Accepted author manuscript, 738 KB, PDF document
Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License
Research output: Contribution to conference - Without ISBN/ISSN › Conference paper › peer-review
Research output: Contribution to conference - Without ISBN/ISSN › Conference paper › peer-review
}
TY - CONF
T1 - The Multilingual Corpus of World’s Constitutions (MCWC)
T2 - The 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation
AU - El-Haj, Mahmoud
AU - Ezzini, Saad
PY - 2024/3/25
Y1 - 2024/3/25
N2 - The “Multilingual Corpus of World’s Constitutions” (MCWC) is a rich resource available in English, Arabic, and Spanish, encompassing constitutions from various nations. This corpus serves as a vital asset for the NLP community, facilitating advanced research in constitutional analysis, machine translation, and cross-lingual legal studies. To ensure comprehensive coverage, for constitutions not originally available in Arabic and Spanish, we employed a fine-tuned state-of-the-art machine translation model. MCWC prepares its data to ensure highquality and minimal noise, while also providing valuable mappings of constitutions to their respective countries and continents, facilitating comparative analysis. Notably, the corpus offers pairwise sentence alignments acrosslanguages, supporting machine translation experiments. We utilise a leading Machine Translation model, fine-tuned on the MCWC to achieve accurate and context-aware translations. Additionally, we introduce an independent Machine Translation model as a comparative baseline. Fine-tuning the model on MCWC improves accuracy, highlighting the significance of such a legal corpus for NLP and Machine Translation. MCWC’s diverse multilingual content and commitment to data quality contribute to advancements in legal text analysis within the NLP community, facilitating exploration of constitutional texts and multilingual data analysis.
AB - The “Multilingual Corpus of World’s Constitutions” (MCWC) is a rich resource available in English, Arabic, and Spanish, encompassing constitutions from various nations. This corpus serves as a vital asset for the NLP community, facilitating advanced research in constitutional analysis, machine translation, and cross-lingual legal studies. To ensure comprehensive coverage, for constitutions not originally available in Arabic and Spanish, we employed a fine-tuned state-of-the-art machine translation model. MCWC prepares its data to ensure highquality and minimal noise, while also providing valuable mappings of constitutions to their respective countries and continents, facilitating comparative analysis. Notably, the corpus offers pairwise sentence alignments acrosslanguages, supporting machine translation experiments. We utilise a leading Machine Translation model, fine-tuned on the MCWC to achieve accurate and context-aware translations. Additionally, we introduce an independent Machine Translation model as a comparative baseline. Fine-tuning the model on MCWC improves accuracy, highlighting the significance of such a legal corpus for NLP and Machine Translation. MCWC’s diverse multilingual content and commitment to data quality contribute to advancements in legal text analysis within the NLP community, facilitating exploration of constitutional texts and multilingual data analysis.
KW - Constitutions
KW - Corpus
KW - Fine-tuning
KW - Machine Translation
KW - Legal Documents
M3 - Conference paper
Y2 - 20 May 2024 through 25 May 2024
ER -