| Sumario: | National bibliographies gather information on printed publications connected to a specific country or ethnic group, including works published within modern borders or by diaspora communities. Initially compiled for cataloging, these collections are increasingly used in cultural historical research, contributing to bibliographic data science (e.g., Lahti et al. 2019a). Such data supports studies on intellectual networks (Hill et al. 2019) and the historical popularity of written works (Tolonen et al. 2021). A central task in bibliographic data science is transforming catalog data into standardized, usable datasets via scalable workflows (Tolonen et al. 2019b, Mäkelä et al. 2020). However, catalog data entry practices vary over time, often prioritizing precision over harmonization for analysis, making standardization challenging for analysts seeking to use it. To address this, we created a curated dataset from the Estonian National Bibliography (n = 313,565), selecting over 50 data fields with broad interest and sufficient coverage. Entries were standardized, harmonized, and enriched with additional data on authors, place names, and publishers. Data transformation was conducted using a reproducible, modular pipeline that can be scaled for updates or adapted to other collections. In this article, we introduce the pipeline and dataset, highlighting use cases where the curated data facilitates cultural historical research. We recommend curated datasets as a strategy for GLAM institutions to enhance research access and leverage in-house expertise to support academic partnerships.
|
|---|