Several Critical Techniques of Genetic Programming and Their Applications for Data Mining
A common problem in data mining is to find accurate data fitting and trend-based forecasting for a dataset. Genetic Programming (GP for short) was accordingly applied, which can particularly induce parse trees with a linear combination of variables in each function node. Different methods of selection, crossover and mutation were also adopted which can be used to avoid the undesirable growth of program size.Additionally, ordinary differential equations and the Particle Swarm Optimization (PSO for short) were used to improve the accuracy of data fitting and forecasting. The results indicate that the improved GP approaches can be applied successfully for accurate data fitting and forecasting.
GP program size ordinary differential equation PSO data mining
Yongqiang Zhang Huashan Chen
School of Information and Electricity-Engineering, Hebei University of Engineering Handan, Hebei Province, P.R.China
国际会议
杭州
英文
440-444
2006-10-12（万方平台首次上网日期，不代表论文的发表时间）