TECHNICAL FEASIBILITY
User Interface
The user interface of this system is a user friendly python Graphical User Interface.
Hardware Interfaces
The interaction between the user and the console is achieved through python
capabilities.
Software Interfaces
3.4.3.SYSTEM REQUIREMENTS
HARDWARE REQUIREMENTS:
SOFTWARE REQUIREMENTS:
4. SYSTEM DESIGN
The class diagram is the main building block of object oriented modeling. It is used
both for general conceptual modeling of the systematic of the application, and for
detailed modeling translating the models into programming code. Class diagrams can
also be used for data modeling. The classes in a class diagram represent both the main
objects, interactions in the application and the classes to be programmed. In the diagram,
classes are represented with boxes which contain three parts:
Dataset Preprocessing
Upload Fake Currency Dataset
User
Run SVM
COLLABORATION DIAGRAM
Upload Fake
Currency Dataset
Dataset
Preprocessing
Run KNN
Algorithm
Run Naïve
Bayes
Run Decisio
User n Tree
Run SVM
Run Rando
m Forest
Run Logistic
Regression
Compariso
n Graph
Fig:4.4.Component Diagram
Upload
Fake
Dataset
Preproc
Run KN
N
Run
Naïve
Run
Decisio
Run
SVM
User
Run
Random
Run
Logistic
Run
Extensi
Fake Compari
Currenc son
Fig:4.5.DEPLOYMENT DIAGRAM
Fig:4.6.Activity Diagram
As the name suggests, Data Flow Diagram (DFD) is an illustration that explicates the
passage of information in a process. A DFD can be easily drawn using simple symbols.
Additionally, complicated processes can be easily automated by creating DFDs using
easy-to-use, free downloadable diagramming tools. A DFD is a model for constructing
and analyzing information processes. DFD illustrates the flow of information in a
process depending upon the inputs and outputs. ADFD can also be referred to as a
Process Model..
Fig:4.7.Data Flow
5. IMPLEMETATION
5.1 Python
History of Python:
Python is a fairly old language created by Guido Van Rossum. The design began
in the late 1980s and was first released in February 1991.
In late 1980s, Guido Van Rossum was working on the Amoeba distributed
operating system group. He wanted to use an interpreted language like ABC (ABC has
simple easy-to-understand syntax) that could access the Amoeba system calls.
Features of Python:
Applications of Python:
6. TESTING
Implementation and Testing:
Implementation
Testing
Testing is the process where the test data is prepared and is used for testing the
modules individually and later the validation given for the fields. Then the system
testing takes place which makes sure that all components of the system property
functions as a unit.
System Testing
Testing has become an integral part of any system or project especially in the
field of information technology. The importance of testing is a method of justifying, if
one is ready to move further, be it to be check if one is capable to with stand the rigors
of a particular situation cannot be underplayed and that is why testing before
development is so critical.
Module Testing
Integration Testing
After the module testing, the integration testing is applied. When linking the
modules there may be chance for errors to occur, these errors are corrected by using this
testing. In this system all modules are connected and tested. The testing results are very
correct. Thus the mapping of jobs with resources is done correctly by the system.
Acceptance Testing
When that user fined no major problems with its accuracy, the system passers
through a final acceptance test. This test confirms that the system needs the original
goals, objectives and requirements established during analysis without actual execution
which elimination wastage of time and money acceptance tests on the shoulders of users
and management, it is finally acceptable and ready for the operation.
7. SCREENSHOTS
Now-a-days all countries are facing major financial blow due to fake currency and to
overcome from this problem author is employing various machine learning algorithms to
predict weather currency is fake or not. As machine algorithms already proves it success
in almost all fields such as Healthcare prediction, cyber-attack prediction, credit card
fraud detection and many more. So author is suggesting to add machine learning power
on fake currency detection.
In propose project author has used many traditional algorithms such as KNN, Decision
Tree, SVM, Random Forest, Logistic Regression and Naïve Bayes but author has not
used any advance machine learning algorithms such as ELM, XGBOOST, MLP or
LightGBM so as extension we have added LightGBM algorithm and compare its
performance with existing algorithms.
