A hybrid machine learning system and its application to insurance underwriting
Christos Nikolopoulos, S. Duvendack · 2002
This paper describes the application of evolutionary learning and classification tree techniques to the insurance underwriting domain. These machine learning techniques are used to build a knowledge base of rules for an expert system which determines when an insurance policy should be terminated. The effectiveness of each method is compared with the other and a hybrid method is proposed, which combines both approaches and seems to overshadow the performance of any other single method.>