Center for Digital Research in the Humanities

 

ORCID IDs

Elizabeth Lorang

Date of this Version

5-2019

Document Type

Article

Citation

Lorang, Elizabeth, Yi Liu, Chulwoo Pack, Leen-Kiat, Delaram Rahimighazikalayeh, and John O'Brien, "Application of the Image Analysis for Archival Discovery Team’s First- Generation Methods and Software to the Burney Collection of British Newspapers." May 2019.

Abstract

The current study, “Application of the Image Analysis for Archival Discovery Team’s First- Generation Methods and Software to the Burney Collection of British Newspapers,” is the first test of our approaches—methods and software—to a different newspaper corpus, specifically the 17th and 18 Century Burney Newspapers Collection. This study stands as the first complete attempt at applying Aida’s software and methods to non-Chronicling America newspapers, as a step toward understanding the potential of our approaches across digitized historic newspapers. In taking this step, our goals were (1) to test how well the software and a classifier model developed on Chronicling America newspapers performed on newspapers from a different corpus, a corpus that represents both a different geographical region and time period as well as newspapers digitized at an early stage in newspaper digitization history; (2) to explore whether classification results would be improved by training a new classifier model on Burney Collection images. Overall, we sought to explore how robust and extensible the first-generation Aida approach is and to better understand which parts of our methods might be brought over to new corpora “as is,” and which may need to be calibrated for specific contexts.

Share

COinS