Exploring a corpus of Indigenous Australian autobiographical works with word embedding modeling
Ekaterina Kamlovskaya · 2022
A methodological reflection 1 IntroductionData-intensive research, 1 data-rich literary history, 2 distant reading, 3 macroanalysis, 4 algorithmic criticism, 5 cultural analytics, 6 digital literary studies 7these are just some of the names that could describe the field to which my project relates.All these names refer to the use of computational tools and methods to investigate research questions from the humanities and, more precisely, to analyze literary and historical textual sources.The research pipeline in such investigations usually consists of standard steps like research question formulation, source identification, data collection and analysis, and visualization and interpretation of the results.One peculiarity, though, is that researchers often borrow tools and methods from a field to which they are not "native"; in most cases, we see historians, linguists, or literary scholars applying methods developed in computer science.As digital tools transform traditional source types (e.g.text) and therefore inevitably change the way we interact with our sources, new perspectives may be created, new doors opened.However, phrases like "applying computational/digital methods," "running an algorithm on your dataset," or "running your data through an algorithm" often make the research process 1