Discovery scientific laws by hybrid evolutionary model

Abstract Constructing a mathematical model is an important issue in engineering application and scientific research. Discovery high-level knowledge such as laws of natural science in the observed data automatically is a very important and difficult task in systematic research. The authors have got some significant results with respect to this problem. In this paper, high-level knowledge modelled by systems of ordinary differential equations (ODEs) is discovered in the observed data routinely by a hybrid evolutionary algorithm called HEA-GP. The application is used to demonstrate the potential of HEA-GP. The results show that the dynamic models discovered automatically in observed data by computer sometimes can compare with the models discovered by humanity. In addition, a prototype of KDD Automatic System has been developed which can be used to discover models in observed data automatically.