First Phase Project Report
On
“Credit Card Fraud Detection Using Machine Learning”
Submitted in partial fulfilment of the requirements to
Malla Reddy University
For the Award of Bachelor of Technology In
Computer Science & Engineering
Submitted By
[Link] TIWARI 2011CS010145
[Link] REDDY 2011CS010099
M. SAI SHIVADEEP 2011CS010166
[Link] RAO 2011CS010113
Under the Guidance of Dr. M. RAVIKANTH,Ph.D(CSE),
JNTUH,LMISTE, Associalte Professor, Department of CSE
TELANGANA STATE PRIVATE UNIVERSITY) MAISAMMAGUDA,KOMPALLY,HYDERABAD-500100
INTRODUCTION
A s w e a re m o v i n g to wa rd s t h e d i g i ta l w o r l d —
c y b e rs e c u r i t y i s b e co m i n g a c r u c i a l p a r t o f o u r l i fe .
W h e n w e ta l k a b o u t s e c u r i t y i n d i g i ta l l i fe t h e n t h e
m a i n c h a l l e n g e i s to fi n d t h e a b n o r m a l a c ti v i t y.
W h e n w e m a ke a ny t ra n s a c ti o n w h i l e p u rc h a s i n g
a ny p ro d u c t o n l i n e — a g o o d a m o u nt o f p e o p l e
p refe r c re d i t ca rd s . T h e c re d i t l i m i t i n c re d i t ca rd s
s o m e ti m e s h e l p s u s m e m a k i n g p u rc h a s e s e ve n i f
w e d o n ’ t h ave t h e a m o u nt at t h at ti m e . B u t , o n t h e
o t h e r h a n d , t h e s e fe at u re s a re m i s u s e d b y c y b e r
att a c ke rs .
To d ay, w e h ave m a ny m a c h i n e l e a r n i n g a l g o r i t h m s
t h at ca n h e l p u s c l a s s i f y a b n o r m a l t ra n s a c ti o n s .
T h e o n l y re q u i re m e nt i s t h e p a st d ata a n d t h e
s u i ta b l e a l g o r i t h m t h at ca n fi t o u r d ata i n a b e tt e r
fo r m .
OUR APPROACH
Our Aim Is To Create A Machine Learning Model
Which Uses The Past Data Of Credit Card
Transacti ons Of The Year 2013 Of Europian Card
Holders.
HOW CAN THIS BE HELPFUL ?
• From the moment the e-commerce payment systems came to
existence, there have always been people who will find new
ways to access someone’s finances illegally. This has become a
major problem in the modern era, as all transactions can easily
be completed online by only entering your credit card
information. Even in the 2010s, many American retail website
users were the victims of online transaction fraud right before
two-step verification was used for shopping online.
Organizations, consumers, banks, and merchants are put at risk
when a data breach leads to monetary theft and ultimately the
loss of customers’ loyalty along with the company’s reputation.
• Unauthorized card operations hit an astonishing amount of 16.7
million victims in 2017. Additionally, as reported by the Federal
Trade Commission (FTC), the number of credit card fraud claims
in 2017 was 40% higher than the previous year’s number. There
were around 13,000 reported cases in California and 8,000 in
Florida, which are the largest states per capita for such type of
crime. The amount of money at stake will exceed approximately
$30 billion by 2020. Here are some credit card fraud statistics:
HOW CAN WE SOLVE THIS ?
• Machine learning models can recognise unusual credit card
transactions and fraud. The first and foremost step involves
collecting and sorting raw data, which is then used to train the
model to predict the probability of fraud
• Classifying whether credit card transactions are authentic or
fraudulent using algorithms such as logistic regression, random
forests, support vector machines (SVMs), deep neural networks
along with autoencoders, long short-term memory (LSTM)
networks, and convolutional neural networks (CNNs)
• Predicting whether it is the cardholders or the fraudsters using the
credit cards through credit card profiling
• Using outlier detection methods to identify considerably different
transactions (or 'outliers') from regular credit cards transactions to
detect credit card fraud.
MACHINE LEARNING MODELS IN OUR PROJECT
1. DECISION TREE CLASSIFIER
2. K-NEAREST NEIGHBORS
3. LOGISTIC REGRESSION
4. SUPPORT VECTOR MACHINES
5. RANDOM FOREST CLASSIFIER
THE PROCEDURE…
1. DATA COLLECTION
2. IMPORTING MODULES AND
LIBRARIES
3. EXPLORATORY DATA ANALYSIS
4. TRAINING AND TESTING THE
DATASET
5. APPLYING ALGORITHMS
6. SELECTING THE BEST MODEL
BASED ON ACCURACY AND F1-
SCORE
MODEL ACCURACIES
• DecisionTreeClassifier=99.91%
• K-Nearest Neighbors=99.92%
• Logistic Regression=99.89%
• Support Vector Machines=99.93%
• Random Forest Classifier=99.91%
F1-SCORES
• DecisionTreeClassifier=0.75
• K-Nearest Neighbors=0.79
• Logistic Regression=0.66
• Support Vector Machines=0.78
• Random Forest Classifier=0.73
WHICH ONE IS TO BE CONSIDERED
• Accuracy: One of the more obvious metrics, it is the measure of all the
correctly identified cases. It is most used when all the classes are equally
important.
• F1-score: This is the harmonic mean of Precision and Recall and gives a better
measure of the incorrectly classified cases than the Accuracy Metric.
CONCLUSION
• 5 MODELS THAT WE CAME WITH WERE UP TO THE MARK
• BY SEEING AT THE F1-SCORE WE CAN ELIMINATE LOGISTIC
REGRESSION AND CAN CONSIDER KNN MODEL AS THE BEST SUITED
MODEL FOR OUR PROJECT
• THE REASON WHY KNN WAS EFFICIENT BECAUSE THE DATA HAS TO
LEARN THE METRICS FROM ITS NEIGHBORS AND THE FRAUDULENT
TRANSACTION WILL BE DIFFERENT FROM THOSE NEAR IT
• SO THE NEGATIVE VALUE CAN BE EASILY PREDICTED AND THUS KNN
GIVES MORE ACCURACY
• BY GIVING INPUT VALUES WE CAN PREDICT WHETHER THE
TRANSACTION IS FRAUD OR NORMAL
THE TEAM
~KISHAN TIWARI
~MANISH REDDY
~SHIVADEEP
~MANISH RAO