Scribd
Upload
63 views

Credit Card Fraud Detection

This document proposes a fraud detection system (FDS) to detect fraudulent credit card transactions. The FDS uses a rule-based filter and Dempster-Shafer theory to combine evidence from different rules. It also includes a transaction history database and Bayesian learner. The system aims to automatically detect if a transaction is genuine or fraudulent to prevent credit card abuse and theft.

Uploaded by

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

Credit Card Fraud Detection

This document proposes a fraud detection system (FDS) to detect fraudulent credit card transactions. The FDS uses a rule-based filter and Dempster-Shafer theory to combine evidence from different rules. It also includes a transaction history database and Bayesian learner. The system aims to automatically detect if a transaction is genuine or fraudulent to prevent credit card abuse and theft.

Uploaded by

Armin Aqelpoor
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 10

Credit card FDS

A fusion approach using Dempster–Shafer theory and Bayesian learning

Vahid Taheri – Armin Aghel poor


AAI981 – Winter 2019

1 Credit card FDS | A fusion approach using Dempster–Shafer theory and Bayesian learning
What is FDS?

FDS is an evidence-based detection technology to automatically

verify if a file or image (also data) is genuine or fraudulent based on

scientific methodologies.
“https://bullwall.com/solutions/fds/”

2 Credit card FDS | A fusion approach using Dempster–Shafer theory and Bayesian learning
What is the problem?
Imagine, your credit card being stolen or lost and somebody that

knows your password, use it !

Now, if your bank can detect that this transaction isn’t genuine , it

can automatically prevent abuse of using it more.

3 Credit card FDS | A fusion approach using Dempster–Shafer theory and Bayesian learning
What is the solution?

We can build a DFS to solve this problem !

4 Credit card FDS | A fusion approach using Dempster–Shafer theory and Bayesian learning
FDS Components

1 Rule Based Filter Dempster–Shafer adder 2

3 Transaction History Database Bayesian learner 4

5 Credit card FDS | A fusion approach using Dempster–Shafer theory and Bayesian learning
FDS Components
1 Rule Based Filter

R1 : Address mismatch

R2 : Outlier detection using ɛ-neighbor-hood clustering

h : fraud

~h : non-fraud

U : suspicious

6 Credit card FDS | A fusion approach using Dempster–Shafer theory and Bayesian learning
FDS Components
2 Dempster–Shafer adder

Combine evidences from two rules.

7 Credit card FDS | A fusion approach using Dempster–Shafer theory and Bayesian learning
FDS Components
3 Transaction History Database

Some Data are used to construct R2 rules.

8 Credit card FDS | A fusion approach using Dempster–Shafer theory and Bayesian learning
FDS Components
4 Bayesian learner
Relation between evidences and predicts.

9 Credit card FDS | A fusion approach using Dempster–Shafer theory and Bayesian learning
attention

you
Thank
for

10 Credit card FDS | A fusion approach using Dempster–Shafer theory and Bayesian learning

You might also like