Computational Methods for Russian Literature : Current State and Future Directions

Computational text analysis has seen increasing adoption in English departments as a rapidly-improving set of methods for supporting the analysis of literature at scale. While some methods use algorithms that are developed specifically for English, even theoretically “language-neutral” methods that rely, in some form or another, on counting words can perform badly when applied to Russian-language texts, due to the complex inflection system that has no parallel in English. Recent developments within the field of natural language processing have begun to put some of these computational methods within reach for the scholar of Russian. In addition, new interfaces for annotating data, and using that data to fine-tune existing algorithms and models have made it easier for scholars to develop tools optimized for their own texts — without having to develop extensive programming skills or hire a developer. This talk will provide an overview of some of the most promising tools and resources for humanists interested in applying computational methods to Russian-language text (including such challenging tasks as digitizing Slavic-language manuscripts), and a preview of work in progress in this area.Non UBCUnreviewedOthe