MeDAL

该数据集名为MeDAL，包含了超过1000个独特标签的医疗缩写。该数据集主要用于评估标记分类模型的性能，尤其是在SciBERT模型上的应用，该模型取得了77.29%的宏观F1分数。数据集规模较大，其任务是对医疗缩写进行消歧。

Named MeDAL, this dataset contains over 1,000 medical abbreviations paired with more than 1,000 unique labels. It is primarily used to evaluate the performance of label classification models, especially when applied to the SciBERT model, which achieved a macro-F1 score of 77.29%. With a large scale, the core task of this dataset is medical abbreviation disambiguation.