Newspapers and Geographical Text Analysis: Examples from the UK

Ian Gregory, Lancaster University

One of the major challenges facing Historical Geographers working in the digital era is how to unlock and analyse the geographical information held in large digital textual corpora. Newspaper corpora provide a good example of these – they are frequently contain many hundreds of millions of words of text that both reflected and shaped the public opinion of the day. Within this, there is information on both what and where newspaper editors felt was important to their readers. Geographical Text Analysis is a set of techniques that helps us to understand these issues. It works on the principle that place-names can be automatically identified in a text and allocated to a coordinate. From here, analysis of the words near to the place-name can be used to identify the themes that are associated with that place-name. From here, using a combination of techniques from spatial analysis and corpus linguistics, we can begin to understand the geographies within the text. This paper explores these issues based on a range of examples drawn from digital collections of British newspapers.

No extended abstract or paper available

 Presented in Session 174. Geographies of Qualitative Sources