Insurance Policy Advisor Using Fuzzy Decision Tree Technique

摘要

People now realized how crucial the insurance is for their protection against irregular and uncertainty situations that will affect life and assets. Customers commit to insurances with various needs based on different backgrounds, economic conditions and risk preference degrees. However, determining the suitable insurance, would be troublesome. For that reason, a specific analysis should be done in matching the protection needs that would be provided by the right insurance to the customers. In understanding the requirements by customers, we explored the patterns in existing records. We have done a thorough experiment using Decision Tree to determine a suitable insurance policy based on the given information from the user. Eight variables were identified from literatures as the inputs, which are; sex, age, body mass index, occupation, annual income, smoking, drinking and contribution per month. These experiments managed to produce meaningful rules in determining four insurance types: life, health, annuity and investment-oriented. The combination of Decision Tree and fuzzy inference methods did improve the accuracy of selection of insurance type for the customers. These combination methods were applied in our model which is applicable for Malaysian population in general. This application would be beneficial to customers to manage and protect their assets in a more proper way.