Taiwan Lectures on Chinese Studies 台灣漢學講座:
A Distant Exploration of Literary Writings and Historical Documents in Classical Chinese
對傳統中文文學作品及歷史文獻的長遠探索
11 Nov, Wednesday, 2-3 pm (UK time)
Speaker: Prof Chao-lin Liu 劉昭麟, National Chengchi University, Taiwan
Chair: Dr Christopher Rosenmeier, University of Edinburgh
Zoom link for registration:
Abstract:
We design and employ software to analyze the contents of literary writings and historical documents. In the burgeoning stage of digital humanities in Taiwan, relatively simple yet interesting analysis of famous novels of Ming and Qing dynasties were used to stimulate the interests in digital humanities of the humanities researchers. Gradually, we move onto more practical analysis of the poetic works of the Tang and Song dynasties. Methods of text analysis assist us to look for useful information in historical documents in both classical and modern Chinese, and part of which represents our collaboration with the China Biographical Database Project of Harvard University. In recent years, we apply similar methods to analyze Taiwanese poems and local gazetteers. More recently, we extend our efforts to optical character recognition to digitise hard copies of classical Chinese texts. Aiming at achieving a more advanced level of algorithmic competence, we also explore the applicability of deep learning methods to the problems of word segmentation for classical Chinese poems and the problems of sentence splitting for historical documents that were also written in classical Chinese. We shall share selected experience in this brief presentation.
我們設計並使用軟體對文學作品和歷史文獻的內容進行分析。臺灣學界在數位人文學科的萌芽階段，便對著名的明清小說進行相對簡單而有趣的探討，以激發人文學者對數位人文研究的興趣。此後臺灣學界開始就唐宋詩詞作品展開更廣泛的研究，並藉著文本分析的方法幫助我們在古典和現代中文的歷史文獻中尋找有用的訊息，我們亦採用類似的方法來分析臺灣本土地地名詞典。其中最具代表性的成果之一便是我們與哈佛大學合作的「中國歷代人物傳記資料庫」。近年來我們致力於提升對紙本進行光學字符辨識的運算能力，以期提高繁體中文文本數位化的程度。我們還探討了深度學習方法在古文詩詞分詞，和歷史文獻分句等課題中的適用性。我將在這個簡短的演講中分享一些上述的研究和工作經驗。
About the speaker:
Prof. Chao-lin Liu obtained his Ph.D. from the Institute of Electronic Engineering and Computer Science, University of Michigan. His research focuses on data analysis and knowledge exploration, digital humanities, computational linguistics. Now he is the Professor and Chairperson of the Department of Computer Science and the Associate Dean of the College of Science, National Chengchi University. Prof. Liu is the incumbent president of the Taiwanese Association for Digital Humanity.
劉昭麟教授於美國密西根大學獲得博士學位，後致力於數位人文、資料分析與知識探勘、計算語言學、自動推理與模型建構、人工智慧等層面研究。劉教授目前擔任政治大學理學院副院長及資訊科學系教授兼系主任，同時也是臺灣數位人文學會的理事長。
Comments