A hybrid approach to auto-insurance claim processing system

In this paper, we describe the implementation of auto-insurance claim processing system by using neural network and fuzzy technique. The important factor of compensation rate calculation is determining accident responsibility rate between the injured and the assaulter. Generally, the determination of responsibility rate mainly depends on the expertise of auto-insurance experts and the accident report by the police. However, inconsistency between experts' opinions and unattainability to assure the allocation of certainty factor bring out various uncertainty problems. For resolving these uncertainty problems, we introduce the expert system which is able to compute the more precise compensation rate by applying neural network and fuzzy techniques based on the results of rule based system.<
>