السلام عليكم ورحمة الله وبركاته،
يسعدني أن أشارككم الإعلان عن إطلاق مدونتنا الجديدة "مُراد MURAD"، وهي – بحسب علمنا – أول مجموعة بيانات عربية تجمع ما يقارب 100 تعريف متخصص من مجالات متعددة تشمل العلوم الطبيعية، والعلوم اللسانية، والعلوم الحوسبية.
تم تطوير هذه المدونة على مدار عامين من العمل البحثي في مجال القاموس العكسي (Reverse Dictionary) ونماذجه، وتهدف إلى دعم أبحاث معالجة اللغة العربية، لا سيما في مهام فهم التعريفات، النمذجة الدلالية، والاسترجاع المعجمي.
🔹 رابط الورقة البحثية:
https://arxiv.org/pdf/2601.21512
🔹 رابط المدونة على Hugging Face:
https://huggingface.co/datasets/riotu-lab/MURAD
نرحب بأي ملاحظات أو اقتراحات تطوير، ونتطلع إلى أن تكون هذه المدونة إضافة مفيدة لمجتمع البحث في معالجة اللغة العربية.
مع خالص التحية،
سري السباعي
Dear colleagues,
I am pleased to share the release of our new dataset, MURAD, which—to the best of our knowledge—is the first Arabic dataset compiling around 100 specialized definitions spanning multiple domains, including natural sciences, linguistics, and computational sciences.
This dataset is the outcome of two years of research in reverse dictionary modeling and related approaches. It is designed to support Arabic NLP research, particularly in tasks related to definition modeling, semantic representation, and lexical retrieval.
🔹 Paper link:
https://arxiv.org/pdf/2601.21512
🔹 Dataset on Hugging Face:
https://huggingface.co/datasets/riotu-lab/MURAD
We welcome feedback, suggestions, and potential collaborations, and we hope this resource will be valuable to the Arabic NLP research community.
Best regards,
Serry Sibaee