Modified UCI Adult Dataset

该数据集是对UCI成人数据集的修改版本，其收入预测任务源自美国社区调查公共使用微数据样本（ACS PUMS），经过筛选，包含至少16岁以上、每周至少工作一小时，以及过去一年内收入至少100美元的个体信息。数据集还包括年龄、收入、教育、性别、血统和就业等属性。为了回归建模，收入目标变量经过了对数转换。该数据集的规模为1,664,500个个体，任务是对收入进行预测（回归任务）。

This dataset is a modified version of the UCI Adult Dataset. Its income prediction task is derived from the American Community Survey Public Use Microdata Sample (ACS PUMS). After strict screening, it includes individual records of subjects who are at least 16 years old, work a minimum of 1 hour per week, and have an annual income of at least $100 in the past year. The dataset encompasses attributes including age, income, education, gender, ancestry, and employment status. For regression modeling purposes, the income target variable has undergone logarithmic transformation. The dataset comprises 1,664,500 individual samples, with the task being income prediction, a regression task.