Evaluating the Creditworthiness of a Client in the Insurance Industry Using Adaptive Neuro-Fuzzy Inference System
DOI:
https://doi.org/10.5755/j01.ee.28.1.14194Keywords:
Insurance industry, creditworthiness, decision-making, neuro-fuzzy model, fuzzy logic, artificial neural network.Abstract
The article describes the decision-making neuro-fuzzy model for evaluation of creditworthiness of the client in insurance industry. Formation of the neuro-fuzzy model, its verification and implementation is the main objective of this article. The proposed neuro-fuzzy model consists of two sub-models: neuro-fuzzy sub-model for evaluating insurance indicators and neuro-fuzzy sub-model for evaluating accounting indicators. The neuro-fuzzy sub-model for evaluating insurance indicators has got three input variables (damages, insurance length, insurance penetration), one rule block and one output variable (evaluation of insurance indicators). The neuro-fuzzy sub-model for evaluating accounting indicators has got two input variables (earnings, liquidity 2nd degree), one rule block and one output variable (evaluation of accounting indicators). The total neuro-fuzzy model for evaluating total creditworthiness of the client has got two input variables (evaluation of insurance indicators, evaluation of accounting indicators), one rule block and one output variable (evaluation of creditworthiness of the client). All inputs variables have got three attributes. A membership function of type “gaussmf” was used in the neuro-fuzzy model. Research results show that the proposed neuro-fuzzy model is verified above input data and can be used as a tool for supporting decisions concerning client's credibility in the insurance industry.
