Moroccorp: Tien miljoen woorden uit twee Marokkaans-Nederlandse chatkanalen

Abstract

In this article we introduce a new corpus of computer-mediated communication in Dutch by Maroccan-Dutch language users, consisting of ten million words of chat material. We treat the background, the compilation method and the inner structure of the corpus, and we relate our efforts to previous attempts to build a corpus of Dutch chat language. We also conducted a Stable Lexical Marker analysis and a case study on a well-known morphosyntactic feature of Moroccan Dutch to assess in this manner the representativity of the corpus.