There are four primary reasons for committing a data interpretation project to LegendBurster:
To undertake the interpretation in an environment of formally structured knowledge, admissive of hierarchical data structures, and capable of explicitly managing negative information;
To be able to run queries against data that are impossible, or extremely difficult to express in normal SQL-based queries.
To be provided with, in addition to items which exactly match queries, those items which are close matches to input queries.
To be able to share with other workers highly structured knowledge, of the type that it embedded in classification hierarchies, thereby contributing to the development of language standards.
All of these depend on the development of a high quality ontology for the domain of knowledge represented on the map to be studied. This ontology will often require the introduction of words which do not appear in the original map legend, or in the attribute tables associated with the map.
Nevertheless, it will usually be possible to produce a mapping between most of the terms used in the original map, and the terms of the ontology. Armed with this mapping, it is possible to use the "Add Attributes to All ..." function in LegendBurster to automate many of the steps in building semantic nets for each mapObject.
The "Add Attribute" procedure can then be used to "fill in the gaps" left by attributes whose required values could not be inserted with the "Add Attributes to All ..." function.
Finally, the "Direct Add" function can be used for the transfer of information into the semantic net descriptions which are not needed for querying, but which nevertheless bear information which should appear as part of any query results <view example of query results being displayed, where "deposit name" is not used for matching>.
Proceed to Adding Attributes to Individual MapObjects.