Text information processing depends critically on the proper document representation. Traditional models, like vector space
model, have significant limitations because they do not consider semantic relations amongst terms. In this paper we analyze
a document representation that uses an association graph scheme model called Global Association Distance Model or GADM, the
significance of the formal distance for the association strength, and the application of several distance-strength functions
in this model. We evaluate this significance for topic classification tasks.
Keywords Document modelling - Document processing - Document re-presentation