@cscjournals

Online Transaction Fraud Detection using Hidden Markov Model & Behavior Analysis

N. Patel, Y. Li, and A. Hadaegh. International Journal of Computer Science and Security (IJCSS) 15 (3): 59-72 (June 2021)

Abstract

Card payment are mostly preferred by many for transactions instead of cash. Due to its convenience, it is the most accepted payment method for offline as well as online purchases, irrespective of region or country the purchase is made. Currently, cards are used for everyday activities, such as online shopping, bill pays, subscriptions, etc. Consequently, there are more chances of fraudulent transactions. Online transactions are the prime target as it does not require real card, only card details are enough and can be stored digitally. The current system detects the fraud transaction after the transaction is completed. Proposed system in this paper, uses Hidden Markov Model (HMM), which is one of the statistical stochastic models used to model randomly changing systems. Using Hidden Markov Model, a fraud transaction can be detected during the time of transaction itself and after 3 attempts of verification card can blocked at the same time. Behavior Analysis (BA) helps to understand the spending habits of cardholder. Hidden Markov Model helps to acquire high-level fraud analysis with a low false alarm ratio.

Links and resources

URL:
https://www.cscjournals.org/library/manuscriptinfo.php?mc=IJCSS-1610
BibTeX key:
patel2021online
search on:

Comments and Reviews  
(0)

There is no review or comment yet. You can write one!

Tags


Cite this publication