Authors:
Daniela Xhemali
;
Christopher J. Hinde
and
Roger G. Stone
Affiliation:
Loughborough University, United Kingdom
Keyword(s):
Genetic programming, Genotype-phenotype mapping, XML, Regular expressions, Software programs.
Related
Ontology
Subjects/Areas/Topics:
Artificial Intelligence
;
Computational Intelligence
;
Evolution Strategies
;
Evolutionary Computing
;
Genetic Algorithms
;
Informatics in Control, Automation and Robotics
;
Intelligent Control Systems and Optimization
;
Representation Techniques
;
Soft Computing
Abstract:
This paper presents an adaptable genetic evolutionary system, which includes an innovative approach to mapping genotypes to phenotypes through XML rules. The evolutionary system was originally created to evolve Regular Expressions (REs) to automate the extraction of web information. However, the system has been adapted to work with a completely different domain – Complete Software Programs – to demonstrate the flexibility of this approach. Specifically, the paper concentrates on the evolution of ‘Sorting’ programs. Experiments show that our evolutionary system is successful and can be adapted to work for challenging domains with minimum effort.