ICE vs GloWbE: Big data and corpus compilation
Dados Bibliográficos
| AUTOR(ES) | |
|---|---|
| AFILIAÇÃO(ÕES) | University of the Balearic Islands Edifici Ramon Llull (Campus UIB), 07122 Palma de Mallorca Spain |
| ANO | 2017 |
| TIPO | Artigo |
| PERIÓDICO | World Englishes |
| ISSN | 0883-2919 |
| E-ISSN | 1467-971X |
| DOI | 10.1111/weng.12281 |
| CITAÇÕES | 7 |
| ADICIONADO EM | 2025-08-18 |
| MD5 |
500f76166f8ee114b381961049f85e79
|
Resumo
ICE and GloWbE are the two main corpora available for the study of world Englishes, and as such they share several features. However, they also differ in a number of respects owing to the different techniques adopted during compilation: while ICE is a traditional corpus with 32 text types represented, GloWbE only contains material from the Internet. This article compares the two corpora from a number of perspectives: (i) frequency and collocation of modal verbs of necessity, (ii) degree of orality and informality of the blogs included in GloWbE versus the spoken component of ICE; and (iii) the vernacular and globalised nature of both corpora. Findings show that big data cannot be considered as an alternative source to corpus compilation. Rather, the new text type(s) found in Internet material should be included in traditional corpora.