Geospatial issue: Lat Long plotted but no map

[dj2dirk212]

I have a list of addresses including lat/longs and am trying to compare them to another list I have in alteryx. When I load the data the points are created and displayed but not on a map. They seem to be registering as a legitimate location but not the correct one? Im wondering if this is something I am doing wrong in alteryx? 

[alexnajm]

It looks like the latitude and longitude values are not in the correct, numeric decimal format. Can you convert them to that and try again?

[alexnajm]

This site may help - https://www.fcc.gov/media/radio/dms-decimal

[dj2dirk212]

I have tried a few different ways by adding in decimal points and changing it from a string to an integer. Do you have any suggestions on the correct format? 

[alexnajm]

The latitude ranges from -90 to 90 and the longitude ranges from -180 to 180. Have you converted those values into this kind of number?

[ed_hayter]

Ive tried some rather arbitrary ways of converting - just guessing so don't use this without looking yourself.

I had mixed results some looked right (ones in DC, South Dakota and Arizona looked reasonable) While I had others that wildly missed the mark (Hawaii and Alabama)

Figured I'd leave this here anyway in case its some help

[ed_hayter]

Although looking at the longitude for the hawaii home it starts with 8 whereas for it to be hawaii it should really start with 15 i would have thought to convert to the correct longitude so either its a completely different format to what i was expecting or the data is not correct at entry?

Edit:

I think the top flow is correct parsing it as Degree Minutes Seconds - but that some of the input information is not correct would be my gut instinct on this

[dj2dirk212]

Im starting to suspect the same thing. Do you know if there is a way to do distance based on the address information? My end goal is to identify duplicates from two datasets. 

[alexnajm]

Not without converting it in someway to a spatial point. Unless you used some sort of API to take addresses and calculate distance that way! Check the community for Geocoder macros

[ed_hayter]

You might be able to use a zip code to lat long lookup table to get some more accurate points then Group by the points to see how many have the same location. Obviously might be multiple with different names in the same zip but you can build in logic to check for that afterwards.

I'd avoid fuzzy match if possible as it can be pretty inconsistent at delivering the desired result.

If you are confident that the address information does not include areas then you could parse the name of the home and its zip code and check for duplicates on those two fields if confident.