10

Ok, so with ArcGIS 10, ESRI has depricated 'Address Standardizing' in the new locators, I know you can still use the 9.3.1 locators to do this; but I heavily suspect that after the 10.1 version we will see this available less and less.

The FGDC Addressing Standard places a heavy value on good addresses having the Prefix, Prefix-Direction, Street-Name, Suffix, Suffic-Direction values for a good clean form. And my agency has a need for the same.

Are there other tools out there that people are using to take structured addresses and parse them out fairly-reliably so they can be stored? Are any of them free?

*Update:*I came across this on CodePlex (US Address Parser) and it has given me a really good jumping off point. Its really nice in that its managed code that I can use all over the place and so far it has had a very high sucess rate at breaking down parts to be reviewed line by line and part by part.

Improve this question
2
  • 1
    My install of ArcGIS v10 has ArcToolBox, Geocoding Tools, Standardize Addresses. The inputs can be a table or a feature class so this should work with the spatial reference data and/or raw addresses. Is this what you are looking for?
    – klewis
    Commented Aug 26, 2011 at 20:57
  • 1
    Yes, but when you had the long running ability to pass a address into a geocoder, then get back the result with a standardized response then you could make a SOAP level call for a single record at a time. Now; you are left to build a geoprocessing service, the host it seperate of the rest of the geocoding system. So they made the process more complex and with the current system having such issues; its not reliable.
    – D.E.Wright
    Commented Aug 26, 2011 at 21:05

5 Answers 5

Reset to default
7

Parsing of an address is a complicated process, as I'm sure you are well aware.

Using ZIP+4 data from the USPS, you can determine if a street exists within a given city/state/zip code. You can even verify that a primary number (house number) falls within the correct ZIP+4 range. Adjusting city names and street names to correct for spelling issues is also possible using spelling lists as well as "sounds like" matching. Taking an address and parsing it into the individual components and then comparing it against a database of known addresses is the only way to know that the parsing has been done correctly.

Knowing that an address fits within the assigned area and knowing that it is a real and deliverable address are distinct objectives. The first, address approximation, is something that Google Maps does very well. However, it is just that, approximation. Google Maps doesn't let you know if the address is actually deliverable, they show you where it would lie on the map if it were real. This is immensely valuable from a mapping standpoint and they have varying degrees of accuracy.

Certainly the USPS database has flaws (many of them) but they are also certainly more accurate and correct than any other single database of US addresses and it's that degree of accuracy (and the fact that it is constantly updated) that we rely on.

I also work at SmartyStreets and wanted to add to the conversation. If you need to validate fewer than 250 addresses per month, the API is free. If your organization is a nonprofit group, the service is completely free with no limits.

Improve this answer
2
  • We are building a seperate service at my agency to work on the Geocoding portions; where will will aggregate County and or City data for a first layer of validation; then pass into commercial sources like TomTom or NavTeq before Tiger. Where we are taking the basis of a ESRI composite geocoder and adding additional steps or branches to compare on source score with another and return the higher score of the two or three. Our use of AIS/AMS DBs from the USPS helps with the 'Deliverable' part, coupled with County data.
    – D.E.Wright
    Commented Jan 24, 2012 at 18:26
  • Thank you for sharing, Jeffrey. I will (shortly) edit your reply to conform to our site guidelines; please review our FAQ for more information.
    – whuber
    Commented Jan 25, 2012 at 15:08
2

I work in the address verification field, actually (a company called SmartyStreets) -- where we do postal address verification (in the United States).

You'll need a CASS-Certified vendor of address data. These companies' software has been approved by the USPS to append missing information onto an address and return accurate results. One such service is US Street Address API -- or if you have a list of addresses, there's the Bulk Address Validation Tool -- which will give you the information you need. Both are free to an extent, but pretty affordable after that.

In ArcGIS, I believe the LiveAddress API can be implemented programmatically. Alternatively, a list of addresses can be exported to a spreadsheet or CSV and uploaded into the list scrubbing service.

Improve this answer
2
  • 1
    Im not looking really for Bulk-Mailing correction or certifications which is what CASS is driven towards; more for just better parsing and lookup to help correct bad addresses. I already have USPS AIS and AMS databases and am now starting to build logic to parse and query a record from those sources if not matched fully. I will take a look at the links you reference; but I am driving towards building a solution that can be shared for most common US addresses for a lower cost. Just because USPS AIS/AMS doesn't find it doesn't mean its incorrect.
    – D.E.Wright
    Commented Jan 23, 2012 at 20:57
  • 1
    @D.E.Wright You are correct. There are addresses that exist outside the domain of the USPS, but the vast majority of them can be found within the USPS database of deliverable addresses. Fedex and UPS both have some addresses (usually remote ones) that they deliver to and USPS will not deliver to. It would be nice to have an authoritative source that combined all of those addresses. The difficulty with that is that the USPS doesn't exactly share their entire database, and neither do Fedex or UPS. For the most part it is proprietary data. Address verification is for more than just mailing.
    – Jeffrey
    Commented Mar 27, 2012 at 14:41
1

Yes there are tools, free would depend on where your addresses are located. There are many not free address scrubbers licensed by the USPS. One of the ones I use is ZP4. I find the input and output to be flexible based on how you store your data currently and how you would like to store your data in the future. You can choose to output your data, for those addresses which match to deliverable addresses within the USPS data, into 7 fragments. These fragments correspond to house number, predirection, street name, street suffix, postdirection, apartment unit abbreviation, and apartment number. It is the lease expensive solution I have found.

Improve this answer
1
  • 1
    While ZP4 is very cheap ($600/yr), the terms make me a little wary and it sounds like you have to be careful of how you are using it.
    – blord-castillo
    Commented Oct 14, 2011 at 20:27
1

I have had great success using the us address parser library. I built a python tool that uses the library to create the parsed out address fields. https://github.com/datamade/usaddress

Improve this answer
0

Another option for US based addresses is YAddress. It also corrects and validates the address before splitting it into components. (I work for the company that makes it.)

Improve this answer

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.