electricsheepeurope/europe-ilo-how-xees-eco-ocu-nb-mean-weekly-hours-actually-worked-per-employee-by
收藏资源简介:
该数据集包含434,371个观测值,记录了38个欧洲国家从1991年至2025年期间,按经济活动和职业划分的雇员实际平均每周工作时间。数据来源于国际劳工组织(ILO)的ILOSTAT数据库,通过REST API获取并过滤为欧洲国家代码,使用国际劳工统计学家会议(ICLS)定义进行协调。数据集包含一个核心指标HOW_XEES_ECO_OCU_NB,表示按经济活动和职业划分的雇员实际平均每周工作时间。数据列包括国家代码、国家名称、数据来源、指标代码、指标标签、分类变量(如经济活动和职业)、观测年份、观测值(以小时为单位)、观测状态标志以及相关注释。数据为年度频率,涵盖多个欧洲国家,适用于表格分类、回归分析和时间序列预测等机器学习任务。数据集由Electric Sheep Europe重新打包,以Parquet格式发布,便于使用Hugging Face的`load_dataset()`加载。
This dataset contains 434,371 observations of mean weekly hours actually worked per employee by economic activity and occupation across 38 European countries from 1991 to 2025. The data is sourced from the International Labour Organizations (ILO) ILOSTAT database, retrieved via the REST API and filtered to European ISO3 country codes, harmonized using International Conference of Labour Statisticians (ICLS) definitions. It features one primary indicator HOW_XEES_ECO_OCU_NB, which represents Mean weekly hours actually worked per employee by economic activity and occupation. Columns include country codes, country names, data sources, indicator codes, indicator labels, classification variables (e.g., economic activity and occupation), observation year, observed values (in hours), observation status flags, and related notes. The data is annual in frequency and covers multiple European countries, suitable for machine learning tasks such as tabular classification, regression, and time-series forecasting. Repackaged by Electric Sheep Europe in Parquet format for easy loading with Hugging Faces `load_dataset()`.




