“Big” digital behavioral data increasingly allows large-scale and high-resolution analyses of the behavior and performance of persons or aggregated identities in whole fields. Often the desired system of study is only a subset of a larger database. The task of drawing a field boundary is complicated because socio-cultural systems are highly overlapping. Here, I propose a sociologically enhanced information retrieval method to delineate fields that is based on the reproductive mechanism of fields, able to account for field heterogeneity, and generally applicable also outside scientometric, e.g., in social media, contexts. The method is demonstrated in a delineation of the multidisciplinary and very heterogeneous Social Network Science field using the Web of Science database. The field consists of 25,760 publications and has a historical dimension (1916–2012). This set has high face validity and exhibits expected statistical properties like systemic growth and power law size distributions. Data is clean and disambiguated. The dataset with 45,580 author names and 23,026 linguistic concepts is publically available and supposed to enable high-quality analyses of an evolving complex socio-cultural system.

中文翻译：

---

绘制不可能的边界：社会网络科学的领域划分

“大”数字行为数据越来越多地允许对人的行为和表现或整个领域的聚合身份进行大规模和高分辨率的分析。通常，所需的研究系统只是更大数据库的一个子集。绘制田野边界的任务很复杂，因为社会文化系统高度重叠。在这里，我提出了一种社会学增强的信息检索方法来描绘基于领域再生机制的领域，能够解释领域异质性，并且通常也适用于科学计量学之外，例如，在社交媒体，上下文中。该方法在使用 Web of Science 数据库对多学科和非常异构的社会网络科学领域的描述中得到了证明。该字段由 25 个，760 篇出版物，具有历史维度（1916-2012）。该集合具有高表面效度，并表现出预期的统计特性，如系统增长和幂律大小分布。数据干净且无歧义。包含 45,580 个作者姓名和 23,026 个语言概念的数据集是公开可用的，应该能够对不断发展的复杂社会文化系统进行高质量的分析。