𝒟 b a l

该数据集包含了我们队列中所有阳性患者以及同等数量的随机抽取的阴性患者，从而创建了一个平衡的数据集，适用于分类任务。此外，该数据集用于评估不同自然语言处理方法在处理类别不平衡问题上的表现，并包含了用于模型训练和评估的训练集、验证集和测试集。该数据集的规模涉及肺癌患者队列，任务目标是预测患者是否患有肺癌。

This dataset includes all positive patients from our cohort and an equal number of randomly selected negative patients, resulting in a balanced dataset suitable for classification tasks. Additionally, this dataset is employed to evaluate the performance of various natural language processing (NLP) methods when addressing class imbalance problems, and it contains training, validation, and test sets for model training and evaluation. This dataset is built upon a cohort of lung cancer patients, with the task objective being to predict whether a patient has lung cancer.