The process of of transforming a multi-relational dataset, containing structured examples, into a
propositional dataset with derived attribute-value features, describing the structural
properties of the examples. The process can thus be thought of as summarising data
stored in multiple tables in a single table (the target table) containing one record per
example. The aim of this process, of course, is to pre-process multi-relational data for
subsequent analysis by attribute-value learners.
