Genetic programming for service creation in intelligent networks

Abstract

Intelligent Networks are used by telephony systems to offer services to customers. The creation of these services has traditionally been performed by hand, and has required substantial effort, despite the advanced tools employed. An alternative to manual service creation using Genetic Programming is proposed that addresses some of the limitations of the manual process of service creation. The main benefit of using GP is that by focussing on what a service is required to do, as opposed to it’s implementation, it is more likely that the generated programs will be available on time and to budget, when compared to traditional software engineering techniques. The problem of closure is tackled by presenting a new technique for ensuring correct program syntax that maintains genetic diversity.