This article presents a corpus study of lexical chains in the register of popular scientific writing, based on the English and German original texts included in the CroCo Corpus. This register is characterized by patterns resulting from the combination of accessible knowledge and scientific validity. The study aims at investigating cross-linguistic differences in terms of frequency and length of lexical chains. Chain elements are extracted by identifying the most frequent noun per text (and any lexical items associated with it by synonymy, hyponymy, or meronymy). The sentence IDs in the corpus markup are used to automatically calculate distances between occurrences of the target items. The analysis of immediacy chains and discontinuities in chains supports the investigation of contrasts in terms of topic continuity patterns.

中文翻译：

---

流行科学写作记录中的词汇链和主题连续性

本文基于 CroCo 语料库中包含的英语和德语原文，对流行科学写作注册中的词汇链进行了语料库研究。该注册簿的特点是由可获取的知识和科学有效性相结合产生的模式。该研究旨在调查词汇链频率和长度方面的跨语言差异。通过识别每个文本中最常用的名词（以及通过同义词、下义词或分词与之关联的任何词汇项）来提取链元素。语料库标记中的句子 ID 用于自动计算目标项目出现之间的距离。对直接链和链中不连续性的分析支持在主题连续性模式方面的对比研究。