February 19, 1995, Los Angeles, California
1:30 - 5:00 pm in Hoffman Hall 401
(Note: this is on the University of Southern California Campus
NOT in the Hotel)
in conjunction with The Eleventh IEEE Conference on Artificial Intelligence for Applications
Workshop Co-chairs: Carol Brown and Alan Sangster
Some present and future uses of World Wide Web by Alan Sangster
Presentation by: Alan Sangster
, University of Aberdeen, Department of Accountancy
Email: (a.sangster@abdn.ac.uk)
Abstract: World Wide Web contains an enormous amount of information. Some is useful in the general sense, some is only useful to a few, some is not useful at all. Accessing World Wide Web is like having dozens of libraries and art galleries at your fingertips - a vast repository, but what can you do with it? This web-paper presents a brief overview of some of ways in which World Wide Web is being used, of some of the types of information available on World Wide Web, and some pointers on how it can be explored without getting buried in a deluge of information.
Facilitating the Utilization of Intelligent Decision Support Systems for Polymer Composites Design with Distributed Hypermedia Documentation by Timothy Lenz, James K McDowell, Jon Sticklen and Martin C. Hawley
Presentation by: Timothy Lenz
Composite Materials & Structures Center, Michigan State
University
Email: (lenz@m31.cps.msu.edu)
Abstract: Researchers in the Intelligent Systems Laboratory at Michigan State University have developed intelligent decision support systems for the design of polymer composite materials as part of an NSF-funded Cooperative Research Center for Low-Cost, High Speed Polymer Composites Processing. One of the main responsibilities of the researchers, apart from actually developing the systems, continues to be the distribution of these systems to member companies of the NSF Center. As these member companies are scattered through several states, an effort was undertaken to capitalize on the multimedia documentation capabilities of the World-Wide Web. As a result, a preliminary multimedia, hyper-indexed distribution of these systems has been established on the Internet. This distribution includes both the decision support systems as well as accompanying documentation. This paper discusses of the details of this preliminary distribution and the methods utilized to make this documentation available on the Web.
Database access in WWW
by Norman Chen (jiachen@pollux.usc.edu)
Abstract: Currently, I work on database access in WWW. Current WWW is based on hypertext architecture, where the access unit is a document (i.e. a file). My research extends the access unit to a record or an object in a database. I propose a "virtual" document, which is active, dynamic, and interactive. 1) Active. An active document can use hyperlink to execute an application program or a database manager. 2) Dynamic. A dynamic document is generated in run-time to transmit the queried data. 3) Interactive. An interactive document presents a user interface, say a form, to collect the necessary information to construct a query. The data model used by a database system and the schema information can be hidden for a browser user. This promotes the database transparency. This approach brings new development on both the WWW servers and browsers, but leaves the protocol intact.
Applying machine learning to help users browse the web - Work in Progress by Stan Matwin, Rob Holte, Denys Duchier and Chris Drummond
Presentation by: Stan Matwin, Department of Computer Science,
University of Ottawa
Email: (stan@csi.uottawa.ca)
Abstract: Many software tools and products are used by people with a wide variety of demands and patterns of use. Off-the-shelf software cannot be optimal for all these users. At present it is usually adapted to best meet the needs of a hypothetical typical user. When users are varied or their needs and usage patterns change with time, this adaptation process does not meet the needs of many and may in fact diminish most users' productivity and quality of work.
We propose to develop a novel form of software adaptation through learning. Agent-based extensions dynamically adapts the software environment in response to an individual's actions and circumstances. A "learning apprentice" learns how to carry out a particular task by observing user interaction with the software. If the apprentice then notices a situation where the task has been successfully performed before it can suggest doing it. Because it learns the system adjusts to changes in the user's behavior or situation. The cost of these frequent adaptations will be kept low by making it as automatic as possible. Machine learning techniques can help accomplish this.
Information Technology Research Institutes
Presentation by: Kazuyoshi Honda, Science University of Tokyo
& Tokyo Gas Co., Ltd. Information Technology Research Institutes
Enhancing Classes with World Wide Web by Carol E. Brown, Norma L. Nielson and David Sullivan
Presentation by: Carol E. Brown , College of Business, Oregon
State University
Amelia Baldwin-Morgan, Eastern Michigan University
France Belanger, Ph. D Student, ISDS Department, University of
South Florida
Yoshifumi Ooyama, NTT Information and Communication Systems Laboratories
Alan Wilson, University of Strathclyde
Prepared and maintained by:
Email: (honda@iti.tokyo-gas.co.jp)
Abstract: A brief introduction to Mizoguchi's Lab.
Email: (brownc@bus.orst.edu)
Abstract: Material presented on this page and the references herein
summarize the ways in which World Wide Web resources are being
used in classes at Oregon State University. The work summarizes
the use of the Web 1) to provide information about classes to
students, 2) to provide required textbook readings, cases, and
other required material to students, 3) to present work prepared
by students, and 4) to facilitate a peer review process between
students. Future possibilities to use the Web as a testing tool
also is discussed.
Other Participants
Email: (Amelia.Baldwin-Morgan@emich.edu)
Email: fbelange@cfrvm.cfr.usf.edu
Email: ooyama@nttnly.isl.ntt.jp
Email: (alanw@account.strath.ac.uk)
Carol E. Brown
(brownc@bus.orst.edu)
and
Alan Sangster
(a.sangster@abdn.ac.uk)
Page last updated February 16th, 1995.