What to Do With a Million Pages of Digitized Historical Newspapers?

Abstract

Newspapers are rich sources of evidence of history and literally millions of pages of historical newspapers have now been digitized. We aim to develop tools for effectively browsing this rich resource. However, the structure of newspapers is highly complex and a complete analysis will involve many interacting components. We demonstrate two approaches for extracting advertisements from other types of material in the newspaper. We also describe preliminary results of interviews with historians about features which they would find particularly useful for conducting research on collections of digitized historical newspapers.