Title :
A genetic constructive induction model
Author_Institution :
Sch. of Cognitive & Comput. Sci., Sussex Univ., Brighton, UK
Abstract :
A hybrid model which uses genetic programming as part of a constructive induction system for supervised learning tasks is presented. The results of the experiments suggest that the model is an effective tool for increasing the generalisation performance of backpropagation in solving parity problems. The model also offers a potentially strong approach to solve problems of data mining
Keywords :
backpropagation; data mining; generalisation (artificial intelligence); genetic algorithms; backpropagation; data mining; generalisation performance; genetic constructive induction model; genetic programming; hybrid model; parity problems; supervised learning tasks; Backpropagation algorithms; Data mining; Decision trees; Feedforward neural networks; Feeds; Genetic programming; Learning systems; Neural networks; Supervised learning; Testing;
Conference_Titel :
Evolutionary Computation, 1999. CEC 99. Proceedings of the 1999 Congress on
Conference_Location :
Washington, DC
Print_ISBN :
0-7803-5536-9
DOI :
10.1109/CEC.1999.781928
