ACSIncome-Adult

该数据集基于2018年美国社区调查（ACS）的公开使用微数据样本，由美国人口普查局提供的高质量数据组成，旨在预测美国工薪成年人的年收入是否超过5万美元。数据集包括人口统计特征，如年龄、婚姻状况、教育程度、职业、工作时间、出生地、种族和性别，并按照60%/20%/20%的比例分为训练集、验证集和测试集。该数据集包含160万行数据。任务是基于人口统计特征预测收入水平，同时确保公平性约束。

This dataset is built upon the public-use microdata samples from the 2018 American Community Survey (ACS), comprising high-quality data provided by the U.S. Census Bureau. The primary objective of this dataset is to predict whether the annual income of working American adults exceeds $50,000.
It includes demographic characteristics such as age, marital status, educational attainment, occupation, working hours, place of birth, race, and gender. The dataset is split into training, validation, and test sets with a 60%/20%/20% ratio respectively, and contains a total of 1.6 million rows of data.
The core task is to predict income levels based on demographic features while adhering to fairness constraints.