On 3/11/2015 3:56 AM, Hassan Aït-Kaci wrote:
>> http://cedar.liris.cnrs.fr/papers/ctr12.pdf
>> Design and implementation of an efficient Semantic Web reasoner
>
> See http://cedar.liris.cnrs.fr/papers/ctr16.pdf for implementation details. (01)
I agree that CEDAR TR 16 is important, and I should have cited it
in my previous note. But this note gives me an opportunity to say
why it's important: (02)
1. TR 16 presents HOOT, a well-designed language for expressing and
querying Hierarchical Ontologies, Objects, and Types. (03)
2. It's a well-written presentation of the general principles with
many examples. They include easy-to-read diagrams, English
explanations, and an analysis of the implications. (04)
3. It relates HOOT to the diagrams and to the T-Box for types,
A-Box for assertions, and Q-Box for queries. It also relates
HOOT to RDF, OWL, SPARQL, SQL, and other logic-based systems. (05)
4. It presents sufficient detail about the implementation and the
algorithms that anyone with some experience in designing and
implementing SW tools can implement HOOT or something like it. (06)
5. The HOOT notation is quite readable. But anyone who prefers an
XML version or a controlled NL version could design a translator
for mapping HOOT to and from their preferred syntactic sugar. (07)
6. HOOT illustrates the kinds of developments that the SW and related
technologies will undergo during the next few years. It shows why
the word 'the' in discussions of OWL should be changed to 'a'. (08)
Following are two paragraphs from the conclusion (page 39). (09)
John
______________________________________________________________________ (010)
5 Conclusion (011)
This document has justified the design of, and detailed an
implementation specification for HOOT, a computer language designed
for processing taxonomic attributed ontologies. We put this work in
the context of the Semantic Web. It is a contribution in the nascent
area of the intelligent processing of distributed knowledge and data
represented as RDF triples. An initial prototype of this specification
was realized as the CEDAR reasoner. Initial performance comparison
tests with the state of the art show that the ideas developed in this
specification are valuable in boosting TBox and ABox querying by orders
of magnitudes. (012)
This initial design is meant as a starter. Later versions of HOOT can
be developed to support more sophisticated OSF theories (ontologies
of concepts abiding by rule defined constraints), focusing always
on efficiency. In the eventuality that such theories may lead to
undecidable inference, we will stay on the pragmatic side and opt
for an incomplete but efficient inference system. Indeed, this is
a trade-off worth making as it enhances semantic expressiveness
while retaining efficient reasoning—just as Prolog is w.r.t.
first-order Horn logic: while semi-decidable it can be implemented
efficiently, and has much greater expressive power than plain
(decidable, but NP-complete) Propositional Logic. (013)
_________________________________________________________________
Message Archives: http://ontolog.cim3.net/forum/ontolog-forum/
Config Subscr: http://ontolog.cim3.net/mailman/listinfo/ontolog-forum/
Unsubscribe: mailto:ontolog-forum-leave@xxxxxxxxxxxxxxxx
Shared Files: http://ontolog.cim3.net/file/
Community Wiki: http://ontolog.cim3.net/wiki/
To join: http://ontolog.cim3.net/cgi-bin/wiki.pl?WikiHomePage#nid1J (014)
|