Artificial Neural Networks have achieved satisfactory
results in different fields such as example
classification or image identification. Real-world
processes usually have a temporal evolution, and they
are the type of processes where Recurrent Networks have
special success. Nevertheless they are still
reluctantly used, mainly due to the fact that they do
not adequately justify their response. But, if ANNs
offer good results, why giving them up? Suffice it to
find a method that might search an explanation to the
outputs that the ANN provides. This work presents a
technique, totally independent from ANN architecture
and the learning algorithm used, which makes possible
the justification of the ANN outputs by means of
expression trees.
%0 Conference Paper
%1 conf/asc/GestalRDP06
%A Gestal, Marcos
%A Rabuñal, Juan R.
%A Dorado, Julian
%A Pereira Loureiro, Javier
%B Artificial Intelligence and Soft Computing
%C Palma de Mallorca, Spain
%D 2006
%E Pobil, Angel P. Del
%I IASTED/ACTA Press
%K Algorithm Artificial Capabilities, Example Extraction, Generalisation Generation, Networks, Neural Prediction Recurrent Rule Series algorithms, genetic of programming,
%P 323--328
%T Description of RANNs and their generalisation
capabilities by means of rule extraction by genetic
programming
%U http://sabia.tic.udc.es/sabia/secciones/publications/?id=311
%X Artificial Neural Networks have achieved satisfactory
results in different fields such as example
classification or image identification. Real-world
processes usually have a temporal evolution, and they
are the type of processes where Recurrent Networks have
special success. Nevertheless they are still
reluctantly used, mainly due to the fact that they do
not adequately justify their response. But, if ANNs
offer good results, why giving them up? Suffice it to
find a method that might search an explanation to the
outputs that the ANN provides. This work presents a
technique, totally independent from ANN architecture
and the learning algorithm used, which makes possible
the justification of the ANN outputs by means of
expression trees.
%@ 0-88986-612-0
@inproceedings{conf/asc/GestalRDP06,
abstract = {Artificial Neural Networks have achieved satisfactory
results in different fields such as example
classification or image identification. Real-world
processes usually have a temporal evolution, and they
are the type of processes where Recurrent Networks have
special success. Nevertheless they are still
reluctantly used, mainly due to the fact that they do
not adequately justify their response. But, if ANNs
offer good results, why giving them up? Suffice it to
find a method that might search an explanation to the
outputs that the ANN provides. This work presents a
technique, totally independent from ANN architecture
and the learning algorithm used, which makes possible
the justification of the ANN outputs by means of
expression trees.},
added-at = {2008-06-19T17:35:00.000+0200},
address = {Palma de Mallorca, Spain},
author = {Gestal, Marcos and Rabu{\~n}al, Juan R. and Dorado, Julian and {Pereira Loureiro}, Javier},
bibdate = {2007-01-26},
bibsource = {DBLP,
http://dblp.uni-trier.de/db/conf/asc/asc2006.html#GestalRDP06},
biburl = {https://www.bibsonomy.org/bibtex/211deaf85862c43e3fa3580e36d5f09ec/brazovayeye},
booktitle = {Artificial Intelligence and Soft Computing},
editor = {Pobil, Angel P. Del},
interhash = {331576d28119c1dbefe272c4a9a888f9},
intrahash = {11deaf85862c43e3fa3580e36d5f09ec},
isbn = {0-88986-612-0},
keywords = {Algorithm Artificial Capabilities, Example Extraction, Generalisation Generation, Networks, Neural Prediction Recurrent Rule Series algorithms, genetic of programming,},
month = {August 28-30},
pages = {323--328},
publisher = {IASTED/ACTA Press},
timestamp = {2008-06-19T17:40:11.000+0200},
title = {Description of {RANNs} and their generalisation
capabilities by means of rule extraction by genetic
programming},
url = {http://sabia.tic.udc.es/sabia/secciones/publications/?id=311},
year = 2006
}