← All examples

Data collection & scraping

Downloading over 1,200 declassified meeting minutes

An archive that lets you download one PDF at a time, and a script that handled the rest.

UCLA Library Data Science Center 2026

A researcher in the International Institute is building a Hebrew corpus of Israeli government meeting minutes, from the founding of the state in 1948 through the 1980s. All of the declassified meetings are online, each as a separate PDF, and the Israeli State Archives only allows them to be downloaded one file at a time.

Project manager Lian Elsa Linton met with the researcher in two consulting sessions and wrote a script that handled the downloading of the 1,200-plus PDFs. The code is kept in a repository so it can be rerun.

A follow-on request is now under way: some of the documents are poor scans, and the researcher and a postdoc want to feed them to an AI service that can read them only ten pages at a time, so the next step is a script that feeds the ten-page chunks to the service and collects the output.

Years
2026
Partner
International Institute
Team
Lian Elsa Linton
Source
Originally published ↗