Kakyoin03/Health_QA_Darija

该数据集包含翻译并文化适应到摩洛哥达里贾语的医学问答对，专门设计用于微调大型语言模型（LLMs），使其能够充当摩洛哥患者的医疗助手。数据集包含8,132个高质量的医学问答对，每个条目包含以下字段：问题（达里贾语）、上下文问题（达里贾语）、答案（达里贾语）、专业领域（如心脏病学、皮肤病学）、紧急程度（低、中、高）以及提取的临床实体（年龄、症状、疾病、药物）。数据集经过严格的“LLM-as-a-judge”评估框架和自动化NLP指标验证，确保语义和事实完整性。

This dataset contains medical Questions and Answers translated and culturally adapted into Moroccan Darija. It is specifically designed to fine-tune Large Language Models (LLMs) to act as medical assistants for Moroccan patients. The dataset includes 8,132 high-quality medical Q&A pairs, with each item containing fields such as question (Darija), context_question (Darija), answer (Darija), speciality (e.g., Cardiology, Dermatology), urgency (Faible, Moyen, Fort), and entities (age, symptoms, diseases, medications). The dataset underwent rigorous evaluation using an LLM-as-a-judge framework and automated NLP metrics to ensure semantic and factual integrity.