Smart Financial Fraud Detection and Customer risk Profilling in Finalcial Institutions to Identify Potential Criminals using Genetic Markov Algorithm

Abstract

keep growing with tremendous growth of electronic transaction, so will the rate of electronic transaction fraud since people will rely more and more on computerized process for their daily activities. Hence, there are needs for more accurate and reliable approach for electronic transaction fraud detection which will help to reduce the illegal activity to the lowest minimum. The use of electronic transaction has increased to a great extent and it caused an explosion in the electronic fraud. Fraud has become one of the major ethical issues in the financial industry. Fraud associated with electronic transaction are also rising today as it is the major mode of payment for both online as well as regular purchase. In order to detect frauds from the mix of genuine as well as fraudulent transactions, efficient fraud detection techniques to detect them accurately are vital rather than simple pattern matching techniques. Here an approach is done to detect the electronic transaction fraud and classify the fraud as either low risk, medium risk or high risk transaction to the financial institution using a fusion approach of genetic and hidden markov algorithm which involve stages of pre-processing in which anonymous transactions were used, genetic algorithm was modelled for feature selection and hidden markov model for classification of fraud as low, medium and high risk transaction. The proposed model is done on existing electronic transaction dataset (anonymous and imbalanced). This research work propose the use of hidden markov model and genetic algorithm to build a model that is able to detect fraud and categorize the customer transaction into three risk levels as low risk, medium risk or high risk transaction to the financial institution to serve as a mechanism which can effectively detect and prevent fraud with great accuracy.