A parameter optimized approach for improving credit card fraud detection
The usage of credit cards has highly increased due to high-speed innovation in the electronic commerce technology. Since credit card turns out to be the majority well-liked manner of payment for mutually online as well as habitual purchase, cases of fraud correlated through it are as well increasing. In normal Hidden Markov Model the problem of cannot find an optimal state sequence for the underlying Markov process also this observed sequence cannot be viewed as training a model to best fit the observed data. In this research, the main aim is to model the sequence of observations in credit card transaction processing using an Advanced Hidden Markov Model (AHMM) and show how it can be utilized for the exposure of frauds. In this process an AHMM is initially trained with the regular manners of a cardholder. If an incoming credit card transaction is not recognized by the trained AHMM with adequately high probability, it is believed to be fraudulent. This proposed work desire to regulate the model parameters to best fit the observations. The ranges of the matrices (N and M) are fixed but the elements of A,B and #960; are to be decided, focus to the rank stochastic condition. The information that can efficiently re-estimate the model itself is one of the more incredible features of HMMs this referred here as AHMM.
Keywords: Hidden Markov Model (HMM), Advanced Hidden Markov Model (AHMM), Hill Climb, and credit card fraud detection
Download Full-Text
ABOUT THE AUTHORS
Prakash
done M.Sc (CT), from Periyar University Salem in 2001, completed M.Phil.(CS), from Manonmaniam Sundaranar University,Tirunelveli in 2003. Received MCA, from Periyar University Salem in 2011. Currently working as a Asst. Professor in Dept. of Computer Applications, Hindusthan College of arts and science, Coimbatore. His research area is data mining.
Chandrasekar
Dr. C. Chandrasekar received his Ph.D. degree from Periyar University, Salem, TN, India. He has been working as Associate Professor at Dept. of Computer Science, Periyar University, Salem – 636 011, Tamil Nadu, India. His research interest includes Wireless networking, Mobile computing, Computer Communication and Networks. He was a Research guide at various universities in India. He has been published more than 50 research papers at various National / International Journals.
Prakash
done M.Sc (CT), from Periyar University Salem in 2001, completed M.Phil.(CS), from Manonmaniam Sundaranar University,Tirunelveli in 2003. Received MCA, from Periyar University Salem in 2011. Currently working as a Asst. Professor in Dept. of Computer Applications, Hindusthan College of arts and science, Coimbatore. His research area is data mining.
Chandrasekar
Dr. C. Chandrasekar received his Ph.D. degree from Periyar University, Salem, TN, India. He has been working as Associate Professor at Dept. of Computer Science, Periyar University, Salem – 636 011, Tamil Nadu, India. His research interest includes Wireless networking, Mobile computing, Computer Communication and Networks. He was a Research guide at various universities in India. He has been published more than 50 research papers at various National / International Journals.
