A page of the census

Artificial Intelligence and Humans Work Together to Free the Census

The Open Genealogy Data project will make historical census databases available for free

PROVO, Utah — January 15, 2019 — Amateur family history buffs and professional historians alike can celebrate the Open Genealogy Data census transcription project. Census records are some of the most valuable resources for genealogy, but they are usually only searchable one name at a time and according to the restrictions set by large genealogy websites. The project at opengendata.org is changing that by combining the power of artificial intelligence with volunteer transcribers to release entire historical census databases to the public for free as “open data,” which means they can be re-used and distributed without restrictions.

“In the past, making a census database was super expensive and time-consuming because people had to read every handwritten census page and type the information they saw into a computer,” explained Doug Kennard, creator of the Open Genealogy Data project. “Since there are hundreds of millions of census records, only companies and organizations with a lot of money could afford to do that. But by using artificial intelligence, we can do most of the work automatically, at a fraction of the cost.”

For those parts of the census that are too difficult for the current technology to read automatically, anyone can volunteer to help transcribe over the internet at the opengendata.org website. In addition to improving the accuracy of the census database itself, the transcriptions by volunteers will be used to improve the artificial intelligence technology for future open-data projects.

Besides helping genealogists, releasing the census as open data is also helpful to technologists who have innovative ideas. “I've seen a lot of people with great ideas for improving how we do family history, but sometimes they just don't have the databases to power those ideas. Open-data projects like this will help them turn their ideas into reality because there are no strings attached like there usually are with data from large genealogy companies that have an interest in preventing competition,” said Kennard.

To volunteer or to access data as it becomes available, visit the website at opengendata.org

About Open Genealogy Data:

Open Genealogy Data is an online community project to make data that is important to genealogists more open and accessible. The project is sponsored and hosted by Historic Journals, a Provo startup company that develops technologies meant to make family history easier and more interesting.

###