ROGER

Loredana Bercuci attended three Digital Humanities workshops at Cornell University on 10, 17, 18 July 2019, Ithaca, SUA: „Text Analysis with Digital Methods,” „Introductions to Visualizing Networks,” „Scrivener for Academic Writing.”

This workshop „Text Analysis with Digital Methods” introduced the attendees to two types of digital techniques: (1) counting and simple probability methods and (2) complex modelling and machine learning techniques.

The first part of the workshop discussed definitions of and differences between concepts such as data mining and natural language processing. The benefits of approaching text as data and digitizing text was also discussed at this stage (e.g. conduct research that would otherwise be beyond the scope of a single research project, expand reach and access to material, easily organize information). Finally, the pre-stages of digital data analysis were presented: gathering data, pre-processing, cleaning/parsing.

The second part of the workshop focused on counting and simple probability methods. The data used in the workshop was extracted from the digital archive “Documenting the American South” (DocSouth), a digital publishing initiative sponsored by the University Library at the University of North Carolina at Chapel Hill. The attendees learned to process the data using AntConc to discover collocates, N-grams, and concordances.

The third part of the workshop discussed complex modelling and machine learning techniques, such as sentiment analysis, topic modelling, stylometrics, clustering, and classification. The attendees were introduced to jsLDA: In-browser topic modeling, and discussed the Robots Reading Vogue project conducted at Yale.

Impressions: