KEYNOTE 1:  Conceptual modeling? When we are awash in information?

November 6th, 16:30 - 17:30

We challenge the traditional who/what/why of conceptual modeling of information in a world where structured data is ubiquitous.
Who (defines conceptual models?) Analysts? Developers? Ontology specialists?   All of the above.  But non-traditional users such as scientists, journalists, educators, and almost anyone with data to share are being empowered to define their own information with easy to use data storage and web management systems.
What (is being modeled?)  A database as part of an information system or software system?  Information that supports a business process?  Definitely.  But some users define their structured information directly – for display and processing.
Why (is a conceptual model defined?)  To describe information and processing of an information system or a software system?  To promote collaboration and communication?  To increase understanding of a domain?  To document a system?  Certainly.  But let’s consider the goals of people who define and publish their own structured information directly; perhaps we can use a conceptual model to offer them useful functionality for their information (e.g., for browsing, mapping, calculations).
We suggest that domain users are doing conceptual modeling.  And we believe that they can relate their conceptual model to a domain model when they are enticed by sophisticated information widgets that can select, display, and process their information.  We also highlight a problem that has been present since conceptual models (or database schemas) were first created: information of interest to a user might be present  in the “data” (such as “Oregon” being part of someone’s address) or in the “schema” such as “Oregon” or “California” being attribute names (for a sport fishing registry).  Finally, we show that users (who understand their own information) can perform schema integration, including complex operations such as pivot and unpivot, when guided with examples (of the widgets) using sample data.
  • Lois

    As a faculty member at The Center for Advanced Computer Studies at the University of Louisiana, Lafayette (in Lafayette, Louisiana), The Oregon Graduate Institute of Science and Engineering, Oregon Health & Science University, and Portland state university (all in Portland, Oregon), she has graduated more than 20 PhD students and published more than 100 papers in refereed conferences and 40 papers in journals. She has served as a principal investigator in more than 50 projects with funding from the US National Science Foundation, the US Air Force, the US National Library of Medicine (of the US National Institutes of Health), the US Defense Advanced Research Agency, and others where the projects were inspired by a broad range of applications including: managing documents in the USDA Forest Service, the design processes for satellite systems, and information in patient medical records; supporting retrospective research over an archive of patient endoscopy reports; improving information retrieval in domain-specific websites (e.g., health information sites in Denmark); and providing what we call semantic widgets for websites that house educational materials. Her research is in the area of database systems, conceptual modeling, digital libraries, and information retrieval.  In 2013, she received the 2013 Portland State University, George C. Hoffman Award for Faculty Excellence for "distinguished contributions to the University in the areas of instruction, university service, and scholarship (in that order) which are done in the spirit of humanism, civility, and collegiality." In 2016, she was named an ER Fellow.

    Lois Delcambre
    Professor Emerita of Computer Science, Portland State University