Foundations of Computational Intelligence

0342905 - ÚI 2011 RIV DE eng M - Monography Chapter
Jiřina, Marcel - Jiřina jr., M.
Genetic Selection Algorithm and Cloning for Data Mining with GMDH Method.
Foundations of Computational Intelligence. Vol. 5. Berlin: Springer, 2009 - (Abraham, A.; Hassanien, A.; Snášel, V.), s. 359-376. Studies in Computational Intelligence, 205. ISBN 978-3-642-01535-9
R&D Projects: GA MŠMT(CZ) 1M0567
Institutional research plan: CEZ:AV0Z10300504
Keywords : GMDH method * genetic selection * cloning * classifier
Subject RIV: IN - Informatics, Computer Science

The GMDH MIA is modified by use of the selection procedure from genetic algorithms while including cloning of the best neurons generated. The selection procedure finds parents for a new neuron among already existing neurons according to the fitness and also with some probability from the network inputs. The essence of cloning is slight modifying the parameters of the copies of the best neuron. The genetically modified GMDH network with cloning (GMC GMDH) can outperform other powerful methods. It is demonstrated on some tasks from the UCI MLR.
Permanent Link: http://hdl.handle.net/11104/0185513