Open Access Open Access  Restricted Access Subscription Access

Online Transaction Fraud Detection on Global Behaviour for E-Commerce

Ms. P. Gomathi, Mr. S. Gopinath, Ms. R. Umamaheswari


Banks and financial firms use investigation to separate false cooperation from authentic business exchanges. Skeptical analysis is still in reach to manipulate either to apply analytics and machine learning to define normal activity based on a customer’ history and distinguish it from unusual behaviour indicating fraud. One of the key values of the banking industry has been its 'Customer-Focused' mindset but in the present world the movement is leaning towards 'Customer-Centric'. The challenges the Banking and financial industry face is in the form of fraudulent transactions. The massive improvement in technology and communication which is combined with an explosive growth of data and information had given rise to global consumer an awareness and empowerment. With the forecast in shopper elements the financial business has a chance to build up an improved client commitment technique. Hence, the implementation of defensive misrepresentation recognition framework is irreplaceable for all banks to limit their misfortunes. Different methodologies dependent on Machine learning, Sequence Alignment have advanced in recognizing different false exchanges. We emphasize the efficient way of fraud detection mechanism and are very much compulsory rather than using simple classification techniques. A KNN algorithm is an evolutionary search and non-parametric technique that impersonate natural evolution to find the best solution to a problem. This algorithm is used for regression and classification problem in deducting fraudulent transaction and minimizing the number of false alerts. If this algorithm is applied into bank fraud detection system, the expectations of fraud transactions can be anticipated after their transactions.

Full Text:



  • There are currently no refbacks.