alkishir/AIULEC: Initial release of AIULEC

This release contains the first version of AIULEC, a learner corpus of written English produced by Libyan Arabic (L1) university students majoring in English at Al-asmarya Islamic University (Spring 2024). Corpus highlights: 272 documents (~20,000 words) from English majors across all semesters Anonymized learner texts written under 60-minute timed conditions without grammar/spell-check tools XML files with embedded metadata and CLAWS C5 POS tags UCLEE error-tagged files for manual error annotation Plain-text (TXT) versions of all documents Combined metadata table (CSV) and PDF versions for teacher-friendly reading Intended use: Learner corpus research SLA studies Error analysis NLP tasks (tagging, classification, grammatical error correction) License: Creative Commons Attribution–NonCommercial 4.0 International (CC BY-NC 4.0) Citation: To be added via Zenodo DOI after publication. For questions or contributions, please open an issue or contact @alkishir on GitHub.