Sub-menu Validate->Extract Domain from Use Cases brings a Domain Model Extractor wizard that helps identify domain elements from use cases text elements.
The text element can be accepted as a non-entity bound condition or as an entity bound condition/use case operation.
To accept as non-entity bound condition, select the box labelled ``Accept as non-entity bound condition''.
Use section ``Select parts to match string'' to identify entities and operations. The section shows the condition, use case actor or the use case operation which needs to be analyzed in order to extract system concepts, concepts, aggregates, possible values or operations. The following are the parts of the wizard interface.
There can't be both a Concept and a System Concept within a string to be resolved.
With the wordnetdata directory under the working directory (containing all wordnet data files), the system will detect the operation for you automatically. But this detection is based on extracting the operation under the assumption that its first word is a verb. But this is not guaranteed to work perfectly at all times as there could be words that belongs to both the noun and verb categories. In such a case you have to manually select the correct operation. If you don't have the wordnet data files then the system will require the user to select the operations manually.
Stephane 2007-09-03