Antwerp Street Stories: ±90,000 machine-readable pages of handwritten local police reports (1876-1945)

This repository holds all scripts and data referenced in the paper:

Lith Lefranc, Mike Kestemont & Ilja Van Damme: “Antwerp Street Stories: ±90,000 machine-readable pages of handwritten local police reports (1876-1945)”

This README describes all assets in this repository, which are made available under a Creative Commons license: CC-BY-NC-SA. This open-source license is meant to encourage the scholarly reuse of these resources. If you reuse the data, we kindly request that you provide a proper citation to the paper mentioned above.

Assets

/scripts

/scripts/requirements.txt: all code was applied and tested using Python (version 3.11.10)

/scripts/postprocess-xml.ipynb: scripts used to postprocess the raw xml files after layout analysis and HTR of the manuscript images

/scripts/enrich-metadata.ipynb: scripts used to analyze the postprocessed xml files and enrich the metadata of each incident book

/scripts/visualize-metadata.ipynb: scripts used to make visualizations of the analyses of the metadata

/data

/data/xml-raw: folder with the raw xml files after layout analysis and HTR of the manuscript images (organized by incident book)

/data/xml-raw: folder with the post-processed xml files created by the scripts in /scripts/postprocess-xml.ipynb (organized by incident book)

/data/IB-metadata: folder with csv files of the original and enriched metadata of the incident books

/data/visualizations: folder with png images created by the scripts in /scripts/visualize-metadata.ipynb