How to extract points in the same location?

5641
6
Jump to solution
06-15-2016 11:35 PM
HectorBorro1
Occasional Contributor

Hello,

I have a very large dataset (5m records) and several of my coordinates are overlapping.

For example in one exact X,Y I have 232 attributes, which is plainly wrong. I am guessing the error to a bad geocoding but I need to prove my hypothesis by only keeping the locations that have several (more than 20) points in the same XY...

Any ideas?

Many thanks!

0 Kudos
1 Solution

Accepted Solutions
RebeccaStrauch__GISP
MVP Emeritus

I haven't tried it, but I think if you look at some of the samples in the Find Identical link Jayanta Poddar​ list, there is a what to keep dupes only....then maybe just do a count...relate or join the info back to the table and then select.

FindIdentical example 3: Output only duplicate records (stand-alone script)

Demonstrates the use of the optional parameter Output only duplicated records. If checked on tool dialog box, or if set, the value of ONLY_DUPLICATES, then all unique records are removed. keeping only the duplicates from the output/

then play with example 4 to read the output and do what you need.

FindIdentical example 4: Group identical records by FEAT_SEQ value

Reads the output of FindIdentical tool and groups identical records by FEAT_SEQ value.

View solution in original post

6 Replies
JayantaPoddar
MVP Esteemed Contributor

     Take a backup of your dataset before you proceed.

- If you want to delete the identical points, you could use Delete Identical—Help | ArcGIS for Desktop . Select the shape field for comparison.

- In case you just want to compare without deleting the features, use Find Identical—Help | ArcGIS for Desktop .

     NOTE: The above tools will require ArcGIS Desktop Advanced license.



Think Location
HectorBorro1
Occasional Contributor

Many thanks for your reply Jayanta Poddar

How can I do a "negative delete"? I mean to keep only the identical points and delete the single or unique?

Taking it one step further... Can I deep the ones that occur only more than 20 times?

cheers!

0 Kudos
RebeccaStrauch__GISP
MVP Emeritus

I haven't tried it, but I think if you look at some of the samples in the Find Identical link Jayanta Poddar​ list, there is a what to keep dupes only....then maybe just do a count...relate or join the info back to the table and then select.

FindIdentical example 3: Output only duplicate records (stand-alone script)

Demonstrates the use of the optional parameter Output only duplicated records. If checked on tool dialog box, or if set, the value of ONLY_DUPLICATES, then all unique records are removed. keeping only the duplicates from the output/

then play with example 4 to read the output and do what you need.

FindIdentical example 4: Group identical records by FEAT_SEQ value

Reads the output of FindIdentical tool and groups identical records by FEAT_SEQ value.

HectorBorro1
Occasional Contributor

Ok, so I ran the analysis and it worked! Thanks for the pointers

I tried extracting the table to run statistical analysis in Excel or FME just to find that the file does not exist in the location I asked it to be saved. (My Documents)

On the file manager, the .lock file that ArcGIS creates was there but with size 0bytes.

Any idea on where the table can befound?

Ps. I closed ArcGIS in order to "free" the locked file and it disappeared...

0 Kudos
LeileiDuan1
New Contributor III

In addition to the find identical points idea above, Collect Events tool may be helpful in this situation since it is useful on incident points that share the exact same XY location. Basically in the output feature class there will be a new field documenting how many points are overlapping at that location. In your case, if the purpose is to identify where there are more than 20 points, select that query in the output feature class, make that selection a separate layer, then use Select by Location where your original point data are identical to the selection set.

HectorBorro1
Occasional Contributor

aha! Just the tool I was looking for, many thanks Leilei Duan for sharing!