Evolving neural-symbolic systems guided by adaptive training schemes: Applications in finance

Abstract

The article presents a hybrid and adaptive intelligent methodology, based on neural logic networks and grammar-guided genetic programming. The aim of the study is to demonstrate how to generate efficient neural logic networks with the aid of genetic programming methods trained adaptively through an innovative scheme. The proposed adaptive training scheme of the genetic programming mechanism leads to the generation of high-diversity solutions and small-sized individuals. The overall methodology is advantageous due to the adaptive training scheme proposed for offering both accurate and interpretable results in the form of expert rules. Moreover, a sensitivity analysis study is provided within the article, comparing the performance of the proposed evolutionary neural logic networks methodology with well-known competitive inductive machine learning approaches. Two financial domains of application have been selected to demonstrate the capabilities of the proposed methodology: (a) classification of credit applicants for consumer loans of a German bank and (b) the credit-scoring decision-making process in an Australian bank. Results seem encouraging since the proposed methodology outperforms a number of competitive existing statistical and intelligent methodologies, while it also produces handy decision rules, short in length and transparent in meaning and use.