To implement this project author has used UCI Machine Learning Fake currency dataset
which contains height and width of currency and below is the dataset screen shots
In above dataset screen first row contains dataset column names and remaining are the
dataset values and first column contains counterfeit class label where 0 means Genuine
and 1 means Fake
In above screen click on ‘Upload Fake Currency Dataset’ button to upload dataset and to
get below output
Fig:Note Lengths
In above screen dataset loaded and in graph x-axis contains class label as 0 and 1 and y-
axis contains number of records found in that class label and now close above graph and
then click on ‘Dataset Preprocessing’ button to read dataset and then normalize dataset
and then replace missing values with 0 and then split dataset into train and test
Fig:Data Prepocessing
Fig:Knn Accuracy
In above screen with KNN we got 97% accuracy and similarly run all algorithms by
clicking button
Fig:Run Algorithms
Fig:Light Gbm
In above screen with extension LightGBM we got 100% accuracy and now click on
‘Comparison Graph’ button to get below output
Fig:Table
In above screen in tabular format we can see accuracy, precision, recall and FSCORE
for each algorithms and we can see its comparison graph also and in all algorithms
extension LIGHTGBM got high accuracy. Now click on ‘Fake Currency Detection from
Test Data’ button to upload fake currency test data and get below output
Fig:Upload TestData
In above screen selecting and uploading ‘testData.csv’ file and then click on ‘Open’
button to get below output
In above screen in square bracket we can see test data and after square bracket we can se
predicted result as ‘Genuine or Fake’
8. CONCLUSION
In this project, SML algorithm SVM, LR, NB, DT, RF and KNN are applied to
the banknote authentication dataset taken from UCI ML repository on three different
train test ratio 80:20, 60:40, 70:30. The dataset contain 1372 and 5 attributes and out of
which 4 are the features and one is the target attribute that have value as genuine bank
currency or forged note. Initially, we have visualized the data by KDE, Box plot and par
plot to study the correlation between the features and the target class see Section III (See
Fig. 1, 2 and 3). From this section it is concluded that all features are important and have
relation with the target class as well as other features, so we have not dropped out any
features. Further in Next section III we have analysed the performance of six SML
algorithms based on the ROC curve and Learning curve on train test ratio 80:20. For the
train test ratio 80:20 Accuracy of KNN is highest i.e., 100 % and NB is having lowest
accuracy i.e., 84% see Fig. 4, 5, 6, 7 and 8. Further in Next section we have analysed the
performance of SML algorithm SVM, LR, NB, DT, RF and KNN on the basis of
standard quantitative analysis parameter like MCC, F1 Score, NPV, NDR, accuracy and
others. For 80:20 and 70:30 train test ratio accuracy is highest in KNN. As MCC value is
near +1 then it is perfect model and F1m is also1 for both train test ratio. And Naïve
Bayes is having the lowest accuracy i.e., 84% in 80:20 and 86% in 70:30 and its MCC is
lowest as well for both the train test ratio. For train test ratio 60:40 highest accuracy is
seen in DT i.e., 100%, MCC value is also +1 that shows that decision tree is performing
better than five SML algorithms. The lowest accuracy is seen in Naïve Bayes. To
visualize the evaluation parameter of SVM, LR, NB, DT, RF and KNN histogram is also
drawn.
9. REFERENCES
[1] M. Aoba, T. Kikuchi, and Y. Takefuji, “Euro Banknote Recognition System Using a
Three-layered Perceptron and RBF Networks”, IPSJ Transactions on Mathematical
Modeling and it's Applications, May 2003.
[5] H. Hassanpour and E. Hallajian, “Using Hidden Markov Models for Feature
Extraction in Paper Currency Recognition.
[6] Z. Huang, H. Chen, C. J. Hsu, W. H. Chen and S. Wuc, “Credit rating analysis with
support vector machines and neural network: a market comparative study”, 2004
. [7] C. Kumar and A. K. Dudyala, “Banknote Authentication using Decision Tree rules
and Machine Learning Techniques”, International Conference on Advances in Computer
Engineering and Applications(ICACEA), 2015.
