The ASK corpus - a learner corpus of Norwegian as a second language; design annotation and search interface

In my talk I will present the ASK corpus which contains 1739 texts written in Norwegian as a second language and personal data about the learners. The texts are written essays from two different tests measuring language performance at two different levels (supposed to be at or above level B1 or B2). A reassessment of the texts according to CEFR level descriptions, were performed in 2009. The texts and personal data are marked up in XML according to the TEI Guidelines. Error coding is done manually using a relatively simple system developed for ASK. To compensate for this simple system, the texts are grammatically tagged using an automatic tagger developed for standard Norwegian, “The Oslo-Bergen tagger”. The latest version of the ASK corpus is accessible in a newly designed and implemented corpus management platform (Corpuscle).