The Multilevel Hypermap Architecture (MHA) is an extension of the Hypermap introduced by Kohonen. By means of the MHA it is possible to analyze structured or hierarchical data (data with priorities, data with context, time series, data with varying exactness), which is difficult or impossible to do with known self-organizing maps so far. A new adaptation of the learning algorithm and its implications for data analysis is the main aspect of this paper. With the generation of hypotheses the MHA is able to detect untrained data relationships in data sets. Beside the advantages in data analysis this approach can also be a contribution to the field of artificial intelligence.
展开▼