ParlaMint-TZ: ParlaMint-ready Corpus of Tanzanian Hansards
DOI:
https://doi.org/10.55492/dhasa.v7i02.7678Keywords:
Tanzanian Hansards, ParlaMint, Swahili corpus, parliamentary dataAbstract
This paper presents ParlaMint-TZ, a standardized parliamentary corpus constructed from Tanzanian Hansards spanning 2003 to 2025, alongside select archival records from 1975. Comprising 1,642 unstructured PDF documents (nearly 300,000 pages) and approximately 90 million words, primarily in Swahili with occasional English, this resource enables systematic, large-scale research on parliamentary discourse, political communication and multilingual practices. To overcome structural heterogeneity, inconsistent speaker naming and non-speech annotations, an open-source, deterministic rule-based processing pipeline was developed using ParlaMint-PL architecture to extract metadata and structure proceedings into ParlaMint-compliant TEI XML format. Finally, the corpus is integrated into an online platform featuring full-text search capabilities, term highlighting and structured transcript rendering, bringing African parliamentary data into a globally interoperable research ecosystem.
Downloads
Published
Issue
Section
License
Copyright (c) 2026 Beata Wójtowicz, Maciej Ogrodniczuk, Michał Rudolf, Dariusz Czerski

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.