Some people eat, sleep and chew gum, I do genealogy and write...

Wednesday, February 5, 2014

#InnovatorSummit -- Tapping into Billions of Records with MyHeritage.com Matching API

This presentation was by Richard Stauffer and the entire name of the presentation was "Tap into Billions of Historical Records and Family Tree Profiles with the MyHeritage Matching API. Of course, I couldn't miss this one. Just a brief review, an API is an Application Programming Interface. It is important to know about these, even if you are not a developer, because they are the ways that third-party programs can access the data in the host database, in this case, MyHeritage.com. This is one of a series of classes focusing on API issues. This is where the rubber hits the road so to speak in the world of large online genealogy database programs. The large companies can expand access to their records while at the same time controlling who and under what terms that access may be given.

Matching in MyHeritage.com is matches individuals in your family tree with individuals and the same individual in other family trees. Once a match has been confirmed by the user, the match can be added to your own family tree. The other matching technology is Record Matches where the search engine matches the user's individual with the millions of records already on the MyHeritage.com. There is also a confidence level displayed when a match is obtained.

In this case MyHeritage is seeking additional online partners from online database applications. This API allows developers to provide MyHeritage.com matches on their own website. In this case, the number of matches can be displayed on the website. MyHeritage uses JSON (JavaScript Object Notation) data-interchange format.

Most of this presentation dealt with the mechanics of how the API worked. It is interesting that this will work in any of the 40 languages supported by MyHeritage. This is essentially a way for third-party websites to benefit from the matching capabilities of the MyHeritage.com website while still keeping the third-party users within their own website.


No comments:

Post a Comment