On taxonomy of evolutionary computation problems

Taxonomy is the practice of classifying members of a group based on their measurable characteristics. In evolutionary computation the problem of telling when two problems are similar is both challenging and important. An accurate classification technique would yield large benefits by permitting a researcher to rationally choose algorithm and parameter setting based on past experience. A good classification technique would also permit the selection of diverse test suites that would give a useful sense of the proper domain of application of a new technique. This study uses a standard taxonomic technique, hierarchical clustering, on a set of taxonomic characters derived from a comparative study using graph based evolutionary algorithms. The result is a cladogram that classifies the problems used in a reasonable fashion. Based on this we then argue that the technique given here can be used to provide an objective, automatic, extensible classification tool for any collection of evolutionary problems and discuss possible methods for improving the technique.