作者: Bernd Brückner
DOI: 10.1007/11893295_29
关键词:
摘要: In the analysis of biological data artificial neural networks are a useful alternative to conventional statistical methods. Because its advantage in analyzing time courses Multilevel Hypermap Architecture (MHA) is used for stimulus related data, exemplified by fMRI studies with auditory stimuli. Results from investigations MHA show an improvement discrimination comparison With interface well known BrainVoyager software and GUI MATLAB easy usability good visualization results possible. The extension introduced Kohonen. By means it possible analyze structured or hierarchical (data priorities, context, series, varying exactness), which difficult impossible do self-organizing maps so far.