Clustering based credit card fraud detection system using validation method with modified K-mean
Author(s):
Shweta
Keywords:
credit card fraud, centroids, clustering,k-mean
Abstract
As the credit card has become the most prevailing mode of payment for both offline as well as online mode of transaction. This leads to acceleration of frauds. Fraud is a million dollar business which is rising rapidly. Thus there is a need of fraud avoidance system. Fraudster many a time try to generate credit card numbers by just altering the digits. Clustering can be done to effectively implement such systems. The K-mean is one of the most popular clustering algorithm used in data mining for real world applications, especially for the numerical dataset. K-mean performance and efficiency is greatly affected by the initial cluster centers, as different initial centers may leads to different cluster formations. In this research paper, K-mean is used in application with a modification which has an additional step for selecting better cluster centroids. This is done by keeping intercluster similarity as low while intracluster similarity as high.
Article Details
Unique Paper ID: 144414
Publication Volume & Issue: Volume 3, Issue 11
Page(s): 99 - 102
Article Preview & Download
Share This Article
Join our RMS
Conference Alert
NCSEM 2024
National Conference on Sustainable Engineering and Management - 2024