Cloud-Native Big Data AI/ML Framework for Risk Intelligence and Fraud Control in Banking and Insurance Ecosystems

Abstract

AI/ML technologies are exploited to detect, mitigate, control, and minimize risks in banking activities as they create, transmit, lend, invest, insure, and secure value. AI/ML paradigms and technologies based on microservice architecture support dynamic delivery of probabilistic native support of banking actions and transactions. Notable categories of AI/ML technologies are based on risk intelligence, fraud detection and prevention, and fraud control. Risk intelligence examines the risk area of the banking ecosystem. Risk intelligence warns participants about the risks involved in their actions or transactions, predicts a credit card operation's logical behaviour in order to set dedicated limits, and determines the best approach to set a limit for a credit score. Fraud detection identifies fraud in real time using analysis of users’ attributes and account history, monitor behaviour changes or sudden changes of an attribute, and classify whether an operation will be done by a legit user or not. Fraud prevention detects fraud before it occurs using knowledge of historical cases to find the relationship between data in order to alert when a similar case may happen and detect unusual patterns to create rules for risky operations. Fraud control deals with the management of fraudulent approaches, actions or transactions for the banking sector. Practices for fraud control address identifying tendencies in the banking sector for fraud detection—vent consumers’ intolerance for crime, minimize social disparity, increase competence to minimize price of fraud.