A new model for credit approval problems: A quantum-inspired neuro-evolutionary algorithm with binary-real representation

Author

De Pinho, Anderson Guimarães ; Vellasco, Marley ; da Cruz, A.V.A.

Author_Institution

Dept. of Electr. Eng., PUC-Rio, Rio de Janeiro, Brazil

fYear

2009

fDate

9-11 Dec. 2009

Firstpage

445

Lastpage

450

Abstract

This paper presents a new model for neuro-evolutionary systems. It is a new quantum-inspired evolutionary algorithm with binary-real representation (QIEA-BR) for evolution of a neural network. The proposed model is an extension of the QIEA-R developed for numerical optimization. The Quantum-Inspired Neuro-Evolutionary Computation model (QINEA-BR) is able to completely configure a feed-forward neural network in terms of selecting the relevant input variables, number of neurons in the hidden layer and all existent synaptic weights. QINEA-BR is evaluated in a benchmark problem of financial credit evaluation. The results obtained demonstrate the effectiveness of this new model in comparison with other machine learning and statistical models, providing good accuracy in separating good from bad customers.

Keywords

credit transactions; evolutionary computation; feedforward neural nets; learning (artificial intelligence); optimisation; quantum computing; binary-real representation; credit approval problems; feedforward neural network; financial credit evaluation; machine learning; neuro-evolutionary systems; numerical optimization; quantum-inspired neuro-evolutionary algorithm; statistical models; Computational modeling; Computer networks; Evolutionary computation; Feedforward neural networks; Feedforward systems; Input variables; Machine learning; Neural networks; Neurons; Quantum computing; classification; genetic algorithms; hybrid neuro-genetic systems; quantum-inspired algorithms;

fLanguage

English

Publisher

ieee

Conference_Titel

Nature & Biologically Inspired Computing, 2009. NaBIC 2009. World Congress on

Conference_Location

Coimbatore

Print_ISBN

978-1-4244-5053-4

Type

conf

DOI

10.1109/NABIC.2009.5393327

Filename

5393327