An autonomous GP-based system for regression and classification problems

The aim of this research is to develop an autonomous system for solving data analysis problems. system, called Genetic Programming-Autonomous Solver (GP-AS) contains most of the features requ by an autonomous software: it decides if it knows or not how to solve a particular problem, it construct solutions for new problems, it can store the created solutions for later use, it can improve existing solutions in the idle-time it can efficientlymanage the computer resources for fast running sp and it can detect and handle failure cases. The generator of solutions for new problems is based o adaptive variant of Genetic Programming. We have tested this part by solving some well-kn problems in the field of symbolic regression and classification. Numerical experiments show that the AS system is able to perform very well on the considered test problems being able to successf compete with standard GP having manually set parameters. 2008 Elsevier B.V.. All rights reser