Science on the Semantic Web: Building the Next Generation of Environmental Information Systems

Rutgers University, October 24, 25

 

 

Background: The term “semantic web” refers to web pages which are understandable by computer programs such as search engines, query agents, data mining software, etc. The prospect of a semantic web holds forth many promises, including the ability to improve information retrieval by using background knowledge and inference; to extract and fuse information from multiple, heterogeneous sources in response to a query; to dynamically compose data and services in response to a query; to track information pedigree and provenance; and to mine data distributed in databases across disciplines.

 

Agenda: This short and focused workshop will look at information systems from a variety of environmental  science disciplines which share the requirements of having to integrate, model, and archive disparate datasets, as well as provide intelligent query capabilities over both the data and the models. Our goal will be to identify problems common to all these systems, and to define how a semantic web approach may overcome these problems. We will also look at barriers to a semantic web approach, and will attempt to elucidate an account of the work that needs to be done to build such systems. To enable us to move quickly into technical discussions, we have decided to focus on a particular technical/research area: the integration of database components and techniques into a semantic web system.

 

The workshop will begin with presentations describing environmental information systems from three organizations of varying size and scope: the National Biological Information Infrastructure (NBII); the Meadowlands Environmental Research Institute (MERI); and the Environmental Protection Agency (EPA).Theses descriptions will discuss data sources, data products, capabilities, and limitations of each system.

This will be followed by discussions of the semantic web vision applied to environmental information systems.

These discussions will have two main themes:

  1. What types of semantic web components can be deployed today in an environmental information system?
  2. What can/should the semantic web learn from research achievements of the database community? For example, it has been suggested that approaches to schema integration are applicable to the semantic web problem of ontology heterogeneity. At the workshop, we will seek to identify specific algorithms/approaches, and consider their relevance to specific problems. Our discussion will begin where the recent NSF workshop on database systems for the semantic web [http://lsdis.cs.uga.edu/SemNSF/ ] left off.

 

 

 

Presentations: In addition to our initial three presentations, we will be seeking presentations which describe a semantic web vision for environmental information systems, or which delve into some aspect of incorporating database research into a semantic web architecture, including schema integration, XML databases, XML wrappers for relational databases, universal schemas, etc.

 

 

Workshop Co-chairs

Nabil Adam (Rutgers) and Yelena Yesha (UMBC)

 

Steering Committee

Vijay Atluri (Rutgers)

Irfan Bora (New Jersey Meadowlands Commission)

Robert Ceberio (New Jersey Meadowlands Commission)

Tim Finin (UMBC)

Susan Hoban (UMBC)

Anupam Joshi (UMBC)

Joel Sachs (UMBC)

Amit Sheth (University of Georgia)

Jim Quinn (University of California Davis)

John Schnase (NASA)

Sue Stendebach (NSF/EPA)

Bhavani Thuraisingham (NSF)