Interface 2003

Interfaces to a Statistical Knowledge Network
Gary Marchionini, (University of North Carolina at Chapel Hill),


As a result of national and local government efforts to leverage WWW technology, the public has increasingly better access to government statistical data. Two significant challenges to promoting appropriate usage of this data are to make the data understandable by people with diverse interests and statistical expertise, and to integrate relevant data from different sources and agencies. As a first step toward a statistical knowledge network that supports access and use, we focus on creating and testing human-computer interfaces that serve to integrate different data sources and provide on-demand help. In this talk, examples of highly interactive interfaces to find US federal agency data and understand statistical concepts relevant to that data will be discussed. These interfaces depend on well-structured underlying metadata and our approaches to integrating cross-agency data underneath the interface will also be discussed.

Take me back to the main conference page.