Scribd
Upload
189 views

Online Credit Card Fraud Detection Using Big Data: A Project Review On

This document provides an overview of a project to develop an online credit card fraud detection system using big data technologies. The proposed system would use a hybrid framework with neural networks and implement algorithms like HMM to analyze transaction data in real-time and detect fraudulent transactions. It discusses the background of credit card fraud, literature reviews previous related work, and provides details of the proposed system architecture including use case diagrams, algorithms, and planned outcomes like improved accuracy and efficiency through leveraging big data.

Uploaded by

shubhangi wable
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
189 views

Online Credit Card Fraud Detection Using Big Data: A Project Review On

This document provides an overview of a project to develop an online credit card fraud detection system using big data technologies. The proposed system would use a hybrid framework with neural networks and implement algorithms like HMM to analyze transaction data in real-time and detect fraudulent transactions. It discusses the background of credit card fraud, literature reviews previous related work, and provides details of the proposed system architecture including use case diagrams, algorithms, and planned outcomes like improved accuracy and efficiency through leveraging big data.

Uploaded by

shubhangi wable
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 16

Vidya Pratishthan’s Kamalnayan Bajaj Institute of Engineering and Technology

A project review on,


ONLINE CREDIT CARD FRAUD
DETECTION USING BIG DATA

Head Of Department Guided By


Prof. D. A. Zende Prof. D. S. Gaikwad
Agenda :

 Introduction  Algorithms

 Problem statement  Proposed system

 Abstract  Proposed Outcomes

 Literature survey  Project plan 2.0

 Use case Diagram  Bibliography

 Sequence Diagram
Introduction :

 Credit cards are usually used as a virtual card.


 Increase in online trading could attract criminal activities.
 Attackers only need to obtain basic information about cardholder.
 3 major contributions are made :
 Improve accuracy.

 Four Layer framework.

 Use of Big Data technology.


Problem Statement :

An hybrid structure framework using neural network that enables to


improve accuracy and efficiency which handles large amount of
data and detecting real time fraud while transacting credit card.
Abstract :

 Credit card fraud is causing billions of dollars in losses.

 Credit card are widely used in day to day life.

 frauds related with it are also growing.

 HMM is used to find fraudulent transaction.

 Main outcome is find fraudulent transaction before


completion of transaction.
Literature survey 1.1 :

TECHNIQUES/
AUTHORS YEAR RESULTS
ALGORITHMS

You Dai, Jin Yan, 2016 Big Data Technologies Credit Card Fraud
Xiaoxin Tang Detection System

Raghvendra Patidar, 2013 Neural Network Neural network-based


Lokesh Sharma pattern recognition

Avinash Ingole, 2013 Clustering Algorithm Fraud Detect using


Dr.R.C. Thool spending profile
Literature survey 1.2 :

TECHNIQUES/
AUTHORS YEAR RESULTS
ALGORITHMS

Suvasini 2009 Dempster-Shafer Fraud detect by applying


Panigrahi,Amlan Theory and Bayesian several test cases.
Kundu
theory
Dr. R. Dhanapal 2012 Decision Tree Fraud detect by using
Tracing Email and IP

Prashant Gupta, 2016 HMM Secures transaction


Afsari Raut, using OTP via SMS.
Farhan Saudagar
Use Case Diagram :
Sequence Diagram :
Algorithm 1.1 :`

Algorithm 1:Mapper function for key aggregation on Hadoop.


Input : tji //the jth transaction of card i
Output : <i, tji> //card ID and its transaction
i=SplitCardID(tji)
<i, tji> = generate KV pair(i, tji)
Emit <i, tji> .

Algorithm 2:Reducer function for model training on Hadoop.


Input : list (<i, tji> ) //list of all transaction of card i.
Output : <i,model i > //trained model of card I
for <i, tji> in list (<i, tji> ) do
listi.add( tji)
end for
modeli = train(list i)
<i,model i >= generate KV pair(i,model i )
Emit <i,model i > .
Algorithm 1.2 :`

Algorithm 3: RDD function for Model training on spark .


Input : T // transaction data.
Output : TMP // train model parameters
inputRDD=read (T).cache()
initialize model parameters in array TMP
while convergence condition is not satisfied do
tmpParamsRDD=inputRDD.map(TMP)
TMP=tmpParamsRDD.reduce()
end while
save params.
Proposed System :`

T
Proposed Outcomes :`

• Accuracy and Efficiency.


Project Plan 2.0 :`

 Gantt Chart
System Implementation
Bibliography :`

[1]Online Credit Card Fraud Detection: A Hybrid Framework with Big


Data Technologies.
[2]Credit card fraud detection system : A Survey
-by Dinesh Talekar and K.P. Adhiya.
[3]Survey paper on credit card fraud detection.
-by Suman and Mitali Bansal.
[4]ATM card fraud detection using Hidden Markov Model.
-by Sonawane V.D. ,Prashant Gupta ,Afsari Raut,Farhan Saudagar.

You might also like