International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056

Volume: 08 Issue: 04 | Apr 2021 www.irjet.net p-ISSN: 2395-0072

Online Fraud Detection on E-Commerce

Ashitosh Gole1, Nitesh Kharatmol2, Prof.Sonali Patil3
1,2
Student, Dept. of Computer Engineering, M.G.M. College of Engineering and Technology, Kamothe,
Maharashtra, India
3
Prof. Dept. of Computer Engineering, M.G.M College of Engineering and Technology, Kamothe,
Maharashtra, India
----------------------------------------------------------------------***---------------------------------------------------------------------
Abstract - The frequency of online transaction has raised can include usernames, password, credit card details, social
significant in last some of the years due to popularization of e- security numbers, etc. Often times the e-mails directly
commerce. We also noticed the significantly increasing the connect you to a website where you can update your
online fraud cases, resulting in billions of dollars losses each personal information. Because these sites often look
year worldwide. Hence it is important and necessary to “Official” they hope you’ll be tricked into disclosing valuable
developed and apply techniques that can assist in fraud information that you normally would not reveal. This often
detection. Which motivate our research. This work aims to results is identify the theft and financial loss. Spyware and
apply and evaluate the computational intelligence Viruses are both malicious programs that are loaded on your
techniques to identify and detect the fraud and make more computer without your knowledge. The purpose of these
secure web payment gateway and another online payment programs uses to capture or destroy information, to ruin
system. In order to evaluate the techniques, we apply and computer performance or to overload you with advertising.
evaluate them in an actual data set of the most popular Viruses can spread by infecting computers and then it
Brazilian Electronic payment System. Our project shows good replicates. Spyware disguises itself as a legitimate
performance in fraud detection and it helps to gain 43% of application and embeds itself into your computer where it
economics matric. they monitor your activity and collects information.

Key Words: E-Commerce, Online payment, E- 1.2 Types of Online Fraud

Business, dataset, payment gateway.

1. INTRODUCTION
Today due to rapid growth online transaction is grown
day by day. The mode of payment is done by credit card or
online payment method. The also credit card and Gpay users
are increasing day by day. It was reported that there are
almost 430 million credit and debit card users across whole
Europe. As the number of credit/debit card/Gpay users
increasing, the fraudulent users are also increasing.

The user has to inform the details while making

payment. In this type, if fraudulent user wants to access
his/her card or any of bank related details then he just
needs to steal that details. In virtual card, the fraudulent
user needs to know the information about details
information about credit card such as, CVV no, Secure code,
Credit card number. Therefore, the secure payment
gateway is needed to identify the user and to verify that the Fig -1: Types of Fraud
user is legal or attacker. The most useful and appropriate
technique used for fraud detection is Behavior and Location
Online Payment Account Generation
Analysis (BLA).

1.1 Online Fraud There was open to online account they need the Users
credit card details to open the account, need to enter the CVV,
Types of online fraud are called phishing and spoofing. Card number and most important thing have an OTP (one
Phishing is the process of collecting your personal time password).
information through e-mails or websites. This information

© 2021, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 3671
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 08 Issue: 04 | Apr 2021 www.irjet.net p-ISSN: 2395-0072

1.3 RANDOM FOREST In proposed system, while registration we take required

information from user which is efficient to detect fraudulent
Random forest algorithm is useful for detect the user activity,
unwanted activity faster and it help to sort out the thing in • In proposed system, I present a Behavior and
the object efficiently to help the things. the method is useful Location Analysis (BLA).
for find unwanted or suspicious activity in multiple class • Which does not require fraud signatures and yet is
object. In this it detects the one fraudulent user from many of able to detect frauds by considering a cardholder’s spending
the user. habit.
• Card transaction processing sequence by the
2. SCOPE and OBJECTIVE stochastic process of a BLA.
• The details of items purchased Individual transactions
The system prevents on fraudulent users from misusing
are usually not known to any Fraud Detection System (FDS)
the details of the credit-card of the genuine users for their
running at the bank that issues credit cards to the
personal gain. The spending habits of the owner is detecting
cardholders.
the fraud. As the fake user might not be aware of the
• Hence, I feel that BLA is an ideal choice for addressing
spending habits of the owner, there will be an irregularity in this problem.
spending pattern, which the system will detect. The owner is • Another important advantage of the BLA - based
immediately alerted about the attempted fraud and the approach is a drastic reduction in the number of Fail Pass
transaction is blocked performed by attacker. Thus, the transactions identified as malicious by an FDS although they
system protects legitimate users from financial loss. The are actually genuine.
system helps making electronic payment safer and more • An FDS runs at a credit card issuing bank. Each incoming
reliable. The principles in the proposed system can also transaction is submitted toward the FDS for verification.
adopted and implemented in other electronic payment • FDS receives the card details and the value of purchase to
services such as online banking facility and payment verify, whether the transaction is genuine or fake.
gateways. • The types of goods that are bought in that transaction are
 Creating an application to detect fraud Credit Cards. not known to the FDS.
 Implementing Hidden Markov model. • It tries to find any anomaly in the transaction based on
 Creating database containing all relevant information of the spending profile of the cardholder, shipping address, and
Customer. billing address, etc.
 Providing security to the customers at the time of • If the FDS confirms the transaction to be of fraud, it raises
transaction. an alarm, and the issuing bank declines the transaction.
The credit card fraud detection feature use user behavior
3. EXISTING SYSTEM and location scanning to check for unusual patterns. These
patterns include user characteristics such as users spending
Credit card frauds are increasing rapidly nowadays
patterns as well as usual user geographic locations to verify
because of fraud financial loss is increasing drastically.
his identity. If any unusual pattern is detected, the system
Billions of amounts is lost every year due to fraud. To analyze
requires re-verification of the user.
the fraud there is lack of research. Many machine learning The system analyzes user credit card data for various
algorithms are implemented to detect credit card fraud. characteristics. These characteristics include user country,
Logistic Regression, Decision Tree Model, Artificial Neural usual spending procedures. Based upon previous data of
Network, Gradient Boosting Algorithm and Hybrid user the system recognizes unusual patterns in the payment
algorithms are applied. The objective of the project is to procedure. So now the system requires the user to login
detect credit card fraud detection by implementing machine again or even block the user for more than 3 invalid
learning algorithm with relevance to time and amount of attempts.
transaction.
4. PROPOSED SYSTEM 5. ALGORITHM
The aim of proposed system is to develop a website Step1: - Initial state
which has capability to restrict and block the transaction This state is indicating the first stage of payment process for
performing by attacker from genuine user’s credit card e.g. of opening payment gateway window
details. The system is developed for the transactions higher Step 2: - If find out some suspicious activity else go to step 5
than the customers current transaction limit. As we seen If user cross its payment limit or it may enter incorrect
existing system detects the fraud after fraud has been details then it calls for the security check. The dataset
occurred i.e., based on customers complained. Proposed consists of credit card transactions whose features are the
system tries to detect fraudulent transaction before product of PCA analysis and thus we don’t know what they
transaction succeed. represent except from the ‘Amount’, ‘Time’ and ‘Class’ ones.
‘Amount’ is the price of each transaction, ‘Time’ is the
<<seconds elapsed between each transaction and the first

