MARCH/APRIL 2003 (Vol. 18, No. 2) pp. 16-17
1541-1672/03/$31.00 © 2003 IEEE
Published by the IEEE Computer Society
Published by the IEEE Computer Society
Guest Editors' Introduction: IFIP Conference on Intelligent Information Processing
PDFs Require Adobe Acrobat
The International Federation for Information Processing is an umbrella organization for national societies working in this field. IFIP was established under the auspices of the United Nations Educational, Scientific, and Cultural Organization in 1960. It continues to maintain close ties to the UN and to nongovernmental computer-science-related organizations. IFIP, as an "organization of organizations," helps disseminate information technology knowledge internationally, in both the developed and developing world. Through its technical committees and associated working groups, IFIP fosters research in all areas of computer science. Every other year, it organizes the World Computer Congress, which is the oldest and best-known international conference featuring the entire spectrum of work in information processing. At a time when computer science is becoming increasingly specialized, IFIP is unusual in its attempt to address and bring together broad areas of interest within a diverse, international community.
The most recent IFIP World Computer Congress was held in Montreal in August 2002. Within WCC 2002 was a conference on intelligent information processing. IIP 2002 was organized by IFIP Working Group 12.5—dedicated to "knowledge-oriented development of applications"—of Technical Committee 12 on artificial intelligence. The group was founded to bring together an international community of scientists concerned with using knowledge-based techniques in real-world software system engineering.
IIP 2002 emphasized discussions of knowledge-based system architectures and intelligent information management. Invited talks and panels highlighted important new topics, such as the use of ontologies to support knowledge-based applications, the emergence of knowledge-based techniques on the Web, agent-oriented architectures, and fundamental questions of knowledge representation. The complete Proceedings of IIP 2002 are available as a bound volume from Kluwer.
This special issue of IEEE Intelligent Systems presents expanded versions of the four papers that received the highest ratings from the IIP 2002 scientific program committee. To prepare for this issue, we also asked authors to respond to an additional round of peer review. The articles provide a perspective on what was an extremely stimulating "meeting within a meeting" that emphasized research on pragmatic aspects of AI.
Interestingly, despite the considerable heterogeneity of the World Computer Congress, the four articles share common themes and a common world view. Each article emphasizes in a different way the importance of representing domain knowledge using explicit ontologies. Each submission almost takes for granted the core notion of defining precise concepts and relationships among concepts as the central element in building intelligent application systems. A theme in these articles—which was echoed throughout the conference—is that domain knowledge must be available online; the next generation of knowledge-system architectures must be able to accommodate distributed knowledge bases, distributed problem solvers, and the heterogeneity in both representational choices and ontological distinctions that is inevitable with the expanding World Wide Web.
In "Identifying Communities of Practice through Ontology Network Analysis," Harith Alani, Srinandan Dasmahapatra, Kieron O'Hara, and Nigel Shadbolt describe a tool called Ontocopi. Ontocopi incorporates an algorithm for ontology-based network analysis to extract relationships among entities that—in the case of the particular ontologies studied—can identify communities of practice (informal groups of individuals who interact, sharing common work practices and interests). The article defines a formal, ontology-based algorithm for a task that is traditionally performed through relatively informal ethnographic analysis.
Alexander Maedche, Boris Motik, Ljiljana Stojanovic, Rudi Studer, and Raphael Volz concentrate on ontology management issues in "Ontologies for Enterprise Knowledge Management." They present an architecture for managing ontology evolution, which will be an essential feature of the next generation of ontology development tools. The Semantic Web promises to strain our ability to deal with problems of ontology change and ontology mapping; the authors' Ontologging system offers a step toward addressing these significant challenges.
"Configuring Online Problem-Solving Resources with the Internet Reasoning Service," by Monica Crubézy, Enrico Motta, Wenjin Lu, and Mark A. Musen, concentrates on issues of problem solving. The article discusses how the Web can provide a vehicle for executing reusable problem-solving methods and how such methods can implement various Web services. The use of reusable problem-solving methods has been an important aspect of work on knowledge-based systems for more than a decade. The Internet Reasoning Service this article describes now lets such reusable problem solvers interoperate in cyberspace.
"Using JessTab to Integrate Protégé and Jess," by Henrik Eriksson, was one of the excellent applications papers presented at the conference. Eriksson describes the integration of support for the popular Jess rule-based system shell with the Protégé-2000 ontology and knowledge base development environment. The article describes the challenges of combining the two systems. It also presents a framework for building intelligent systems that hundreds of users now employ regularly for a wide range of application areas.
In general, IIP 2002 discussions emphasized rapidly evolving topics such as the importance of ontology design and management in constructing intelligent systems, the promise of the Semantic Web and the role of AI in supporting distributed intelligence over the Internet, and the opportunity for machine learning and information extraction that the burgeoning amount of Web content offers. There was a tremendous sense at the conference that the advent of the Web is doing much to change AI practice and the future of intelligent systems. Uniform excitement existed about the new directions in which pragmatic work in AI is headed as the Web creates new content and new infrastructure on which to build the next generation of intelligent application systems.
As the conference and program cochairs of IIP 2002, we are delighted to share this sampling of the conference with the readership of IEEE Intelligent Systems.
Mark A. Musen is a professor of medicine (medical informatics) and computer science at Stanford University and is head of the Stanford Medical Informatics laboratory. His research interests include knowledge acquisition for intelligent systems, knowledge system architecture, and medical decision support. He has directed the Protégé project since its inception in 1986, emphasizing the use of explicit ontologies and reusable problem-solving methods to build robust knowledge-based systems. He has an MD from Brown University and a PhD from Stanford. Contact him at Stanford Medical Informatics, 251 Campus Dr., Stanford Univ., Stanford, CA 94305; firstname.lastname@example.org; www.smi.stanford.edu/people/musen.
Bernd Neumann is a full professor in informatics, head of the Artificial Intelligence Laboratory, and chair of the Hamburg Informatics Technology Center at Hamburg University's Department for Informatics. His research interests include image understanding, knowledge management, and knowledge-based systems for diagnosis and configuration. He received his diploma in electrical engineering from the Technical University of Darmstadt, and his SM and PhD from MIT. He is chair of IFIP's Technical Committee on Artificial Intelligence, a member of the AAAI, the ACM, the IEEE, and the German informatics society (GI). Contact him at email@example.com.
Rudi Studer is a full professor in applied informatics at the University of Karlsruhe's Institute of Applied Informatics and Formal Description Methods), the director of the Knowledge Management Group at the FZI (Research Center for Information Technologies), University of Karlsruhe, and a member of the L3S Learning Lab in Hannover, Germany. His research interests include knowledge management, Semantic Web technologies and applications, ontology engineering, knowledge discovery, and e-learning. He received his Habilitation in computer science and his PhD in mathematics and computer science, both from the University of Stuttgart. He is on the editorial board of IEEE Intelligent Systems and is a member of the AAAI, ACM, IEEE, and the German informatics society (GI). Contact him at Institute AIFB, Univ. of Karlsruhe, 76128 Karlsruhe, Englerstr. 11, 76128 Karlsruhe, Germany; firstname.lastname@example.org; www.aifb.uni-karlsruhe.de/WBS.