Associate professor of the newly established Science Integration Programme at the University of Tokyo in Tokyo, Japan.
Internet-Based Integrated Environmental Assessment, Part II: Semantic Searching Based on Ontologies and Agent Systems for Knowledge Discovery
Version of Record online: 8 FEB 2008
Journal of Industrial Ecology
Volume 10, Issue 4, pages 37–60, October 2006
How to Cite
Kraines, S., Batres, R., Kemper, B., Koyama, M. and Wolowski, V. (2006), Internet-Based Integrated Environmental Assessment, Part II: Semantic Searching Based on Ontologies and Agent Systems for Knowledge Discovery. Journal of Industrial Ecology, 10: 37–60. doi: 10.1162/jiec.2006.10.4.37
- Issue online: 8 FEB 2008
- Version of Record online: 8 FEB 2008
- industrial ecology;
Integrated environmental assessment of technologies and processes associated with industrial ecology requires the sharing of expert knowledge from a wide range of domains of research. A concept for an Internet-based platform that uses Internet and communication technologies to support knowledge sharing in the form of numerical model computational services for solving complex problems related to large-scale systems such as urban ecosystems has been proposed. Here, a software prototype that has been developed for sharing and discovery of expert knowledge from domains related to industrial ecology is described. The prototype uses the combination of a generic upper level ontology and a specific domain level ontology to provide a semantic layer for the process of searching and matching knowledge requests with knowledge advertisements. Knowledge advertisements are provided in the form of semantic descriptions of computational model service capabilities. The prototype uses a multiagent system to support the realization of a distributed knowledge-sharing marketplace on the Internet, populated by agents that advertise model services, agents that make requests on behalf of system users, agents that supply ontologies, agents that conduct searches on behalf of other agents, and agents that manage address and function repositories for other agents. A proof of concept demonstration of the prototype is provided, using a hypothetical scenario of four researchers around the world who have knowledge that could aid each other's research efforts.