menu 1
menu 2
menu 3
menu 4

dg.o Web

DG Researchers Encode the Shipping Lanes
A joint U.S. Coast Guard / Arizona State AI Lab project is developing a language to smooth traffic on the world's waterways
For the DGRC

Navigation Ontology Project

" Project home page
" USCG Intelligent Waterways
" Vocabulary Development for Markup Languages
" Maritime Transportation and the Semantic Web
" W3C Semantic Web home

For thousands of years ships and boats have sailed the oceans.

But how exactly do you define a "boat" versus a "ship" in a way that's consistent for mariners across the United States and the rest of the world? And once you've got that down, isn't there a better, faster way in the 21st century to handle the ship-to-shore transfer of regulatory documents than faxes?

Kathleen Shea, a Computer Specialist and Project Manager with the Coast Guard's Research and Development Center had been considering those very sorts of questions when she was approached by Raphael Malyankar, a Digital Government researcher from Arizona State University's Artificial Intelligence Laboratory.

Malyankar was interested in applying artificial intelligence research to navigation in the hopes of eventually allowing automated systems to make better use of geospatial information. After talking, the two realized they could work together to realize Shea's dream project: MIML, or Maritime Information Markup Language, which would be an essential component of Malyankar's ultimate goal.

"If a buoy gets hit and is knocked off station," says Shea, "the information needs to get out to mariners." Using MIML to make such vital information available over the Internet would let it be transferred much faster - without the mistakes introduced by illegible faxes or radio static. In addition, she says, "Commercial ships are required to announce upon arrival into port, cargo and crew info. MIML could lead ideally to web-based postings of information, rather than faxes."

MIML is a markup language based on XML (eXtensible Markup Language), which is a way of structuring information in documents so that computers can process it easily. The most popular and widely-known of markup language is HTML (Hypertext Markup Language), used to create most Web pages.

MIML would let a computer know that should be displayed to the reader as a particular kind of image, rather than the word "chart."

But creating a mark-up language from scratch requires years of work to classify thousands of terms. It is a job that's part computer science and part linguistics.

In order to capture every term correctly, Malyankar and Shea combine the standard working documents of maritime life, like charts, maps, weather reports with workshops that include, in Shea's words, "Every sector of maritime—owners, operators, agents, cargo handlers/shippers. We are bringing in 14 different categories to our next workshop. It will be a vertical slice, including state port authority people as well as federal government officials."

In these workshops the researchers try to determine which information is absolutely known by each category. That category will then be identified as the authoritative source for that set of information. Then they repeat the workshops in different locations; "You do it in different geographical places, extremes like New York, Miami, Houston, and Seattle, and see if you get the same kind of answers," says Shea. "You try to anticipate every kind of complication you could get."

Latest DG News

" dg.o2005 Convenes May 16-18 in Atlanta
" dg.o2005 Issues Call for Papers
" IEEE ISI Issues Call for Papers
" DG Study Calibrates Mediation
" Swapping Secrets of the Double Helix
" FirstGov BUZZ Reports on Tsunami Relief Work
" UK and DO-Wire Launch e-Gov Best Practices wiki
" DG Team Develops "Virtual Agora" for e-Gov
" Mapping for Times of Crisis
" Exploring Detection of Crisis Hotspots
" Report: Mass eMail Campaigns Harmful
" Scenario-Based Designs for Stat Studies
" US, EU Explore Info Integration
" DG Team Develops Digital Interpreter
" DG Study Gives Teeth to FBI
" Research Smooths Road for Small Businesses
" DG Researchers Parsing in Tongues
" e-Gov Journal Issus Call for Articles

" See all news stories

Contribute to dgOnline

Shea and Malyankar then split the work into an overlapping process, where her side concentrates on individual words and terms, while his side creates ontologies (a group of defined terms and concepts and their relationships) and a taxonomy for the whole field. This is being done in anticipation of MIML's eventual absorption into the Semantic Web, a metadata scheme that lets computers integrate and "understand" information on the Web, instead of merely displaying it.

"The technologies we're using are really being actively developed in the commercial world. We are following the Navy's approach to the technology which is the basis for the current federal government approach," says Shea. She and Malyankar are working on an interface to the Navy's XML registry. In addition, she has already been invited to attend meetings of the European Union's Marine XML project.

Highly concerned with synching up terms and data, the resesarchers have also learned a little bit about timing: "Academic schedules don't merge with industry and government's," says Malyankar. "There are few real-world holidays in May, but academia is engrossed in end-of-term projects, examinations, commencement, and so on," he says.

But as an academic researcher, Malyankar sees several practical benefits coming from their work together: "We learned from them that vocabulary mapping is important," he says. "For example, when we say ‘model' it's a very rigorous description of a domain; for them it's a slightly less formal description of vocabulary items."

For Malyankar's students, it was a chance to get, "exposure to a real-world software development process," he says. "The students learned it was important to write production quality software, not just produce something and get a grade but something that is robust and works."

For Shea too, the collaboration has proved especially beneficial, "As a person running a project with a limited budget, I got a lot more work than I ever paid for," she says. "We sponsored his work and first-hand just provided our time. Later we gave money because he knew the information sources we were going to be using to create the vocabulary. From our perspective, it was free to have someone come up to speed in our language."