Dynamic Split-point Selection Method for decision tree evolved by Gene Expression Programming

Gene expression programming (GEP) is a kind of heuristic method based on evolutionary computation theory. GEP has been used to evolve parsimonious decision tree with high accuracy comparable to C4.5. However, the basic GEPDT do not distinguish different attributes, whose boundaries are usually quite different. The basic GEPDT often fails to find optimal split points for some branches and thus handicapped the learning tasks. In this paper, we proposed a simple but effective split-point selection method for GEP evolved decision tree to improve the performance of tree splitting and classification accuracy. Results show that our method can find better generalized ability rules and it is especially suitable for difficult problems with many attributes in different boundaries.