[8] M. Lee and T. Chang, “Comparison of Support Vector Machine and Back
Propagation Neural Network in Evaluating the Enterprise Financial Distress”,
International Journal of Artificial Intelligence & Applications 1.3 (2010) 31-43.
[11] A. Patle and D. S. Chouhan, “SVM Kernel Functions for Classification”, ICATE
2013.
[17] C.-Y. Yeh, W.-P. Su, and S.-J. Lee, “Employing multiplekernel support vector
machines for counterfeit banknote recognition,” Applied Soft Computing, vol. 11, no. 1,
pp. 1439–1447, Jan. 2011.
Our country's most precious asset is its library and six supervised machine learning
releasing fake notes that seem very identical Support Vector Machine, a Random Forest,
to the genuine thing. There was a glut of a Logical Regression, which is Naïve Bayes,
counterfeit currency in circulation during the Decision Tree, and the k-n Neighbour.
not impossible, for a human to tell the evaluated using metrics like Precision,
difference between a real and counterfeit Precision, Recall, MCC, the F1-S and
note without certain distinguishing features, others. Three different train-test ratios were
as many of the features of both types of taken into consideration, namely 80:20,
notes are similar. Differentiating between 70:30, and 60:40. Some SML algorithms
We suggest a method for recognising Euro Counterfeit notes with high resolution are
banknotes that combines a Radial Basis becoming simpler to fabricate with the
Function, or RBF, network with a three- advancement of digital imaging technology,
layered perceptron. A well-known pattern such as colour scanners and laser printers.
recognition approach and a very efficient The issue of counterfeit money notes affects
tool for identifying banknotes is a three- almost every nation on Earth.There have
layered perceptron. Due to its estimation of been reports of people even getting
the sample data's probability distribution, an counterfeit currency from machine vending
RBF network may reject data that is not machines, ATMs, and even during elections.
legitimate. We validate our model using It is important to develop a system that can
several RBF networks and classify it using a identify fake currency. To identify
three-layered perception. There are two counterfeit banknotes, we provide a system
benefits to the suggested system compared that uses multiple-kernel machine learning
to the one that uses only one the RBF techniques in this article. The system is fed
network. The computation cost remains the brightness histograms of the divisions
constant regardless of the number of classes, that make up each banknote. When merging
and the extraction of features area may be many kernels into a single matrix, linearly
easily determined. Since Euro banknotes balanced combination is used. In order to
exhibit highly noticeable characteristics in make the semi-definite programming (SDP)
infrared (IR) photographs, we therefore approach more efficient, two approaches are
suggest using IR and visible photos as input used. One approach takes the kernel weights
information for the system. Acceptance rates as given and assumes they are non-negative,
for legitimate banknotes and rates of whereas the other sets them all equal to one.
Naïve Bayes using this module. you can normalise the dataset, replace
Fifthly, we will use this module to upload values that are missing with 0, and split it
Run Decision Tree. into a train and test set.
Sixth, we are going to execute SVM with
the help of this module.
7) Execute Random Forest: We will execute
Random Forest using this module.
Step 8: do Logistic Regression This module
will be used to do the logistic regression.
Nineth, Execute the LightGBM Algorithm
Extension: We'll be comparing different
algorithms with the help of this module. The accuracy, precision, recall, and
10) Identifying Fake Currency in Test Data: FSCORE scores for each algorithm are
To do this, we will use this module. shown in tabular style on the above screen.
Additionally, a comparison graph is
provided for each method. Among all the
algorithms tested, LIGHTGBM achieved the
highest accuracy. After that, to acquire the
RESULTS AND DISCUSSION results shown below, submit test data of
false cash by clicking the "Fake Currency
Recognition from Test Data" button.
we can see the anticipated outcome labelled NPV, NDR, accuracy, and more, we have
as "Genuine or Fake." compared the SML algorithms SVM, LR,
NB, DT, RF, and KNN in the following
section. The KNN model has the best
accuracy for 80:20 or 70:30 train-test ratios.
CONCLUSION
Both the train-test ratio and the MCC value
Using three distinct train-test ratios (80:20, are 1 since the model is flawless.
