Retraining the Neural Network for Data Visualization [chapter]

Viktor Medvedev, Gintautas Dzemyda
IFIP International Federation for Information Processing  
In this paper, we discuss the visuaHzation of multidimensional data. A well-known procedure for mapping data from a high-dimensional space onto a lower-dimensional one is Sammon's mapping. The algorithm is oriented to minimize the projection error. We investigate an unsupervised backpropagation algorithm to train a multilayer feed-forward neural network (SAMANN) to perform the Sammon's nonlinear projection. Sammon mapping has a disadvantage. It lacks generalization, which means that new points
more » ... annot be added to the obtained map without recalculating it.
doi:10.1007/0-387-34224-9_4 dblp:conf/ifip12/MedvedevD06 fatcat:725ile6d2jcmfn5hf4v6vwjxkq