The Year of the Newspaper

Text:
Increase font size
Decrease font size

Led by postdoctoral fellow Ashley Reed, this year the DIL is exploring how scholars, graduate students, teachers, and undergraduate students might use the unique collection of digitized North Carolina newspapers now available through the UNC Library.

In 2014, a partnership between UNC and Newspapers.com resulted in the digitization of some 3.2 million pages of pre-nc newspaper example1923 local newspapers held by the UNC Library’s North Carolina Collection.  Keyword search access to all these newspapers is now available through the UNC Library website.  The 3.2 million digital files produced as a part of this process will also become available for text mining, topic modeling, and other “big data” applications.

On Tuesday, March 24, 2015, the UNC Libraries, the Digital Innovation Lab, and the Odum Institute for Social Science Research will host a symposium in the Pleasants Family Assembly Room at Wilson Library to spread the word about the collection and its potential applications for research, teaching, and engaged scholarship. Faculty and graduate students from across the UNC campus are invited to attend; more information can be found here.

A sub-set of the files has been loaded into a “sandbox” environment in which text mining and topic modeling approaches can be applied.  A working group of DIL staff, faculty, and library staff have been meeting regularly since October to ask: “what new questions can we ask if we have access to millions of newspaper pages?”   “What tools can be used?”  “What new tools and approaches need to be developed?”

The sandbox sample includes all N.C. newspapers published during and immediately after World War I.  The working group has tracked the coverage of the influenza epidemic in North Carolina across more than one hundred local newspapers.

For the past year, Robert Allen has been using the newspaper collection in his undergraduate teaching.  Students in his spring ’15 first year seminar on family history (AMST 53H) are using the newspapers from the 1860s and 1870s to “find” African-American family members who were separated under slavery.