Approach
|
Task
|
Ontology exists?
|
Query, Dump
|
Usability
|
OWL Schema
|
Schema/Data Separation
|
D2RQ
|
uni-directional, SPARQL, LinkedData
|
ad-hoc creation
|
query, dump possible
|
automatic mapping
|
automatic mapping generates poor schema, one table per class, only datatype properties, more possible
|
no, everything created at once
|
Triplify
|
uni-directional, LinkedData
|
ad-hoc creation
|
display of LinkedData
|
manual sql queries
|
supports classes and objectproperties, early version
|
no, everything created at once
|
SquirrelRDF
|
uni-directional, access several dbs via SPARQL
|
no, and none is created
|
query, allows SPARQL
|
manual
|
no, on purpose not
|
no
|
CODE (Man Li)
|
conversion
|
ad-hoc creation
|
dump/batch, ?with possibility to update instance data?
|
semi-automatic, but clear engineering process
|
automatic learning of some hierarchy, domain/range of roles, objectproperties, 11 rules
|
yes, separate, data is imported after schema creation
|
DB2OWL
|
several DBs, aims at integration
|
distinction between a global ontology and local ontologies, existence of a global target ontology mandatory
|
query, uses R2O
|
automatic
|
basically same as CODE (Man Li), similar rules
|
yes
|
DartGrid
|
integration of several DBs
|
integration ontology mandatory
|
query
|
|
|
|
Relational.OWL
|
|
|
|
|
|
|
Protege DataMaster
|
|
|
|
|
|
|
Protege OntoBase
|
|
|
|
|
|
|
Protege DataGenie
|
|
|
|
|
|
|
Virtuoso RDF Views
|
|
|
|
|
|
|
R2O+ODEMapster
|
|
|
|
|
|
|
OntER(Trinkunas)
|
verify acquisition
|
no
|
manual acquisition of the ontology based on ER model
|
manual, but with clear directives, ?could be automated?
|
conversion of ER model, similarity to RDF is exploited
|
no, but a verification method seems to exist, to validate the handcrafted model against the db, based on a db integration tool
|
Habegger
|
learning of a mapping
|
no
|
n.a.
|
proposal of a learning algorithm, no evaluation
|
some hierarchy based on theta subsumption in horn clauses, objectproperties
|
??
|
SPASQL
|
|
|
|
|
|
|