The Tagged Icelandic Corpus (MÍM)

In this paper, we describe the development of a morphosyntactically tagged corpus of Icelandic, the MÍM corpus. The corpus consists of about 25 million tokens of contemporary Icelandic texts collected from varied sources during the years 2006–2010. The corpus is intended for use in Language Technology projects and for linguistic research. We describe briefly other Icelandic corpora and how they differ from the MÍM corpus. We describe the text selection and collection for MÍM, both for written and spoken text, and how metadata was created. Furthermore, copyright issues are discussed and how permission clearance was obtained for texts from different sources. Text cleaning and annotation phases are also described. The corpus is available for search through a web interface and for download in TEI-conformant XML format. Examples are given of the use of the corpus and some spin-offs of the corpus project are described. We believe that the care with which we secured copyright clearance for the texts will make the corpus a valuable resource for Icelandic Language Technology projects. We hope that our work will inspire those wishing to develop similar resources for less-resourced