18-08-2012, 11:01 AM
DISTRIBUTED DATAMINING IN CREDIT CARD FRAUD DETECTION
Distributed Data Mining.doc (Size: 57.5 KB / Downloads: 34)
Abstract
Credit card transactions continue to grow in number, taking a larger share of the US payment system, and have led to a higher rate of stolen account numbers and subsequent losses by banks. Hence, improved fraud detection has become essential to maintain the viability of the US payment system. Banks have been fielding early fraud warning systems for some years. We seek to improve upon the state-of-the-art in commercial practice via large scale data mining. Scalable techniques to analyze massive amounts of transaction data to compute efficient fraud detectors in a timely manner are an important problem, especially for e-commerce.
Besides scalability and efficiency, the fraud detection task exhibits technical problems that include skewed distributions of training data and non-uniform cost per error, both of which have not been widely studied in the knowledge discovery and data mining community. Our proposed methods of combining multiple learned fraud detectors under a "cost model" are general and demonstrably useful; our empirical results demonstrate that we can significantly reduce loss due to fraud through distributed data mining of fraud models.
Existing System
The current system will only accept the credit card payments blindly. There is no scientific and intelligent approach to detect the credit card frauds. So it is not possible to verify if the card used by customer is genuine one or a stolen one.
Proposed System
The proposed system is the most intelligent way to analyze and detect the fraud transactions. It uses cost model to analyze the transactions. The system requires more data mining and bridging techniques to connect to the different sites in the distributed databases. The transactions can be identified and classified using classifiers at a single site based on all the sites there is meta classifiers to classify the transactions. There are learning agents called learners which will learn from the classifiers.
Scope of the System
The scope of the system includes designing the learning agents, classifiers, meta classifiers, meta learners, implementing the cost model in differentiating fraud transactions from legitimate ones. Any specification-untraced errors will be concentrated in the coming versions, which are planned to be developed in near future.
Distributed Data Mining.doc (Size: 57.5 KB / Downloads: 34)
Abstract
Credit card transactions continue to grow in number, taking a larger share of the US payment system, and have led to a higher rate of stolen account numbers and subsequent losses by banks. Hence, improved fraud detection has become essential to maintain the viability of the US payment system. Banks have been fielding early fraud warning systems for some years. We seek to improve upon the state-of-the-art in commercial practice via large scale data mining. Scalable techniques to analyze massive amounts of transaction data to compute efficient fraud detectors in a timely manner are an important problem, especially for e-commerce.
Besides scalability and efficiency, the fraud detection task exhibits technical problems that include skewed distributions of training data and non-uniform cost per error, both of which have not been widely studied in the knowledge discovery and data mining community. Our proposed methods of combining multiple learned fraud detectors under a "cost model" are general and demonstrably useful; our empirical results demonstrate that we can significantly reduce loss due to fraud through distributed data mining of fraud models.
Existing System
The current system will only accept the credit card payments blindly. There is no scientific and intelligent approach to detect the credit card frauds. So it is not possible to verify if the card used by customer is genuine one or a stolen one.
Proposed System
The proposed system is the most intelligent way to analyze and detect the fraud transactions. It uses cost model to analyze the transactions. The system requires more data mining and bridging techniques to connect to the different sites in the distributed databases. The transactions can be identified and classified using classifiers at a single site based on all the sites there is meta classifiers to classify the transactions. There are learning agents called learners which will learn from the classifiers.
Scope of the System
The scope of the system includes designing the learning agents, classifiers, meta classifiers, meta learners, implementing the cost model in differentiating fraud transactions from legitimate ones. Any specification-untraced errors will be concentrated in the coming versions, which are planned to be developed in near future.