Financial fraud is on the rise, posing a serious threat to the financial industry. As a result, banks are constantly under pressure to improve their fraud detection systems. Quite a lot of researchers have cast-off machine learning & data mining techniques to solve this problem in recent years. We present the current state of the art on various fraud methods, as well as identification & prevention techniques anticipated in the literature, such as clustering, grouping, & regression. The aim of this research is to find out which techniques & processes have been perfected so far in order to achieve the best results. The projected investigative work has motivated to discover fraud loan application patterns. In this context the different banking processes are used to mine processes & also help in developing the fraud detection technique using process mining. In this article we evaluate the performance of the proposed process mining system.

Process mining, Machine learning techniques, Loan Process, Banking System, Loan Dataset