Metadata design for the first electronic learner corpus of Romanian

Abstract

This paper introduces an ongoing work of collecting, annotating and documenting the first digital Romanian Learner Corpus (LECOR), focusing on its metadata. We shortly describe the institutional context of the project, the current state of the art in the field, the objectives in terms of structure, dimensions and annotations and what work has already been done at this stage of the project. Then we present the modular structure of the metadata scheme and a detailed  account of all the metadata fields and their possible values, from general metadata concerning the whole corpus (Section 3.1), to metadata organised around the student/learner (Section 3.2) and text/composition (Section 3.3). We
will also give some examples of how metadata has been dealt with in various researches (including based on LECOR corpus).