Online genealogical treasure trove was built using advanced machine learning technology and is the smartest of its kind

MyHeritage, the leading global service for discovering your past and empowering your future, announced today the publication of a huge collection of historical U.S. city directories that has been two years in the making. The collection was produced by MyHeritage from 25,000 public U.S. city directories published between 1860 and 1960. It comprises 545 million aggregated records that have been automatically consolidated from 1.3 billion records. This addition grows the total size of MyHeritage’s historical record database to 11.9 billion records.

This press release features multimedia. View the full release here: https://www.businesswire.com/news/home/20200228005389/en/

MyHeritage Releases Massive Collection of Historical U.S. City Directories (Photo: Business Wire)

MyHeritage teams applied innovative technologies to produce this collection and make it as useful and easy-to-use as possible. The city directories in this collection were published by cities and towns all over the U.S., and each directory is formatted differently. To overcome the formatting differences and unify the structures, MyHeritage corrected errors in the Optical Character Recognition of the scanned directory pages, and then employed several advanced technologies, including Record Extraction, Name Entity Recognition, and Conditional Random Fields to parse the data. By training a machine learning model how to parse raw free-text records into names, occupations, and addresses, the company produced a searchable, structured index of valuable historical information.

As an important resource for family history research, city directories can provide fascinating new discoveries for anyone exploring their family history in mid-19th to mid-20th century America. The records contain valuable insights on everyday American life spanning the time period from the Civil War to the Civil Rights Movement. Cities in the United States have been producing and distributing directories since the 1700s, providing an up-to-date resource to help residents find and contact local individuals and businesses. The city directories provide a wealth of information regarding family life during those years, listing names, residences, occupations, and relationships between individuals. Thanks to their exceptional level of detail, city directories can also provide a viable alternative to U.S. census records during non-census years, and can fill in the gaps in situations where census records were lost or destroyed. In 1921, a fire at the U.S. Department of Commerce destroyed most of the records from the 1890 census. Despite the loss of the records in the fire, much of the data can be reconstructed using the 1890 city directories on MyHeritage, which consist of directory books from 344 cities across the country, including 88 of the 100 most populated cities during that year.

“We are harnessing new technologies to make family history research more accessible than ever before,” said Tal Erlichman, Director of Product Management at MyHeritage. “The use of machine learning to process the city directory records highlights the major strides MyHeritage is making in digitizing global historical records.”