Building the Great Recession News Corpus (GRNC): A contemporary diachronic corpus of economy news in English

Abstract

The paper describes the process involved in developing the Great Recession News Corpus (GRNC), a specialized web corpus which contains a wide range of written texts obtained from the business section of The Guardian and The New York Times between 2007 and 2015. The corpus was compiled as the main resource in a sentiment analysis project on the economic/financial domain. A justification of the corpus design is provided, along with the methodology followed for the compilation process. To evaluate its usefulness, we include a sentiment analysis study on the evolution of the sentiment conveyed by the word credit during the years of the Great Recession.