Skip to content

BillLaneCenterStanford/Document-OCR-Quality-Visualization

Repository files navigation

Mapping Texts is a collaboration between the University of North Texas and Stanford University aimed at experimenting with new methods for finding and analyzing meaningful patterns embedded in massive collections of digital newspapers.

Using a collection of 232,500 pages of historical newspapers digitized for the Chronicling America project, we have developed an interactive visualization that allows you to explore both the quality of these digitized newspapers:

Interactive Visualization: Assessing Newspaper Quality
http://mappingtexts.org/quality

This is the source code for this visualization, which links to but does not include the following Javascript libraries:

Google Javascript API: https://www.google.com/jsapi
Google Maps API: http://maps.google.com/maps/api/js?sensor=false
JQuery: http://jquery.com
Simile Timeline: http://www.simile-widgets.org/timeline/
Cufón font substitution script: http://cufon.shoqolate.com 
Protovis Visualization Library: http://vis.stanford.edu/protovis

We have also produced a white paper (http://mappingtexts.org/whitepaper) that details the project, our experiments, and our findings.

For more about this project, see our mission description (http://mappingtexts.org/?page_id=3). This project has been generously supported by the National Endowment for the Humanities.

About

Spatial and temporal browser for assessing text recognition quality in a document collection

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published