© 2021, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 3672
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 08 Issue: 04 | Apr 2021 www.irjet.net p-ISSN: 2395-0072

transaction>> and ‘Class’ represents a fraud 7. PROJECT IMPLEMENTATION AND

transaction when its value equals 1 and a valid TEHCNOLOGIES
transaction when its value equals 0.
Step 3: - It ask the ask the user for security question. The Project is designed and developed in Django
Those question and answer accept at the login activity Framework. We used Django Framework for coding of the
Step 4: - If user fails to answer correctly then it informs or project. Created and maintained all databases into MySQL
alert the bank system about the activity happens. Server, in that we create tables, write query for store data or
Step 5:- else it corrects then it goes and complete the record of project.
transaction
 Hardware Requirement
 Processor –Core i3
 Hard Disk – 160 GB
 Memory – 1GB RAM
 Monitor
 Software Requirement
 Windows 7 or higher
 Python
 Django framework
 Database
 MySQL

8. CONCLUSION

Credit card fraud is a criminal dishonesty. In this paper

describe different types of fraud, such as bankruptcy fraud,
counterfeit fraud, theft fraud, application fraud and
behavioral fraud, and discussed measures to detect them. It
Fig -2: Flow of Fraud Detection has included pair-wise matching, decision trees, clustering
techniques, neural networks, and genetic algorithms. Yet, the
unprofessional fraudster is unlikely to operate on the scale of
6. WORK FLOW the professional frauds ter and the costs to the bank of their
detection may be uneconomic. The bank would be faced with
an ethical dilemma. Should they need try to detect such
fraudulent cases or should they act in shareholder interests
and avoid uneconomic costs? The next step in research
program, the focus will be upon the implementation of a
‘suspicious’ scorecard on a real data-set and its evaluation.
The main tasks is to build scoring models to predict
fraudulent behavior, taking into account the fields of behavior
that relate to different types of fraud identified in this paper,
and to evaluate the associated ethical implications. The plan
is to take one of the European countries, probably Germany,
and then to extend the research to other EU countries.
ACKNOWLEDGEMENT

Working on the project on Online Fraud Detection was a

source of immense knowledge to us. However, this would
not have been possible without the equal support of all
members. I sincerely express my deep sense of gratitude to
our Asst.Prof. Sonali patil, for her valuable guidance,
Fig -3: Workflow continuous encouragement and support whenever required
and timely help given to us throughout the course of this
work

© 2021, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 3673
 International Research Journal of Engineering and Technology (IRJET) e-ISSN: 2395-0056
Volume: 08 Issue: 04 | Apr 2021 www.irjet.net p-ISSN: 2395-0072

REFERENCES
[1] CLIFTON PHUA1, VINCENT LEE1, KATE SMITH1 & ROSS
GAYLER2 “A Comprehensive Survey of Data Mining-
based Fraud Detection Research” published by School of
Business Systems, Faculty of Information Technology,
Monash University,

[2] “Survey Paper on Credit Card Fraud Detection by

Suman”, Research Scholar, GJUS&T Hisar HCE, Sonepat
published by International Journal of Advanced
Research in Computer Engineering & Technology
(IJARCET) Volume 3 Issue 3, March 2014

[3] “Research on Credit Card Fraud Detection Model Based

on Distance Sum –by Wen-Fang YU and Na Wang”
published by 2009 International Joint Conference on
Artificial Intelligence

[4] “Credit Card Fraud Detection through Parenclitic

Network Analysis-By Massimiliano Zanin, Miguel
Romance, Regino Criado, and SantiagoMoral”
published by Hindawi Complexity Volume 2018,

[5] “Credit Card Fraud Detection: A Realistic Modelling and a

Novel Learning Strategy” published by IEEE
TRANSACTIONS ON NEURAL NETWORKS AND LEARNING
SYSTEMS, VOL. 29, NO. 8, AUGUST 2018

[6] “Credit Card Fraud Detection Using Machine Learning

Models and Collating Machine Learning Models by
Navanshu Khare and Saad Yunus Sait” published by
International Journal of Pure and Applied Mathematics,
Volume 118 No. 20, 2018.

© 2021, IRJET | Impact Factor value: 7.529 | ISO 9001:2008 Certified Journal | Page 3674