| 1345-1400 | Welcome by chairs | |
| 1400-1415 | cwm rules | Tim Berners-Lee, W3C |
| 1415-1430 | SweetRules: Tools for RuleML Inferencing and Translation | Benjamin Grosof, MIT |
| 1430-1445 | OO jDREW: A Java-Based Rule Engine for Object-Oriented RuleML | Harold Boley, Marcel Ball, and Bruce Spencer, National Research Council of Canada and U. New Brunswick |
| 1445-1500 | FLORA 2 | Michael Kifer, Stony Brook University |
| 1500-1515 | Discussion | |
| 1515-1530 | Break | |
| 1530-1545 | Hoolet | Sean Bechhofer and Ian Horrocks, University of Manchester |
| 1545-1600 | Use of SWRL for Ontology Translation | Mike Dean, BBN Technologies |
| 1600-1615 | Rule based inference support in Jena2 | Dave Reynolds, HP Laboratories |
| 1615-1630 | ROWL: Rule Language in OWL and Translation Engine for JESS | Norman Sadeh, Fabien Gandon, and Mithun Sheshagiri, CMU |
| 1630-1635 | IBM CommonRules Update | Hoi Chan, IBM |
| 1635-1700 | Discussion | |
The applet version of OO jDREW (http://www.jdrew.org/oojdrew), a Java-based OO RuleML engine, is presented. OO jDREW uses subClassOf taxonomies from RDFS for order-sorted unification and combines forward (data-driven) with backward (query-driven) reasoning. Interaction employs the POSL presentation syntax (http://www.ruleml.org/submission/ruleml-shortation.html) or directly OO RuleML markup. A sorted discount example and the New Brunswick Business Knowledge Base use case (http://www.ruleml.org/usecases/nbbizkb) will be run.
Rule based inference support in Jena2Jena is a well-known Java toolkit for developing semantic web applications. It supports RDF, RDFS, OWL (and legacy DAML) through a rich API and a comprehensive set of tools (including parsers, serializers, stores, and query processing).
In August 2003 we released a substantial new version of Jena, Jena2. Among many internal and external enhancements a key goal for Jena2 was to offer inference capabilities. This currently comprises a simple open API to support multiple inference engines and a suite of built-in rule based engines. These rule engines are used as the implementation vehicle for supporting different profiles of RDFS and OWL inference, but are also directly exposed to developers in the form of a GenericRuleReasoner which can be used for general rule-based processing of RDF data.
The Jena2 rule reasoner offers both a forward production engine and an LP-style (tabled) backward chaining engine. A novel feature of the design is that that these two can coupled together in a hybrid execution mode that allows rule developers to trade-off eager v. lazy processing.
In this talk we'll briefly sketch how inference fits into the Jena2 framework and then outline the rule engines provided.
This presentation will provide an overview of ROWL, a rule language in OWL, and an associated translation engine for JESS. We will discuss our experience using ROWL and extensions we have developed to capture and enforce privacy policies, decision rules and automated service identification rules in the context of several application environments. If time permits, the presentation will also include a description of rule editing functionality we have developed. Finally, we will attempt to contrast ROWL with other similar efforts aimed at adding rules to the the Semantic Web.
Relevant URLs:
ROWL: http://mycampus.sadehlab.cs.cmu.edu/public_pages/ROWL/ROWL.html
MyCampus project: http://almond.srv.cs.cmu.edu/~sadeh/mycampus.htm
Extensions of ROWL for privacy and context awareness:
http://almond.srv.cs.cmu.edu/~sadeh/Publications/Small%20Selection/Semantic_Web_Technologies_ArticleWSJ.pdf
Date: May 22, 2004
Location: New York, NY USA