Next:The
contents of this Up:The
methodology SeSKA and Previous:The
methodology SeSKA and
Forming the models
Knowledge for constructing the DM and the DG can be inserted in small fractions
as SOOKAT takes care of the structuring. Fractions inserted are normally
-
default value suggestions for attributes of concrete subclasses of abstract
concepts, and
-
abstract dependencies between attributes of abstract concepts.
These fractions are elicited from multiple knowledge sources in arbitrary
order. The DM and DG maintain the source of each fraction. Traceability
between models connects the fractions with the final KB.
Also statistical analyses can be utilized in DM formation, as described
in section 2. Statistical analyses are
not integrated with the tool SOOKAT at the moment, so they have to be run
separately.
The IS can be formed automatically based on DGs.
Next:The
contents of this Up:The
methodology SeSKA and Previous:The
methodology SeSKA and
Päivikki Parpola
Sat Oct 14 22:52:14 EEST 2000