sofuslad/TJEN

TJEN数据集包含344,000条来自1750年至1850年丹麦-挪威最大城镇报纸的劳动力广告。这些广告由雇主（需求侧）或求职者（供给侧）提交，用于在报纸上发布，内容涉及劳动力买卖的各个方面，如职业结构、来源、经验、技能和推荐等。广告格式在整个时期保持高度一致，便于高精度提取。数据集是ENO数据集的子集，通过二进制分类模型和正则表达式分词器处理得到，并添加了性别和供需侧分类。验证显示成功率为96.12%。数据集结构包括文本、ID、日期、报纸名称、性别和供需侧指标。数据无敏感信息，但存在历史偏见，反映了18-19世纪丹麦-挪威的等级结构。The TJEN dataset contains 344,000 labour advertisements from the largest town newspapers in Denmark-Norway from 1750 to 1850. These advertisements were submitted by employers (demand-side) or job seekers (supply-side) for publication in newspapers, covering various aspects of labour power buying and selling, such as occupational structures, origins, experience, skills, and references. The advertisements maintain a highly homogenous format throughout the period, allowing for high-precision extraction. The dataset is a subset of the ENO dataset, processed using a binary classifier model and a RegEx-based tokenizer to isolate individual advertisements, with added classifications for sex and supply-demand side. Validation shows a success rate of 96.12%. The dataset structure includes text, ID, date, newspaper name, sex, and supply-demand indicator. The data contains no sensitive information but has historical biases reflecting the hierarchical structure of 18th and 19th century Denmark-Norway.