between the classes, and presents the results in dendrograms of the groups by years. LDA is useful to classify the abstracts automatically under certain topics, while RM is helpful for finding the internal structure of abstracts and extracting their macro-topics. The Editor summarizes findings in the concluding Chapter 11, noting that the distant reading, or computerized operating with texts requires a complementary qualitative selection of the corpora as well. Recent developments of digital methods prove that “In the near future high performance computers, new generation software packages, methods developed in machine learning and text mining fields, and data-driven computational tools will more and more affect the way scholars conduct research in the humanities and social sciences.” (p. 217) Each chapter presents many dozen references to the most recent literature. The book is extremely interesting, innovative, and useful for students and researchers in history of sciences, computational lexicography, automatic detection of patterns in texts, and general artificial intelligence.

中文翻译：

---

史前战争和暴力：定量和定性方法

类之间，并按年在组的树状图中显示结果。LDA 有助于将摘要自动分类到特定主题下，而 RM 有助于查找摘要的内部结构并提取其宏观主题。编辑总结了第 11 章的结论，并指出远程阅读或对文本的计算机化操作也需要对语料库进行补充性的定性选择。数字方法的最新发展证明，“在不久的将来，高性能计算机、新一代软件包、机器学习和文本挖掘领域开发的方法以及数据驱动的计算工具将越来越多地影响学者进行人文学科研究的方式。和社会科学。” （第 217) 每章都提供了许多对最新文献的引用。这本书非常有趣、创新，对科学史、计算词典编纂、文本模式自动检测和通用人工智能领域的学生和研究人员很有用。