Predictive modelling for motor insurance claims using artificial neural networks

• 主要な著者: Mohd. Yunos, Zuriahati, Ali, Aida, Shamsyuddin, Siti Mariyam, Ismail, Noriszura, Sallehuddin, Roselina
• フォーマット: 論文
• 出版事項: International Center for Scientific Research and Studies 2016
• 主題: QA75 Electronic computers. Computer science
• オンライン･アクセス: http://eprints.utm.my/id/eprint/74129/; https://www.scopus.com/inward/record.uri?eid=2-s2.0-85010441001&partnerID=40&md5=af579fb1c71f9ff343e1c7145bd68b79

The expected claim frequency and the expected claim severity are used in predictive modelling for motor insurance claims. There are two category of claims were considered, namely, third party property damage (TPPD) and own damage (OD). Data sets from the year 2001 to 2003 are used to develop the predictive model. The main issues in modelling the motor insurance claims are related to the nature of insurance data, such as huge information, uncertainty, imprecise and incomplete information; and classical statistical techniques which cannot handle the extreme value in the insurance data. This paper proposes the back propagation neural network (BPNN) model as a tool to model the problem. A detailed explanation of how the BPNN model solves the issues is provided.