EWV

Insurance fraudsters and white-collar criminals cause annual losses of 4 to 5 billion euros, and the trend is rising. This is the estimate of the German Insurance Association. These cases of fraud result in significant additional costs for insurers, which ultimately translate into higher premiums for insurance customers. As part of the EWV project - Detection of Economic Crime and Insurance Fraud - lawyers, psychologists, economists, and computer scientists have developed a system for fraud prevention. The project placed particular emphasis on the legally compliant preservation of evidence.

The project’s primary goal was to detect large-scale fraud attempts that could systematically harm companies. To this end, the project participants investigated suspected cases of insurance fraud and developed IT-supported methods for detecting manipulation, for example of images, financial data, and authorship. Using specialized analysis methods for images and financial data, they were able to identify suspicious cases or even uncover fraud attempts. Using psychological analysis methods, the team of scientists examined the behavior of insurance fraudsters. From a data protection perspective, the scientists also examined how and to what extent policyholder data can be accessed.

The interdisciplinary nature of the project ensures a holistic approach to the analysis, prevention, and investigation of insurance fraud. Insurers can use the developed methods and recommendations for action both preventively and to investigate insurance fraud.

The project ran from January 2015 to the end of March 2018 and was funded with a total of 1.5 million euros by the Federal Ministry of Education and Research (BMBF). In addition to Fraunhofer SIT, participants included the University of Kassel with two chairs, Dortmund University of Applied Sciences and Arts, the Institute for Psychology & Threat Management (I:P:Bm), Arvato Financial Solutions, Fraunhofer IAO, and, as associated partners, Mannheimer Versicherung and the Gotha Tax Office.