Discovery scientific laws by hybrid evolutionary model
| Autor: | Xu Tan, Guangming Lin, Tao Hu, Sanfeng Chen, Zuo Kang, Fei Tang |
|---|---|
| Rok vydání: | 2015 |
| Předmět: |
Scientific law
Computer science business.industry Cognitive Neuroscience Evolutionary algorithm Ode Genetic programming Machine learning computer.software_genre Computer Science Applications Task (project management) Artificial Intelligence Ordinary differential equation Natural science Artificial intelligence business computer |
| Zdroj: | Neurocomputing. 148:143-149 |
| ISSN: | 0925-2312 |
| DOI: | 10.1016/j.neucom.2012.07.058 |
| Popis: | 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. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full Text from ScienceDirect