Improved Gene Expression Programming Algorithm With Application

Evolutionary Computation is a hot point in the research area of Artificial Intelligence, Knowledge Engineering and Data Mining. Genetic Algorithm and Genetic Programming are the most important computation models in Evolutionary Computation. In 2001, F. Candida proposed a new Evolutionary Computation method named Gene Expression Programming (GEP). GEP is a lately Genetic Algorithm Programming, simpler representation and the creation of higher levels of complexity than GP and GA. GEP is as simple as Genetic Algorithm, and as functional as Genetic Programming. But for the most problems, GEP is much faster than Genetic Programming in 2 to 4 magnitudes. It is highly parallel, random, auto-adapted search algorithm depend on a kind of which the biological choice and the evolution mechanism.This paper introduces the means of the GEP firstly. Then, Gene Expression Programming is improved and proposed to solve job shop scheduling problem and automatic clustering problem. Job shop scheduling problem is the reduced model of many actual scheduling problem, its research has important theory significance and engineering worth. GEP combines the advantages of Genetic Algorithm and Genetic Programming, it has more powerful ability which solves problem. Gene Expression Programming Algorithm is proposed for applying into job shop scheduling problem efficiently. In order to verify the effectiveness of the proposed method, an example was used for experiment. The obtained scheduling solution time is quite short. The experiment result shows that the proposed algorithm is efficient. The purpose of Data Mining is to abstract potential, valuable knowledge and useful information from plentiful data. Cluster analysis is one of the research domains of data mining,which has important appliances in many domains such as in business, biology, medicine, geography, web archive, and it becomes one of the hot research problems. Without any prior knowledge, Improved Gene Expression Programming is used to solve clustering automatically on various data sets. Finally, two kinds of clustering data are experimented to verify the effectiveness of the proposed approach.