Computational & Quantitative Approaches
These pages are for projects where close reading alone cannot handle the corpus. Computational methods shift the interpretive work to collection, preprocessing, validation, and explanation.
The split with the qualitative approaches is about analytical posture. The same speeches or tweets can support several kinds of analysis, depending on the claim.
Launch the wizard
If the immediate problem is OCR, cleanup, metadata, or analysis-ready files, use the standalone wizard. It routes by scale and compute, then gives you a starter kit for Claude Code or Codex.
Preparation before analysis
Core computational methods
Topic Analysis
Use LDA, STM, or embedding-based models to map recurring themes.
Sentiment Analysis
Estimate tone, then validate the measure against the corpus.
Word Embeddings
Represent words or documents as vectors for similarity, drift, and classification.
In the classroom
I teach these methods in two Leiden courses. If you are enrolled in either one, these pages give you the methodological language that the weekly exercises do not always have time to spell out.
Digital Korea
12-session course in computational text analysis with Orange Data Mining and R, aimed mainly at Korean Studies students. It starts with preprocessing and ends with topic modeling.
BA3Text as Data (DH strand)
Six-seminar digital-humanities strand of the BA3 Contemporary Korea and Digital Humanities course. No programming required. Students work with prepared Korean corpora and learn how the main text-as-data tools behave.
If your thesis draws on either course, use these pages to turn the classroom workflow into methods-chapter prose.
Combining with qualitative methods
Many theses are stronger when a computational measure is checked through close reading. See Qualitative Approaches for that side of the split and for common pairings.
Overview and other methods
Return to the Methods overview if none of the pages here fits your project.