A project of automatically extracting migration records from Finnish church books.
Pipeline: https://github.com/TurkuNLP/htr-table-pipeline
Manually annotated dataset (train/dev/test): https://github.com/TurkuNLP/htr-annotations
Data release: https://zenodo.org/records/15606656
Vesalainen et al. (2025) Creating a Historical Migration Dataset from Finnish Church Records, 1800–1920