relatedTo(d1,d2)
equivalent(d1,d2)
specializes(d1,d2)
generalizes(d1,d2)
cites(d1,d2)

weaklyRelatedTo(d1,d2)
stronglyRelatedTo(d1,d2)
maybeEquivalent(d1,d2)


Difference between weaklyRelatedTo and stronglyRelatedTo
========================================================
weaklyRelatedTo is the transitive hull of stronglyRelatedTo.
stronglyRelatedTo is the association providet by the user.
While stronglyRelatedTo always ensures us a semantic relationship
between two documents, it is possible that weaklyRelatedTo associates
semantically unrelated documents to each other.
  example: A document may discuss different (semantically) unrelated 
topics, because it is a kind of summary of what happend recently. 
Therefore a lot of other documents discussing these topics in detail
should be stronglyRelatedTo this summary (in fact these documents are
likely to specialize the summary). Using the weaklyRelatedTo association
we get relations between all of these documents. The only property these
documents share, is to be mentioned in the same summary (and maybe being
published at about the same time). 
 example: Imagine two sets of documents and summaries in each of them. We
get two disjunct graphs. Now another document - called paper later in this 
example - is published (out of the scope of any summary) and this document 
is stronglyRelatedTo two documents - one in each graph. The weaklyRelatedTo 
association now spans all documents. What does the weakRelatedTo association 
two former unrelated documents in this graph mean? It tells us "these 
documents are related to two documents (summaries) containing both  a 
document being reatedTo by a paper." We would not have been able to state 
this kind of relation between these two documents, if we would olny have 
used the transitive kind of relatedTo association. 