Artificial Neuron Networks versus Cluster Analysis: A comparative study in a sample of children with spina bifida

Abstract

The use of Artificial Neural Networks (ANN) in data analysis in Psychology is not very frequent, The prevalence of multivariate statistical techniques is evident although the majority of these techniques can be implemented in different architectures of ANN. The lack of studies that show the advantages of certain procedures over others could explain the scarce use of ANN. In this work, we have compared the profiles of the clusters obtained using a non-hierarchical classification algorithm (k-means) and those obtained by means of a competitive net. These two procedures have been applied to a sample of children with spina bifida and who are undergoing treatment at the «Virgen del Rocío» Children's Hospital of Seville.