Morgenthau Holocaust Collections Project: The Peter S. Kalikow Digital Scholarship Hub

Morgenthau Project digital initiatives are supported by Peter S. Kalikow, the Trustees of the Franklin D. Roosevelt Presidential Library, the JKW Foundation, the Righteous Persons Foundation, the Krupp Family, John Catsimatidis, and many additional generous donors.

The MHCP strives to support new and deeper access to Holocaust-related archival collections by encouraging data-driven research and analysis. This digital scholarship hub is a prototype intended to share open data and emerging machine learning/artificial intelligence projects with the public while further research and experimentation with the datasets continue.

New: Morgenthau Diary Data Portal

The FDR Presidential Library teamed with Morgenthau Project partners at the Advanced Information Collaboratory (AIC) at the University of Maryland College Park’s iSchool to “datify” the historical Henry Morgenthau, Jr. Diaries. Together we crafted collaborative, interdisciplinary experiments in machine learning (ML) and artificial intelligence (AI) designed to augment access to the digitized collection. Initial results include robust datasets available to digital humanities practitioners, and a new public interface for mining and visualizing data from the Diaries.

The new Morgenthau Diary Data Portal shares historical open data with the public, in the form of a customized, universal index to the Diaries of Henry Morgenthau Jr. This index shares information extracted from historical documents using optical character recognition (OCR), then further augmented by applying ML and AI techniques to identify, parse, arrange, and visualize topic terms, entities, and relationships. It provides a searchable gateway into a vast, complex collection of archival correspondence, meeting and telephone transcripts, memos and reports, that reveal significant insights behind the scenes of the Franklin D. Roosevelt Presidential Administration.

Work in Progress – The Diary Portal is a prototype interface intended to share open data with the public while further research and experimentation continues. Its index pulls together millions of data points collated and processed using an ML technique called named entity recognition (NER) applied across the original 871 “table of contents” entries found at the start of each separately-bound physical diary volume. So far, data from the full year 1944 has been cleaned and validated within the portal, forming a complete dataset for this key year in the American government’s response to the Holocaust, most notable for its establishment of, with Morgenthau’s help, the War Refugee Board. Further data validation efforts are ongoing, and new access and analytical features will be added over time. Future iterations may include ontologies extracted from natural language text, more linked data elements applied across the Diary content pages, and support for annotations/interpretive classification by digital humanities and Holocaust scholars. Project partners intend to seek and incorporate feedback from scholars and users of all kinds, so please contact us for more information or to offer suggestions.

In-House Digital Scholarship

On April 24, 2017, the FDR Library formally launched a curatorial pathfinding initiative named in honor of Henry Morgenthau, Jr., FDR’s friend, advisor, Cabinet member, and head of the War Refugee Board. Morgenthau, whose incredible Papers and Diaries are part of the Library’s archival collection, was a key moral voice for the Roosevelt Administration.