AnsweredAssumed Answered

Address dataset problem

Question asked by ckolaja on Mar 22, 2017
Latest reply on Mar 27, 2017 by ckolaja

Hi all. I have a question that I am hoping you all could help me with.

 

I am using ArcGIS Pro (but I could also use ArcMap 10.5).  I use StreetMap Premium for geocoding.  I have a large dataset with addresses where one person may have multiple addresses.  Sometime these are actually different address and sometime they are the same address repeated.  I want to geocode the address then (if they geocode to the same address/XY coordinate) collapse the address so that each address will only be one row (with the begin and end date from the first and last row respectively). If the address fields are not exactly the same it will not match it to the same geocoordinate and usually skips to the zip code.  You can see in the example below the rows 1 and 2 below will match to the same coordinate but row 3 will probably geocode to the zip code since the street name is not recognized (please take into account I made up the address).  Usually this is because the street name is misspelled but all of the other fields are the same. Is there some tool to clean these address when this happens? I am familiar with SAS programming if anyone know of a way to clean the addresses in that program. I was wondering if anyone on here has ever done this or if there is an easier way to do this in ArcGIS rather than geocoding then exporting and cleaning in SAS and the bringing back into ArcGIS Pro. 

 

Row   ID     Address                       City               State      Zipcode      DateBegin      DateEnd

1         1      123 Northbay Lane     Sometown      CA         12345         1/1/2015         2/28/2015

2         1      123 Northbay Ln         Sometown      CA         12345         3/1/2015         11/30/2016

3         1      123 Nrthby Lane         Sometown      CA         12345         12/1/2016       3/10/2017

 

Change to:

Row   ID     Address                       City               State      Zipcode      DateBegin      DateEnd

1         1      123 Northbay Lane     Sometown      CA         12345         1/1/2015         3/10/2017

 

In this example I just show one example for one person but this could happen several times for one person then the data sets are several hundred thousand observations so I can't manually make the changes.

 

Any advice would be greatly appreciated!

 - Claire

Outcomes