Home Research publications academics about members partnerships contact

OZONE (Zoomable Ontology Navigator)


Project description:

OZONE is a query interface for retrieving knowledge from documents augmented with semantic information which is written in DAML (DARPA Agent Markup Language).

Since most web pages are written in HTML and they are intended for human consumption, it is very hard to automate any task which one has to perform on the Web. For example, the most common way to extract knowledge from the web is to use search engines which is based on string match. But sometimes the result from keyword searching is not necessarily useful - There could be no hits or too many hits. A possible solution for the search problem - and for the general issue of letting automated "agents" roam the web  - is to provide a mechanism which allows a more precise description of things on the web. This, in turn, could elevate the status of the web from machine-readable to so-called machine-understandable. DAML provides a rich set of constructs with which to create ontologies and to markup information so that it is machine readable and understandable. Practically, a small piece of XML data is augmented inside each web page to let "agents" or "machines" understand the precise meanings of the document. The DAML language is being developed as an extension to XML and the Resource Description Framework (RDF).

While this framework can enable intelligent services such as search brokers and search agents, it causes another problem. It is very hard for humans to formulate a valid query for retrieving information. The ontolgy, schematic information for XML data inside web pages, lacks structure (or is semi-structured) and is scattered over the web, which prevents humans from taking advantage of the semantic information. OZONE gathers ontolgy information and provides users with an interface which makes them feel like they are using relational databases even though its underlying data model is totally different. OZONE is also designed to facilitate interactive and incremental query formulation. Users are provided with easy-to-understand graphical representations of ontology artifacts so that they can interact with visual objects to specify query conditions. Also users can check intermediate results to see if they are on the right track.

OZONE is built using HCIL's Jazz Toolkit for Zoomable User Interfaces.


Bongwon Suh, Research Assistant of HCIL, Graduate Student of Computer Science, University of Maryland
Ben Bederson, Director of HCIL, Assistant Professor of Computer Science, University of Maryland


Suh, Bongwon and Bederson, Benjamin, B. (2001) "OZONE: A Zoomable Interface for Navigating Ontology", HCIL Tech Report #2001-04, University of Maryland, College Park, MD 20742.


Currently, OZONE is not available for downloading.

If you would like to receive information about updates on OZONE, or are willing to help us evaluate OZONE with an occasional survey, please send your email address to sbw@cs.umd.edu.

Privacy Statement: This information will be used only by the authors to monitor the use of OZONE and possibly to contact users for follow-up evaluation and feedback. We will not provide any personal information to anyone outside this project. Statistics about this site will contain no traceable personal information.

Related Sites

W3C Semantic web activity: Resource Description Framework (RDF)

DAML (DARPA Agent Markup Language)

The semantic web in Scientific American (May 2001) by Tim Berners-Lee, James Hendler and Ora Lasilla

SHOE (Simple HTML Ontology Extensions)

Last updated June 25, 2001
ęCopyright University of Maryland

Web Accessibility