60:40, and 70:30), this article applies the Additionally, for both the train-test ratio,
SML algorithms SVM, LR, NB, DT, RF, Naïve Bayes has the lowest accuracy,
and KNN on the UCI ML repository's coming in at 84% in 80:20 & 86% in 70:30.
banknote authentication dataset. There are Decision tree outperforms five support
1,372 records in the dataset, with 5 vector machine techniques with an MCC
properties totaling; 4 of these are features, value of +1 and a train-test ratio of 60:40,
and 1 is the target attribute, which yielding an accuracy of 100%. Naïve Bayes
determines whether the money is real or has the worst level of accuracy. Also
counterfeit. To begin, as shown in Section included is a histogram for SVM, for
III (Figures 1, 2, and 3), we have visualised example LR, NB, DT, RF, and KNN
the data using KDE, a box plot, and a par assessment parameters.
plot in order to explore the association
REFERENCES
between the characteristics and the target
class. It is clear from this section that all [1] M. Aoba, T. Kikuchi, and Y. Takefuji,
aspects are significant and relate to the “Euro Banknote Recognition System Using
target class and other features; hence, no a Three-layered Perceptron and RBF
features have been excluded. Using the ROC Networks”, IPSJ Transactions on
and learning curves with a train-test ratio of Mathematical Modeling and it's
80:20, we examined the efficacy of six the Applications, May 2003.
SML algorithms in Section III. With an
[2] S. Desai, S. Kabade, A. Bakshi, A.
80:20 train-to-test ratio, The precision of
Gunjal, M. Yeole, “Implementation of
KNN is 100%, whereas that of NB is 84%
Multiple Kernel Support Vector Machine for
Using industry-standard quantitative
Automatic Recognition and Classification of
analysis metrics such as MCC, F1 rating,
Counterfeit Notes”, International Journal of Intelligence & Applications 1.3 (2010) 31-
Scientific & Engineering Research, October- 43
2014
. [9] C. Nastoulis, A. Leros, and N. Bardis,
[3] C. Gigliaranoa, S. Figini, P. Muliere, “Banknote Recognition Based On
“Making classifier performance Probabilistic Neural Network Models”,
comparisons when ROC curves intersect”, Proceedings of the 10th WSEAS
Computational Statistics and Data Analysis International Conference on SYSTEMS,
77 (2014) 300–312 Vouliagmeni, Athens, Greece, July 10-12,
2006.
[4] E. Gillich and V. Lohweg, “Banknote
Authentication”, 2014. [10] S. Omatu, M. Yoshioka and Y.
Kosaka, “Bankcurrency Classification Using
[5] H. Hassanpour and E. Hallajian, “Using
Neural Networks”, IEEE, 2007.
Hidden Markov Models for Feature
Extraction in Paper Currency Recognition. [11] A. Patle and D. S. Chouhan, “SVM
Kernel Functions for Classification”, ICATE
[6] Z. Huang, H. Chen, C. J. Hsu, W. H.
2013.
Chen and S. Wuc, “Credit rating analysis
with support vector machines and neural [12] E. L. Prime and D. H. Solomon,
network: a market comparative study”, 2004 “Australia’s plastic banknotes: fighting
counterfeit currency.,” Angewandte Chemie
. [7] C. Kumar and A. K. Dudyala,
(International ed. in English), vol. 49, no.
“Banknote Authentication using Decision
22, pp. 3726–36, May 2010.
Tree rules and Machine Learning
Techniques”, International Conference on [13] A. Roy, B. Halder, and U. Garain,
Advances in Computer Engineering and “Authentication of currency notes through
Applications(ICACEA), 2015. printing technique verification,” Proceedings
of the Seventh Indian Conference on
[8] M. Lee and T. Chang, “Comparison of
Computer Vision, Graphics and Image
Support Vector Machine and Back
Processing -ICVGIP ’10, pp. 383–390,
Propagation Neural Network in Evaluating
2010.
the Enterprise Financial Distress”,
International Journal of Artificial
Evaluation of Machine Learning Algorithms for the Detection of Fake Bank Currency
Authored By