The Boston Public Library's Anti-Slavery collection—one of the largest and most important collections of abolitionist material in the United States—contains roughly 40,000 pieces of correspondence, broadsides, newspapers, pamphlets, books, and memorabilia from the 1830s through the 1870s.
The primary production goal of this project is to gain a complete corpus of machine-readable text from these handwritten documents. There are no software programs that can accurately convert handwriting into characters that a computer can understand as an actual letter, number, or symbol. Once the documents have all been transcribed and converted into this machine-readable text, we will upload the text into ourrepository systemand index them along with their corresponding image files. Users will then be able to search the full text of the letters across the entire collection.
We also plan to make the transcriptions available as a complete, open access data set, with the intention that the corpus will be exposed to machine learning, topic modeling, and other natural language processing and computer visualization applications.

https://www.antislaverymanuscripts.org

• United States of America

The primary production goal of this project is to gain a complete corpus of machine-readable text from these handwritten documents. Users will then be able to search the full text of the letters across the entire collection.