Diabetes Screening Dataset
收藏arXiv2025-09-30 收录
下载链接：
https://github.com/YANGWU001/Putree/tree/main/dataset
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是一份新的糖尿病筛查数据集，其中包含了在中国一年时间内53,764位个体的购物行为信息，包括他们购买与糖尿病相关产品的频率。在这份数据中，有4,496个实例被标记为阳性（占总数的8.4%），为了反映糖尿病的患病率，阳性类别先验概率被设定为0.124。该数据集的规模涉及53,764位个体，任务是对慢性疾病进行预测。
This dataset is a novel diabetes screening dataset containing shopping behavior information of 53,764 individuals collected over one year in China, including the frequency of purchasing diabetes-related products. Within this dataset, 4,496 instances are labeled as positive, accounting for 8.4% of the total. To reflect the actual prevalence of diabetes, the prior probability of the positive class is set to 0.124. Involving 53,764 individuals, this dataset is designed for chronic disease prediction tasks.
提供机构：
YANGWU001
