ChroniclItaly 3.0. A deep-learning, contextually enriched digital heritage collection of Italian immigrant newspapers published in the USA, 1898-1936.

This open access collection includes the digitized front pages of 10 Italian language newspapers published in California, Connecticut, Pennsylvania, Vermont, and West Virginia. It totals 8,653 issues and contains 21,454,455 words. The titles are: L’Italia, Cronaca sovversiva, La libera parola, The patriot, La ragione, La rassegna, La sentinella del West Virginia, L’Indipendente, La Sentinella, and and La Tribuna del Connecticut. The material was collected from Chronicling America, an Internet-based, searchable database of U.S. newspapers published in the United States from 1789 to 1963 made available by the Library of Congress. The corpus features mainstream (prominenti), anarchic (sovversivi), and independent newspapers thus providing a very nuanced picture of the Italian immigrant community in the United States at the turn of the twentieth century. To promote transparency, the collection includes two versions of ChroniclItaly 3.0: unprocessed (as it was collected from Chronicling America) and processed (with pre-processing interventions). Users can also find the data-sets including all the outputs from all the enrichment steps and post-intervention: named entity recognition (NER), geo-coding, sentiment analysis, and network analysis in addition to the readme.txt file that helps users navigate the folders and the metadata file containing relevant information. The code used to perform all the interventions is available at this GitHub repository https://github.com/lorellav/DeXTER-DeepTextMiner. Finally, all the enrichment outputs can be explored in the interactive app DeXTER available at https://c2dh.shinyapps.io/dexter/.