Note that these classifications have been done automatically using various algorithms and techniques under development at UNC. Moreover, they are applied to samples of the various agency sites rather than all pages in these sites. The classification labels and placement of pages in various classes are thus estimates for the purposes of demonstrating feasibility.

Following four demos allow you explore topics and geographical coverages of various web sites. Result table shows titles, and URLs of result web pages.

EIA demo (link:http://idl53.ils.unc.edu/~junliang/rb_eia_new.html)
The demo collected about 8000 web pages from Energy Information Administration web site.

SSA demo (link:http://idl53.ils.unc.edu/~junliang/rb_ssa.html)
The demo collected about 15000 web pages from Social Security Administration web site.

NCHS demo (link:http://idl53.ils.unc.edu/~junliang/rb_nchs.html)
The demo collected about 4000 web pages from National Center for Health Statistics web site.

NASS demo (link:http://idl53.ils.unc.edu/~junliang/rb_nass.html)
The demo collected about 4000 web pages from National Agricultural Statistics Service web site.

Old demos

Fedstats demo ( link: http://idl53.ils.unc.edu/~junliang/rb_fedstats.html )
This demo collected about 9000 webpage from vaious government agencies linked from Fedstats website. It allows you to explore three facets of data : Topics, Geograpical coverage, and Agencies . Result table shows titles, and urls of result web pages.

Census demo ( link: http://idl53.ils.unc.edu/~junliang/rb_census.html )
This demo collected a portion of(about 10000 pages) web pages from Bureau of Census web site. It allows you to explore three facets of data : Topics, Geograpical coverage, and document types . Result table shows titles, and urls of result web pages.

BLS demo ( link: http://idl53.ils.unc.edu/~junliang/rb_bls.html )
This demo collected about 15000 web pages from Bureau Labor Statistical web site. It allows you to explore main topic and secondary topic categories. Urls of the web pages are displayed in the result table.

EIA demo ( link: http://idl53.ils.unc.edu/~junliang/rb_eia.html )
This demo collected about 12000 web pages from Energy Information Administration web site. It allows you to explore four facets of data : Fuel types, Geography, sectors, and processes. Result table shows titles, page sizes and descriptions of result web pages.

Quick help on using the RB++:
-----Mouse over the bars to explore !
-----Click on search button to retrieve.
-----Type in keyword in text boxes to filter results.

NOTES:

1. Java plug in 1.4.1 or up is required.

Back to Govstats website More examples on RAVE website