| | | |

Internet Logic

Semantic Web Laboratory

What is the Semantic Web?
The Semantic Web represents the meaning of distributed Web objects based on metadata such that (interactive) engines or (autonomous) agents can search, integrate, and reuse required objects across applications with high precision.

The Semantic Web Laboratory (SemWebLab) at the NRC Institute for Information Technology (NRC-IIT) was established in 2002 to develop Semantic Web tools and applications and to coordinate with similar efforts in Canada and worldwide. SemWebLab is an effort of the Internet Logic Group jointly with the Human Web and e-Learning Groups as well as NRC-CISTI. External partners include the Faculty of Computer Science at the University of New Brunswick, the Competence Center Semantic Web at DFKI, and the international Rule Markup Initiative.

At the basic layer, SemWebLab develops ontologies consisting of taxonomies that classify Web objects along with rules, typed by taxonomies, for integrity checking and knowledge inference. SemWebLab also studies agents that use ontologies to support, e.g., the similarity retrieval and composition of learning objects. We consider Semantic Web Services as the special case of active Web objects. SemWebLab has a focus on metadata extraction to cope with the vast number of Web objects that are natural language documents.

Opportunities

Public-Private Partnerships: SemWebLab has been networking across the public and private sectors, and will be happy to work with new partners in all issues involving the Semantic Web in Canada and worldwide. The Business Development and NRC-Industrial Research Assistance Program Offices at NRC-IIT – e-Business in Fredericton, as well as our Industrial Partnership Facility, offer a unique environment for incubating Canadian Semantic Web businesses.

Semantic New Brunswick: SemWebLab has started an initiative that uses techniques of the Semantic Web to move towards a Semantic New Brunswick (‘Semantic NB’). This is a novel effort for a region: utilising Semantic Web techniques to accelerate e-Business by enabling participants to advertise and search product and service catalogues, contracting policies and rules, as well as governmental regulations and opportunities. The common “semantic interface” for NB will help companies and government agencies as well as customers and citizens.

Standardization efforts: SemWebLab is involved in various standards bodies and initiatives such as W3C, OASIS, and OMG, as well as RuleML. This is because the Semantic Web calls for standardization on all levels, from ontology languages and tools, to the various ontologies themselves, to agent interoperation protocols, to entire e-business platforms. SemWebLab also acts a clearinghouse for Semantic Web software in NRC.

Prototypes

Some prototypes of SemWebLab are listed here with links to descriptions and researchers. More prototypes are in the pipeline, so feel free to contact us.

RuleML: The Rule Markup Language is developed towards a canonical Web interlingua for business rules using XML markup, formal semantics, and efficient implementations. RuleML has recently been extended for Object-Oriented Modelling and redefined in XML Schema. An implementation is available as part of jDREW. RuleML is also contributing the rule part of the Semantic Web Rule Language.

RACOFI: The Rule-Applying Collaborative Filtering system combines multidimensional collaborative ratings with RuleML-based rules to recommend Web objects such as CD albums that best match user queries. The first prototype, RACOFI Music has been on-line since August 2003. Current work on RACOFI Composer extends this platform by a playlist for MP3 songs, advanced filtering rules, etc.

Metaxtract: The Metaxtract framework enables automatic or semi-automatic semantic annotation using linguistic techniques over text-rich resources. Its NBBizMapper and NBBizKB case studies have helped making regional businesses in New Brunswick more visible by putting them on a semantic-geographic map and by formalising facts and rules about them.

AgentMatcher: The Agent Match-Maker architecture employs similarity, pairing, and negotiation techniques to facilitate transactions in virtual markets. It is based on weighted trees that uniformly describe the products advertised / searched by seller / buyer agents. The general architecture has been instantiated to e-Learning and Power Grid Transactions.

Research Contact

Dr. Harold Boley
Research Officer
Internet Logic
NRC Institute for Information Technology
46 Dineen Drive
Fredericton, NB E3B 9W4
Telephone: +1 (506) 444-0385
E-mail: Harold Boley

Business Contact

Marc-Alain Mallet
Business Development Officer
Business Development Office, New Brunswick
NRC Institute for Information Technology
46 Dineen Drive
Fredericton, NB E3B 9W4
Telephone: +1 (506) 444-0394
Fax: +1 (506) 452-3859
E-mail: Marc-Alain.Mallet@nrc-cnrc.gc.ca