Based on machine learning and text mining "Tiangong Kaiwu" knowledge writing and technology, philosophical thought research

• Published in: Mediterranean archaeology & archaeometry Vol. 85; no. 1; pp. 635 - 646
• Main Authors: Zhang, Yixuan, Wang, Yishuo, Ni, Xiaohan
• Format: Journal Article
• Language: English
• Published: University of the Aegean 01.04.2025
• Subjects: Agriculture; China; Computational linguistics; Confucianism; Data mining; Data warehousing/data mining; Language processing; Machine learning; Natural language interfaces; Natural language processing; Natural resources; Technology; China
• ISSN: 1108-9628
• DOI: 10.1007/978-94-007-2627-7_12

Song Yingxing's Tian Gong Kai Wu (Exploitation of the Works of Nature [phrase omitted]) is a comprehensive work in the field of production technology at the end of Ming Dynasty. The book is divided into three volumes and eighteen articles, which covers a wide range of fields such as agriculture, handicraft and mining, as well as various production techniques, processing technologies and scientific principles. It reflects the popular science and technology level of the Ming Dynasty. The chapter structure, knowledge classification, writing background and technical thoughts of this book are important issues in the field of ancient Chinese ideological history and even the history of science. This research adopts an interdisciplinary approach and uses text mining tools such as subject modeling, word embedding and entity recognition to systematically analyze the text catalog and knowledge structure of Tiangong Kaiwu. Research indicates that the sequential order of the chapters in Tian Gong Kai Wu is not a result of chaos caused by Song Yingxing's lack of time for refinement, but rather a deliberate arrangement on his part. Song Yingxing's pragmatic, agrarian-focused, and anti-extravagance mindset fundamentally shaped the intellectual content of Tian Gong Kai Wu. Furthermore, his compilation of this work, fueled by his disdain for the frivolous academic atmosphere prevalent in the late Ming Dynasty, marked a shift towards practical learning. Essentially, this shift embodies the adherence to the Confucian philosophy of governing and applying knowledge to practical ends, critically inheriting the Confucian concept of "investigating things" (Ge Wu [phrase omitted]). Against the backdrop of Ming Dynasty Confucianism, characterized by metaphysical studies of nature and principles ([phrase omitted]), Song Yingxing reconfigured the methods of exploring rational principles and redefined his understanding of the natural order. Keywords: digital humanities, theme modeling, text mining, Tiangong Kaiwu, Song Yingxing, technical thought