Speaker: John Olukuru
Fraud Management is inevitable to insurance companies as this informs the level of risk covered which in turn affects the premiums being charged. With fraudsters advancing their gimmicks continuously, insurers need to embrace big data techniques that can quickly and accurately access, link and analyze massively increasing volume of data at its most granular format.
For insurance companies, big data is a collection of all available data from both traditional sources (proposal forms, sales records, systems used for underwriting, claims, finance etc.) and digital sources both inside and outside the organization. Big data helps insurers have a superior understanding of their business environment and their client behaviors in all dimensions which informs fraud management techniques. This is made possible by the use of big data analytics tools; Machine learning is such a tool that addresses the question of how to build computers that improve automatically through experience. It is one of today’s most rapidly growing technical field lying at the intersection of computer science and statistics and at the core of artificial intelligence and data science.
Machine learning has been a major contributor for detecting different types of financial fraud through its diverse methods. This study proposes a hybrid approach that considers both supervised and unsupervised learning which when combined show superior performance across different applications. By combining both techniques, the intention is to raise detection rates of known fraudulent activities and decrease the false positive rate for unknown fraud cases. We will evaluate two advanced machine learning approaches, Support Vector Machine and Neutral Networks together with self-organizing feature Map as part of an attempt to improve fraud detection in Medical insurance.
By using machine learning technology, the study adapts to automatically flagging new emerging fraud cases in real time with lack of prior information about fraud patterns.