Fuzzy Text Match/Similarity Search

KNIME Extensions Text Processing
#1

Hello,

I have two columns that I would like to “join” or match based on similar words. example…

Col 1
TORONTO

Col 2
ETOBICOKE TORONTO ONTARIO CANADA

I just want to match Col 1 to Col 2 based on similar words without having to specify that “TORONTO” is the only matching word between the two columns … so far this seems to be harder than it should be. I have exhausted all previous posts about this subject (so my apologies for any subject repetition) and node/workflow example suggestions. Am I missing something very obvious? Thanks for your help, Knime Community

#2

With assumption that Col1 has always one word, start a loop for every row use a String Manipulation with substring. Use a new column as a join criteria. Filter out rows with empty columns.

#3

Hi,
On a similar not I want to match how close the strings are in column A and Column B of a particular row but it seems string matching or similarity function does it for all the rows in column A vs Column B.
Which note to use if I junt want to compare two columns on same row and get the similarity on the strings

Regards
Yogesh

#4

or

in a Chunk Loop.
1 Like