Markus and LoGaRT: digital tools for research in Chinese texts
On 22 November we had the pleasure to welcome Brent Ho (Hou Ieong), Berlin State Library, and Shih-Pei Chen, Max Planck Institute for the history of science, as the first speakers in the new cycle of ENP-China conferences. They presented the two digital tools that they developed respectively for text mark up and data extraction in Chinese texts, more particularly for classical Chinese.
Brent Ho holds a Ph.D. degree in Computer Science from National Taiwan University and has been working on developing tools for historical research as well as in the context of library science ever since. He was the main IT specialist for the ERC project “Communication and Empires” led by Hilde De Weerdt. During this project, he designed and developed the online platform MARKUS for tagging Chinese historical texts. MARKUS won the 3rd place in the “Best DH Tools or Suite of Tools” category in the worldwide DH Award in 2016. Brent joined Berlin State Library as a special subject research librarian for Digital Humanities in 2016. He is actively involved in developing new pathways into the rich CrossAsia platform.
Shih-Pei Chen is the Digital Content Curator for Department III “Artefacts, Action, and Knowledge” of the Max Planck Institute for the History of Science (MPIWG). She got her PhD degree in Computer Science from National Taiwan University in 2011 and has been working closely with historians in designing research tools since 2013. Prior to joining MPIWG, Shih-Pei was a postdoctoral researcher at Harvard University from 2011 to 2013, and was the project manager for the China Biographical Database (CBDB), in which she directed and explored the text-mining tasks to automatically extract biographical information from the digital versions of primary and secondary sources and import them into CBDB. She is currently researching and developing digital research methodologies for studying the History of science, technology, medicine, and is interested in applying those methods in broader historical research. She also organizes digital research projects in collaboration with other scholars at MPIWG.
Brent Ho presented the various features of MARKUS and a number of practical applications in research projects across a wide range of sources. MARKUS allows scholars to process texts automatically for markup (individuals, positions, place names, dates, etc.) or to mark up texts manually in order to train MARKUS to enrich its repertoire of classified terms. There will be soon an extension of MARKUS to Korean. Brent Ho emphasized the fact that MARKUS is suited for the micro-reading of texts.
Shih-Pei Chen, conversely, highlighted the different nature of LoGaRT, a tool developed specifically to explore, mark up, and extract data from Chinese local gazetteers (地方誌). Local gazetteers were often compiled by members of the local elite and were produced under the sponsorship of the local officials. They contain all sorts of information on the historical, geographical, economic, administrative, biographical, touristic, etc., aspects of a locality in China. Local Gazetteer constitute a very massive corpus of about 4,000 identified titles (probably more) for which LoGaRT provides a way to carry out extensive searches with the ability to retrieve the information for further processing in statistical, spatial (GIS) and visualization software.