Sagaustus/adh-text-analysis-pack

Computational text analysis for African digital humanities: 33 runnable notebooks on African-language news, public-domain African archives and court records. sklearn + stdlib only. # Computational Text Analysis Pack Computational text analysis for African digital humanities. One notebook per method, where **one row is a document, a passage or a word** rather than a spreadsheet cell. Built for scholars of literature, languages and linguistics who want the statistics behind the tools. ## Using them Each notebook is self-contained — open any one without having run the others. **In Google Colab:** click the badge at the top of a notebook, then run the first cell. It will ask you to upload the CSVs; pick them from this pack's `data/` folder. The upload is remembered for the rest of the session. **Locally:** put the notebook in the same folder as the CSVs and run it. Needs `pandas`, `numpy`, `matplotlib`, `scipy`, `scikit-learn`, `statsmodels`. **Inside CourseHub:** every notebook is also the 📓 Lab Notebook tab on its lesson, next to the quiz. Nothing to install and no upload step. ## Data Real African-language and African-archive text. News in 16 African languages with topic labels, passages from 150 public-domain works about Africa, and passages from 48 African court judgments and government reports. Everything is public domain, open government, or AFL-3.0. Nothing in-copyright is included. See `data/PROVENANCE.md`. | File | What it is | |---|---| | `masakhanews.csv` | news in 16 African languages with topic labels; 3,600 of the 8,400 documents are in Hausa, Igbo, Yoruba or Nigerian Pidgin | | `masakhaner.csv` | | | `colonial_texts.csv` | 5,984 passages from 150 public-domain works about Africa — a colonial-discourse corpus, not an African-authored one | | `african_legal.csv` | 1,692 passages from 48 African court judgments and government reports | ## The notebooks | # | Notebook | Technique | |---|---|---| | | **Module 00 — Text as Data** | | | `00` | Text as Data | Turning a corpus into a table: document, token, type | | | **Module 01 — Framing the Question** | | | `01a` | What a Corpus Can and Cannot Answer | Matching a literary or …