Accession Number : ADA264961


Title :   Unsupervised Splitting Rules for Neural Tree Classifiers


Descriptive Note : Technical rept.,


Corporate Author : BROWN UNIV PROVIDENCE RI


Personal Author(s) : Perone, Michael P ; Intrator, Nathan


Full Text : https://apps.dtic.mil/dtic/tr/fulltext/u2/a264961.pdf


Report Date : 17 May 1993


Pagination or Media Count : 7


Abstract : This paper presents two unsupervised neural network splitting rules for use with CART-like neural tree algorithms in high dimensional data space. These splitting rules use an adaptive variance estimate to avoid some possible local minima which arise in unsupervised methods. We explain when the unsupervised splitting rules outperform supervised neural network splitting rules and when the unsupervised splitting rules outperform the standard node impurity splitting rules of CART. Using these unsupervised splitting rules lead to a nonparametric classifier for high dimensional space that extracts local features in an optimized way.... CART, Unsupervised feature extraction, Neural trees.


Descriptors :   *NEURAL NETS , *NETWORKS , *CLASSIFICATION , *ADAPTATION(PHYSIOLOGY) , ALGORITHMS , IMPURITIES , EXTRACTION , SPLITTING , STANDARDS , ESTIMATES , NODES , TREES


Subject Categories : Psychology


Distribution Statement : APPROVED FOR PUBLIC RELEASE