Big data for the analysis of translation demand in five European capitals

Abstract

The amount of digital global information available is exploding. Huge quantities of unstructured information is being generated by big companies, such as Google, Facebook, Microsoft… The term big data refers to these huge datasets that cannot be managed and analysed by traditional tools and software in a tolerable time. This research aims to show the utility of big data as a source of information for linguistic research, which enables us to carry out investigations that could not be done in a traditional way. In this paper, we use the information available in the microblogging service Twitter to analyse the most spoken languages in five European capitals, in order to know the demands for translation jobs. Thus, we prove the utility of big data as a tool for linguistic research. We have studied the total number of tweets generated in Berlin, Brussels, Paris, Madrid and London, during 21 August 2015 and 21 September 2015. For this purpose, we have developed an authoring tool through which we obtain, stored, processed and analysed the information. Our objective is to obtain results based on millions of data in real time, which saves not only time, but also costs in the research process and allows us to know the languages used anywhere and anytime.