The
Knowledge Annotation Initiative
of the
Knowledge Acquisition Community
(KA)²

1 Introduction

The Knowledge Annotation Initiative of the Knowledge Acquisition Community (KA)² is an initiative officially launched at EKAW-97 to develop an ontology that models the knowledge acquisition community (its researchers, topics, products, etc.). This ontology will form the basis to annotate WWW documents of the knowledge acquisition community in order to enable intelligent access to these documents. (KA)² is an open joint-initiative where the participants are actively involved in (i) a distributive ontological engineering process to model the knowledge acquisition community (a domain ontology), and (ii) annotating webpages relevant for the KA community (the instances of the domain ontology).

(KA)² aims at "intelligent" knowledge retrieval from the Web and automatic derivation of "new" knowledge. In other words, it aims at knowledge-based reasoning on the Web, as opposed to the more usual information retrieval. Another objective of the initiative concerns a distributive ontological engineering process.

2 Research issues

There are three research topics involved in the initiative.

2.1 Ontological engineering

The knowledge acquisition community has to build its own ontology. This will be done in a distributive manner, using the Ontolingua environment. The current version of the ontology can be viewed at the European mirror site of the Ontolingua server of Stanford University. To view the ontology, login as "ontologias-ka2" with password "adieu007". The ontology for the KA community consists of seven related ontologies: an organisation ontology, a project ontology, a person ontology, a research-topic ontology, a publication ontology, an event ontology and a research-product ontology. At the moment (March 1998), we are organising the process how to establish a more definitive version of the ontology. The process will be collaborative and distributed.

2.2 Annotation of web pages

The instances of the ontology have to be provided by so-called "provider agents". Instances appear distributively at the relevant web pages of each provider agent. Web pages are annotated using a new HTML tag, called ONTO. Enclosing web information within this tag makes the information accessible for an ontology-based webcrawler (see below). Examples of annotated web pages can be found at all provider agents. Instructions on how to annotate web pages are also available. The process of how to annotate the web pages of the knowledge acquisition community is still an open issue. There are several possibilities, ranging from one responsible person offering an annotation service to each provider agent taking care of its own pages. Provider agents have to register at the Ontocrawler by sending it a URL of an index file listing the URLs of all annotated web pages.

2.3 Ontocrawler

Given a query, an ontology-based webcrawler (Ontocrawler) has to access the web pages and use the ontology to provide answers. Depending on how rich the ontology is (e.g. the amount of inferencing allowing axioms), Ontocrawler can also deduce "new" information, that is not explicitly stored on the Web. Notice that such inferencing is very common in knowledge-based systems, but not at all for web search engines. Ontocrawler can be accessed at the University of Karlsruhe, and is part of the Ontobroker project. At Ontobroker, provider agents can register and update their web pages so that these will be considered by Ontocrawler. Ontocrawler takes user queries and responds with answers. Ontocrawler reasons with FLogic (Frame Logic), a formal language to represent frames (classes, attributes, values). Because the ontology is built in Ontolingua, translators need to establish the relation between Ontolingua and FLogic. Ontolingua has been used for "visibility" reasons, and because it provides a hypertext environment enabling easy inspection.

3 Agents involved in (KA)²

There are several agent communities involved for getting the (KA)² initiative started, keeping it going, assuring its scientific content, making it a global collaborative effort and attracting industrial interest: coordinating agents, provider agents, ontopic agents, wise agents and business agents.

3.1 The coordinating agents

Coordinating agents are responsible for the daily matters of the initiative.

3.2 The wise agents

Wise agents are concerned with the scientific issues involved in the initiative.

3.3 The provider agents

Provider agents provide the initiative with instances of the ontology. In other words, they have to annotate their web pages, using the KA ontology. At the kick-off meeting during EKAW'97, the following groups and people committed themselves to be a provider agent. The recruiting agent is responsible for attracting more researchers and groups.

How to become a provider agent

Clearly, this initiative can only succeed through active participation of the community. So we welcome each member of the KA community to join (KA)². Please send an email to the recruiting agent in case you want to join our initiative.

3.4 The ontopic agents

Ontopic agents (from ontology topic) are researchers that contribute to the ontological engineering process to establish a consensual ontology of the KA community. This process is a collaborative effort of the KA community. There are about 15 groups of ontopic agents, each group being responsible for a particular research topic of KA. The following researchers form the ontopic agents group, along with the respective research topics.

How to become an ontopic agent

If you are interested in becoming an ontopic agent, send an email to the managing agent, indicating to what the part of the ontology you want to contribute.

3.5 The business Agents

Business agents are responsible for exploring the possibility of external funding of the initiative and raising the interest of possible interested industries.

4. Agenda of the initiative

In order to get the initiative started, an agenda has been setup with the most urgent things to be done.

4.1 Ongoing activities

To keep the initiative going, several ongoing activities have to be taken care of.

5 Additional information

Pointers that add more background on these ideas are:

• The general idea:
  Dieter Fensel, Michael Erdmann, and Rudi Studer: Ontology Groups: Semantically Enriched Subnets of the WWW. In Proceedings of the International Workshop Intelligent Information Integration during the 21st German Annual Conference on Artificial Intelligence, Freiburg, Germany, September 9-12, 1997. (ftp://ftp.aifb.uni-karlsruhe.de/pub/mike/dfe/paper/www.ps)
• A more detailed overview of the (KA)² initiative:
  Richard Benjamins and Dieter Fensel "Community is Knowledge in (KA)²" (postscript HTML).
• The tools and techniques in more detail:
  Dieter Fensel, Stefan Decker, Michael Erdmann, and Rudi Studer: Ontobroker: Transforming the WWW into a Knowledge Base (Draft only). (postscript.Z, postscript)
• The FLogic version of the Ontology for the Knowledge Acquisition Community:
  Richard Benjamins and Dieter Fensel: A Draft Ontology for Knowledge Acquisition (Draft only) (http://www.aifb.uni-karlsruhe.de/WBS/broker/ka-onto.onto)
• A paper submitted to KAW'98, Banff with a detailed explanation of the KA2 initiative.

Pointers to related work include:

• A related initiative for the logic programming community is proposed by Harld Boley (paper.ps).
• The SHOE approach (Luke, Spector, Rager, Hendler of the University of Maryland).
• XML.
• An overview of other related projects can be found at http://www.aifb.uni-karlsruhe.de/WBS/broker/inhalt-wp.html

Acknowledgments

We would like to thank all participants of the (KA)² initiative. Richard Benjamins is supported by the Netherlands Computer Science Research Foundation with financial support from the Netherlands Organisation for Scientific Research (NWO), and by the European Commission through a Marie Curie Research Grant (TMR).