This hands-on workshop introduces HistText, a researcher-oriented application designed for the exploration, datafication, and analysis of large multilingual text corpora. Developed through a close collaboration between historians and computer scientists, HistText provides an integrated environment in which scholars can move step by step from query design and corpus construction to entity extraction, data transformation, and visualization. Unlike tools that focus on only one stage of the workflow, HistText brings these operations together in a single environment oriented toward transparent and reproducible humanities research. The workshop is designed for humanities researchers who wish to engage in computational text analysis without needing advanced programming skills. It will demonstrate how HistText structures research into a transparent and replicable workflow: participants begin with a research question, identify suitable collections, formulate and refine search strategies, build a corpus of relevant documents, and then transform textual materials into analyzable data. The workshop will also introduce the logic of the text-mining pipeline underlying these operations and discuss its strengths and limits when applied to historical materials. Particular attention will be paid to the importance of documenting analytical choices at every stage, so that the research process remains explicit, accountable, and reproducible.
Armand et al. (2026) studied this question.