Wikidata 中人类条目覆盖潜在偏差的初步评估
Initial Evaluation of Potential Bias in Coverage of Humans in Wikidata
Clair Kronk
中文摘要
开放协作知识图谱 Wikidata 正日益成为智能体人工智能、信息检索和语言建模系统的数据基础，因此系统审计其人口统计代表性及整体公平性已成为重要研究需求。开源审计平台通过 QLever 导入逾 10 million 条陈述绑定，覆盖 Wikidata 上逾 6 million 名人类，并考察性别、性取向、地理分布、出生地城乡属性、族裔、标签与描述及别名的多语言覆盖、职业，以及部分维度的交叉组合。分析结合卡方拟合优度检验、95% Wilson 得分置信区间、差异比与 Rubin 缺失分类。结果显示，在标注性别的人类中，女性占 28.71%（CI +/-0.04）；38.26% 的人类有国籍陈述，其中西欧和北美（WENA）约占 53%。在可分类的 1.8 million 个出生地中，农村出生地仅占 2.48%，低于 27.4% 的全球基准。族裔陈述不足 1.2%，非英语 Wikidata 描述占 18.2%。各评估维度均存在显著缺失，族裔和性取向记录尤其不足，农村出生地与非 WENA 国籍的代表性则最低。
关键要点
- 01开源审计平台经 QLever 分析逾 6 million 名 Wikidata 人类实体
- 02考察性别、性取向、地域、出生地、族裔、职业及交叉维度
- 03有明确性别的人类中，女性占 28.71%（CI +/-0.04）
- 04农村出生地仅占 2.48%，远低于 27.4% 的全球基准
- 05族裔与性取向记录最不足，非 WENA 国籍代表性偏低
解读
尚无解读。
原始英文摘要
arXiv:2609.22375v2 Announce Type: replace-cross Abstract: Introduction. Open collaborative knowledge graphs such as Wikidata increasingly ground agentic artificial intelligence, information retrieval, and language modeling systems, making systematic auditing of their demographic representation and overall equity a research imperative. Methods. Herein, we present an open-source auditing platform that ingests over 10 million statement bindings representing over 6 million humans on Wikidata via QLever, and evaluates representation of gender, sexual orientation, geography, birthplace urbanicity, ethnicity, multilingual coverage of labels, descriptions, and aliases, occupation, and select intersectional pairs of these entities. It does so by making use of Chi-square goodness-of-fit tests, 95% Wilson-score confidence intervals, and disparity ratios, in light of Rubin's missingness taxonomy. Results. Women accounted for 28.71% (CI +/-0.04) of all humans in Wikidata with a stated gender. 38.26% of humans had a citizenship statement, with Western Europe and North America (WENA) representing approximately 53% of such statements. Among 1.8 million birthplaces that could be classified, rural birthplaces were observed in 2.48% of cases (in comparison to 27.4% global baseline). Fewer than 1.2% of entities carried an ethnicity statement, and non-English Wikidata descriptions covered 18.2% of items. Discussion. Our findings reveal significant missingness across the evaluated axes. Ethnicity and sexual orientation were the most critically under-documented (missing statements) while rural birthplaces and non-WENA citizenship were the most underrepresented.