The exemplar clusters are embedded in a network of associations among words that map relations of similarity at all levels.
We can overcome the problem of data sparseness by applying not co-occurring words but co-occurring clusters to the similarity of target words.
Is there a group difference in the number and size of clusters produced?
We will then examine whether the resulting clusters are associated with cohort status and relevant sociodemographic variables.
At present the model only contains a limited framework supporting node- and tree-based clusters.
Finally, there are a variety of clustering and data-mining techniques for understanding the structure of the data in other ways.
The variants are not interspersed with one another; but form segregated clusters.
Assuming known categories (classes) of objects are available, we can compare clustering results by assessing the agreement of the clusters with the classes.
