National Archives and Records Administration (NARA) 2021 Artificial Intelligence (AI) / Machine Learning (ML) Information Day
peter.suber's bookmarks 2021-04-18
"The [National Archives] Catalog currently has a large data set (over 100 million digital pages of records, plus archival metadata and other types of records, including electronic databases) and an unsophisticated search. The archival hierarchy of the records is intended to assist the user in discovery, but in the digital realm, users find it difficult to use. The metadata that we have entered manually cannot provide the granular information for users to get the search results they want and it has taken NARA decades to produce. There is a delta between what NARA has been able to describe, and the specific information that users want from our records. Can AI fill the gap?..."