Retrieving the specialised substance from a corpus of professional discourse in the field of Records and Information Management

Abstract

Corpus size is a clear indicator of the sort of professional data that the corpus contains. Corpora representing front stage activities (i.e. activities that are easily accessible such as CEO's letters to shareholders or annual reports) may reach several tens or hundreds of thousands of words as is the case with Corpus of Environmental Impact Assessment (Flowerdew 2008) or Cho and Yoon's earning calls corpus (Cho & Yoon 2013), both containing 250,000 words. Much larger corpora of frontstage discourse have been developed but they tend to be the exception rather than the norm. The Wolverhampton Corpus of Written Business English (Fuertes-Olivera 2007; Cheng 2017) comprises more than ten million words and contains a fairly wide variety of genres including product descriptions, company press releases, annual financial reports, business journals, academic research papers, political speeches and government reports, all dealing with business or professional topics. On the other hand, corpora representing backstage activities (i.e. activities that are not easily accessible such as business conversations or professional emails) are typically small. Handford and Matous' corpus of oral interactions on construction sites contains 12,000 words (Handford & Matous 2011). Gillaerts' corpus of organisational emails hardly exceeds 14,000 words (Gillaerts 2012). Koester's Corpus of American and British Office Talk (ABOT) contains 34,000 words (Koester 2010) which is very similar in size to Millot's corpus of professional emails which reaches 33,000 words (Millot 2017).