Reduce fraud risk Using CNN and LSTM and Time Series Data in E-Commerce
General Material Designation
Dissertation
First Statement of Responsibility
Hayder Al Jassim
.PUBLICATION, DISTRIBUTION, ETC
Name of Publisher, Distributor, etc.
Electrical and Computer Engineering
Date of Publication, Distribution, etc.
1401
PHYSICAL DESCRIPTION
Specific Material Designation and Extent of Item
87p.
Other Physical Details
cd
DISSERTATION (THESIS) NOTE
Dissertation or thesis details and type of degree
M.S.
Discipline of degree
Software Engineering Field of Data Mining
Date of degree
1401/07/19
SUMMARY OR ABSTRACT
Text of Note
Fraud and forgery in bank cards or financial transactions is the illegal activity of a person or people in a bank account that does not belong to them. If a person illegally uses another person's bank account, he has committed credit card fraud. Today, due to the growth of the Internet of Things and virtual banking, such frauds have increased significantly. Artificial intelligence and data mining should be used to identify these frauds with high accuracy. In this thesis, using deep learning, convolutional neural networks, and LSTM, we try to identify fraud in bank transactions with high accuracy. CNNs have a good ability to find spatial features, and in contrast, LSTM can find temporal relationships between data well. The combination of these two deep networks can lead to a very accurate bank card fraud detection system.
OTHER VARIANT TITLES
Variant Title
کاهش خطر تقلب با استفاده از CNN و LSTM و داده های سری زمانی در تجارت الکترونیک
UNCONTROLLED SUBJECT TERMS
Subject Term
Internet of Things (IoT), banks, fraud monitoring, (CNN), deep learning.