About two months ago, we announced that our entire Factual data stack was now working in real-time for US Place entities. Today we’re pleased to report that this also now applies to the 49 additional countries that Factual covers.
As a reminder, Gil previously explained the Seven Steps a contribution travels as it works its way through Factual’s data stack. Now we are able to apply these processes to all global inputs, of which entity resolution is the most complex. Resolving an entity requires an algorithmic process that determines if two strings are referring to the same place. Without a solid Resolve API, our Place data would quickly get hammered with duplicate records. For example, let’s say we want to add the following data about a well-known restaurant in France:
Even though the contribution has no address number and contained “Rue” instead of the current value of “R.”, Resolve is able to add this data to the existing restaurant entry rather than create a duplicate record. Now imagine you want to push millions of contributions through – no problem, we can handle it.
Things get pretty interesting when you try this in double-byte characters. Take “Tea House Takano” (aka ティーハウスタカノ) in Japan – which we have here in our dataset. Let’s say we want to add the following data fields:
Resolve is able to determine that this record already exists in Factual, even though the contribution does not have address data, and is submitted with “Japan” instead of “JP” (as consistent with our schema). Resolve both matches and normalizes the input; another dupe prevented.
What does this mean for our developers? Access to even more fresh, accurate, and comprehensive global place data! It also means that any data you or your users contribute to Factual through the Submit API is incorporated instantly and hassle-free.
US Resolve is currently available as a stand alone API, and we will be making International Resolve available shortly in Q1 2013. So you’ll soon be able to perform place data cleansing and enrichment for global data.
Know the correct phone number of a spa in Argentina, have a better lat/lng for a bar in Tokyo, think a cafe in Jakarta is missing? Throw all the data at us and we’ll sort it out.
Adding data regularly,