Credit rating of natural person by expert knowledge compilation in logic basis of neural networks

International Workshop "Advanced Technologies in Material Science, Mechanical and Automation Engineering – MIP: Engineering – 2019"

材料科学;机械制造;原子能学

Rzayev, R.R.^1^2 ; Aliyev, A.A.^3 ; Kravets, O.Ja.^4

Institute of Control System of Azerbaijan National Academy of Science, B. Vahabzadeh st. 9, Baku

AZ1141, Azerbaijan^1

Azerbaijan State Oil and Industrial University, Azadlyg Ave. 16/21, AZ1010, Azerbaijan^2

Baku State University, Zakhid Khalilov st. 23, AZ1148, Azerbaijan^3

Voronezh State Technical University, 20 let Oktyabrya st. 84, Voronezh

394006, Russia^4

关键词: Credit ratings;    Expert estimates;    Expert knowledge;    External knowledge;    Heuristic knowledge;    Multilayer feedforward neural networks;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/537/4/042028/pdf
DOI  :  10.1088/1757-899X/537/4/042028

学科分类：材料科学（综合）

来源: IOP

【 摘 要 】

The paper discusses the combined use of expert systems and neural network to evaluation the solvency of natural person. Corresponding comprehensive approach to individual credit rating is proposed by compilation of the expert and/or heuristic knowledge about the estimates of the solvency of potential borrowers under uncertainty. Obtained expert estimates of the current solvency of individuals are based on the preliminary expert estimate of influence factors for their ranking and the weights of their relative influence. Adequate translation of the external knowledge relative to weighted summary estimates of natural person solvency in effective internal knowledge is compiled in the logical basis of a multi-layer feedforward neural network.