CoCoScore-BERT sample dataset

Samples of disease-gene association data set used in Master thesis project of Nuttapong Mekvipad. The data sets were automatically generated using distant supervision as described in BERT-based context-aware classifier for biomedical relation extraction. The tar file contain training and test sets usable by fastText and BERT models found in https://github.com/NMekvipad/CoCoScore-BERT.本数据集为Nuttapong Mekvipad硕士论文项目中使用的疾病-基因关联数据集样本。该数据集采用远程监督（distant supervision）方法自动生成，相关实现细节可参考《面向生物医学关系抽取的基于BERT的上下文感知分类器》（BERT-based context-aware classifier for biomedical relation extraction）。该tar文件包含可供fastText与BERT模型使用的训练集与测试集，上述模型的相关代码可通过https://github.com/NMekvipad/CoCoScore-BERT获取。