International Peer-Reviewed Open Access Journal ISSN (Online): 2395-5325
Author Login Track Manuscript Contact
IJCRCST Logo

International Journal of Contemporary Research in Computer Science and Technology

Peer Reviewed Open Access Fully Refereed Journal Since 2015
Submit Paper

OUTLIER DETECTION ON FINANCIAL CARD OR ONLINE TRANSACTION DATA USING MANHATTAN DISTANCE BASED ALGORITHM

V.Kathiresan, Dr.N.A.Vasanthi

Download Full PDF
Article Information
  • Published In Volume 2, Issue 12 (2016)
  • Publication Date July 31, 2026
  • Manuscript ID IJCRCST-DECEMBER16-01
  • Article Type Research Paper
  • Pages 1100 - 1103
  • 5 Views 0 Downloads

Abstract

Outlier detection is a technique in statistics to find anomaly in a dataset or data distribution. Anomaly or outlier is a data object which is deviated from the existing group of data objects. Finding outlier in financial card or online transaction leads fraud detection or fraud suspicion.With the increase in the number of credit and debit card transactions, there has been a substantial increase in the number of fraudulent card transactions too. As the data of RBI, Indian banks are reported close to 27614 and 3835 cases of credit and debit-card related frauds between April 2011 and September 2014. There were additional 2000 cases of internet-banking fraud in that same period. In transaction details two dimensions are taken into account to detect outlier those are transaction number and amount. The outlier is detected based on the MDBA( Manhattan Distance Based Algorithm) Manhattan Distance is mostly used to find distance between two data objects.

Keywords

Outlier detection anomaly Data objects Financial card MDBA

Authors

V.Kathiresan
Dr.N.A.Vasanthi
How to Cite this Article

V.Kathiresan, Dr.N.A.Vasanthi (2016). "OUTLIER DETECTION ON FINANCIAL CARD OR ONLINE TRANSACTION DATA USING MANHATTAN DISTANCE BASED ALGORITHM". International Journal of Contemporary Research in Computer Science and Technology, 2(12), pp. 1100-1103.