Abstract— we collect user’s orders, like purchased and

Abstract—

Due
to the advent of Internet technologies, Ecommerce widely adapted mode of
business in modern times. Fraudulent activity exits in many areas of businesses
and our daily life. Thousands of dollars are loss every day due to the fraud in
purchased items and immediately cancelled orders without any proper reason. In
existing there is lot work in E-commerce fraud detection in credit card fraud
detection, network intrusion etc. But limited work in E-commerce fraud
detection for user orders behavior. Now in this paper we are proposing a
concept of E-commerce fraud detection in user orders behavior. For this we
collect user’s orders, like purchased and cancelled data. For every user data
we will compare with email, mobile, IP address and address for each
transactions. Based on that data we can effect to the users that hide the COD
option, and block the user account. To design potent and efficient fraud
detection concept is the key for reducing the losses in transactions

                 Keywords-Fraud,E-Commerces.

 

            I.INTRODUCTION

E-commerce
payment systems have become popular due to widespread use of the internet-based
shopping and banking. Rapid increment of this era billions of dollars are lost
every year due to credit card fraud. Fraud is an act of betrayal intended for
personal usage or to harm a loss to someone. Fraudster only wants to know the
personal information related to card (card number, card expiry date etc.). It
can be possible physically or virtually. It is commonly understood as
dishonesty to gain some advantage which is often financial, over another
person.

 

The volume of electronic
transactions has raised signi?cantly in last years, mainly due to the popularization
of electronic commerce (e-commerce), such as online retailers like Amazon,
eBay, etc. We also observe a signi?cant increase in the number of fraud cases,
resulting in billions of dollars losses each year worldwide. Therefore, it is
important and necessary to developed and apply techniques that can assist in
fraud

 

 

detection and prevention, which motivates our
research. This project aims to apply and evaluate computational techniques to
identify fraud in electronic transactions.

 

                                    II. LITERATURE SURVEY

 

Ghosh and Reilly 1 have proposed credit card fraud
detection with a neural network. They have built a detection system, which is
trained on a large sample of labeled credit card account transactions. These
transactions contain example fraud cases due to lost cards, stolen cards,
application fraud, counterfeit fraud, mail-order fraud, and non-received issue
(NRI) fraud. Recently, Syeda et al. 2 have used parallel granular neural
networks (PGNNs) for improving the speed of data mining and knowledge discovery
process in credit card fraud detection. A complete system has been implemented
for this purpose. Stolfo et al. 3 suggest a credit card fraud detection
system (FDS) using meta learning techniques to learn models of fraudulent
credit card transactions. Meta learning is a general strategy that provides a
means for combining and integrating a number of separately built classifiers or
models. A meta classifier is thus trained on the correlation of the predictions
of the base classifiers. The same group has also worked on a cost-based model
for fraud and intrusion detection. They use Java agents for Meta learning
(JAM), which is a distributed data mining system for credit card fraud
detection A number of important performance metrics like True Positive—False
Positive (TP-FP) spread and accuracy have been defined by them. Alekerov et al.
4 present CARDWATCH, a database mining system used for credit card fraud
detection. The system, based on a neural learning module, provides an interface
to a variety of commercial databases. Kim and Kim 5 have identified skewed
distribution of data and mix of legitimate and fraudulent transactions as the
two main reasons for the complexity of credit card fraud detection. Based on
this observation, they use fraud density of real transaction data as a
confidence value and generate the weighted fraud score to reduce the number of
misdetections.

        Fan et al. 6
suggest the application of distributed data mining in credit card fraud
detection. Brauset al. 7 have developed an approach that involves advanced
data mining techniques and neural network algorithms to obtain high fraud
coverage. Chiu and Tsai 8 have proposed Web services and data mining
techniques to establish a collaborative scheme for fraud detection in the
banking industry. With this scheme, participating banks share knowledge about
the fraud patterns in a heterogeneous and distributed environment. To establish
a smooth channel of data exchange, Web services techniques such as XML, SOAP,
and WSDL are used.

 

 

III. PROPOSED METHODOLOGY

 

1.       The
unethical actions of the dishonest customers is becoming a major concern among
all the online sellers.

2.      
Our proposed system there
by provides a practical solution to avoid the buyer fraud with the help of
history of their purchased and cancelled products.

·        
So for the first time
user can’t get COD option, even though it is repeated for the second time also,
account will block.

·        
If any user is fraud,
again he is trying to register with new account we will compare with email,
mobile, IP address and address. If it is matched he won’t get COD.

·        
And seller has one
proposed option that if he block any user application won’t get give that
seller products to that user

                   

                                 Fig. Architectural
Design

 

 

                                        

 

                                   IV. CONCLUSIONS

For the fraud detection in
e-commerce various approaches are there. In this paper, we have reviewed some
of the detection approaches. Each approach having its own rule sets to
implement and rules are not clearly described in approach. Based on observation
table we can conclude that Hybridization of BLAST-SSAHA approach is best
suitable for the fraud detection in terms of cost and accuracy. To detect a
fraud is necessity but also to decrease false alarm is also necessary.
BLAST-SSAHA?s True positive (TP) ratio having less than Dempster Shafer theory
and Fuzzy Darwinian detection but cost of both approaches is quite expensive.
So it would not beneficial to implement both together. So by implementing rules
which are used in another approach or implementing advanced rule sets into the
BLAST-SSAHA. So there is possibility to increase True Positive result and
decrease false alarm.

                             V.ACKNOWLEDGMENT

             It gives us great pleasure in
presenting this project report titled “Online blood Connect” and we wish to
express our immense gratitude to the people who provided invaluable knowledge
and support in the completion of this project. Their guidance and motivation
has helped in making this project a great success. We express our gratitude to
our project guide Ass.prof.B.Pravallika, who provided us with all the guidance and
encouragement throughout the project development.  We would also like to express our sincere
gratitude to the respective Project Incharge Ass.prof.k.Lakshmi Narayanamma for
providing us the needed assistance, detailed suggestions and also encouragement
to do the project.. We are eager and glad to express our gratitude to the Head
of the Information Technology Dept. Prof. Dr .K. Srinivasa Reddy, for his
approval of this project. We would like to deeply express our sincere gratitude
to our respected principal Prof. Dr.L.V.N.Prasad and the management of
Institute of Aeronautical Engineering for providing such an ideal atmosphere to
build up this project with well-equipped   
library with all the utmost necessary reference materials and up to date
IT Laboratories. We are extremely thankful to all staff and the management of
the college for providing us all the facilities and resources required. 

Written by