Workshop: Digital approaches towards serial publications (18th-20th centuries)
Where? Brussels, Royal Academies for Sciences and Arts of Belgium, Hertogsstraat 1 Rue Ducale, 1000 Brussels
When? 11-12 September 2017, organised by the Ghent Centre for Digital Humanities and the Centre de Recherche en Information et Communication (ReSIC) in the context of TIC-Belgium and DARIAH-BE.
The past few years have seen an increasing interest in Natural Language Processing and other text-mining techniques in the humanities. This tendency has been sparked by the distant reading approaches for literary theory, which aim to draw general conclusions on larger amounts of text using computational techniques. This way of tackling data has gained momentum, and many (large scale) projects have been set up in order to meet the expectations of humanities scholars wanting to make sense of the vast amount of digitised data that has become available in public domain, primarily, 18th and 19th but also 20th century publications. For some time NLP has been mainly used for the confirmation of existing historical knowledge, but now many techniques and tools have become more mature, it is time to draw an intermediary balance of NLP and text mining in general and the mining of serial publications in particular. Moreover, it also time to question the phenomenon of ‘scientific serendipity’ and NLP as a set of technologies enabling such serendipity.
This workshop brings together humanists, social scientists and computational scientists who have been working with historical serial texts and/or periodicals: including newspapers, journals, book series, congress series, etc. We are mainly interested in reflections upon completed case-studies: the research results compared to the initial expectations, examples of the phenomenon of ‘scientific serendipity’, the theoretical and heuristical implications of the choices made during the research process, comparisons between NLP and other methods such as Semantic Network Analysis. Last but not least, we will reflect upon graphical approaches to text corpora, such as visualisation methods that are en vogue for exploratory searching, analysis and communication of textual analysis. From n-grams through tree diagrams to word clouds: what are the epistemological implications of using graphic displays developed outside the humanities?
The workshop is free, but registration is required. Please register here.
For the full programme and practical details, please click here.