A statistical comparison between an unsupervised neural network and a partially connected neural network in the detection of breast cancer
Smaranda Belciug · Annals of the University of Craiova Mathematics and Computer Science Series · 2010
This paper deals with the comparison of the two neural network methods of learning: supervised (partially connected neural network) and unsupervised (self organizing featuremaps (SOFM), in order to assess their performances on a labeled breast cancer database. A statistical comparison has been made to reveal the di®erences between the two methods regarding diagnosis accuracy and computational time.