I was wondering how to split a column in half in order to speed up a fuzzy match join. Figure A contains all the items. Figure B contains items 1-5 while Figure C contains items 6-10. After the fuzzy matching result has been completed on the separate columns, I want to rejoin them so that they resemble Figure A again with the addition of the Matched result score. Is this possible?
ID | Item a | Item b |
1 | A | a |
2 | B | b |
3 | C | c |
4 | D | d |
5 | E | e |
6 | F | f |
7 | G | g |
8 | H | h |
9 | I | i |
10 | J | j |
Figure A
ID | Item a | Item b |
1 | A | a |
2 | B | b |
3 | C | c |
4 | D | d |
5 | E | e |
Figure B
ID | Item a | Item b |
6 | F | f |
7 | G | g |
8 | H | h |
9 | I | i |
10 | J | j |
Figure C
Solved! Go to Solution.
The split needs to be dynamic. Some values in certain fields will remain the same e.g. State and ZIP will remain 2 and 5 respectively.
I'm not sure that this is splitting the column in half. It seems that it is just creating a new column entirely with all of the entries repeated in the new column.
Sorry about that!! I didn't see the image you posted in your original query when the post came through in my email, and it looked like it was set up like this:
ABCDEFGHIJ | abcdefghij |
So you are totally correct in that my response would not be helpful in the way you had it set up!!
However, I think to do what you're trying to do, you could find the median record number (use a Summarize tool to find the Median of RecordID), then append the median back onto your original data, then filter to split entries that are greater than the median. Would that do it?
Again, sorry about that. The way tables come across in the email format is WAY different than how they look here on the Community!!
No need to apologize for the wrong formatting. I understand completely. Happens to the best of us.
I think your solution will work, however, I am running into a problem when appending the median back onto the original. I am using a union to rejoin the two, and It gives the column with null as all the values. Should I use a different join?
Use the Append tool rather than the Union tool - the median will be your Source (bottom input) and your original info will be the Target. This will add the median field on to every record in your original list, whereas Union will try to join your columns together by field/position. Example attached. Does that help?
Yes that helped. Thank you very much.