Extraction of gender role expressions in Northern Wei epitaphs using correspondence analysis and historical consideration
Ochi, Seiko
Meijo University, Japan
HTML XMLThis study focuses on the Northern Wei Epitaphs.We conducted text mining using KH Coder and performed quantitative analysis. The software uses R, MeCab and mysql to enable various kinds of quantitative text analysis.We were able to extract female and male characteristic words,and show the process of how gender roles are constructed and legitimized.
A Computational Exploration of the Narrative Functions of Poetry in Chinese Qing Vernacular Fiction (1644-1911)
Ma, Rongqian (1); Du, Keli (2)
1: Indiana University Bloomington, United States of America; 2: Trier Center for Digital Humanities, Germany
HTML XMLThis paper presents a pilot study exploring the narrative functions of poems embedded in Chinese Qing vernacular fiction (1644-1911). Analyzing a large corpus of Qing fiction, we identified five narrative functions of the embedded poems and tested the feasibility of applying computational models to classify poems based on their functions.
Topic modeling medieval Buddhist Chinese with BERTopic: Distinguishing concerns in translated and indigenous Chinese texts 500-800 CE
Bingenheimer, Marcus (1); Brody, Justin (2)
1: Temple University; 2: Franklin and Marshall College
We try to answer whether the BERTopic topic modeling framework can be used to obtain topics that can be used to distinguish two corpora in Buddhist Chinese, a low-resource idiom, based on texts translated or authored between 500 and 800 CE. By using a modified BERTopic pipeline and 'virtual paragraphs' we are able to show that topic modeling can identify coherent and meaningful topics, which mark significant differences between the Indian-Chinese and the Chinese-Chinese texts.
Is rhyme meaningful? Examining rhyme words in topic models of nineteenth-century English sonnets
Houston, Natalie
University of Massachusetts Lowell, United States of America
HTML XMLThis paper presents a series of experiments that explore the relationship between rhyme words in English sonnets and their semantic meaning as discovered through LDA topic models in order to examine the relationship between the constraints of poetic form and the ideas, themes, or emotions expressed in poetry.
A Machine Learning Approach to Identify Printing Blocks for Japanese Kokatsuji (Old Movable Type) Books
Li, Yuxiao (1,2); Kitamoto, Asanobu (1,3)
1: National Institute of Informatics, Japan; 2: EPFL, Switzerland; 3: ROIS-DS Center for Open Data in the Humanities, Japan
HTML XMLIn this work, we present a novel machine learning approach to uncover the details of Japanese Kokatsuji (Old Movable Type) printing technology by analyzing digitized book images, empowering humanities scholars to conduct quantitative analyses of Kokatsuji books.