Comparing text strings


  I am new in the world of KNIME and I really would appreciate some expert advise on a current challange.

In my workflow, I am trying to categorize some search terms based on similarities with product names (like for example if someone search for a 'product xyz' with the term 'xyz', i want ideally that KNIME can associate these 2 text strings based on best available match).

How would you suggest to solve this problem? Which noded shall I use? I don't have a lot of thecnical background :)

Kind regards

If you are just getting started, have a look at the Rule Engine node.  This can be quite handy.  You could use a rule like:

$ColumName$  LIKE "*xyz*" => "has xyz"

To find your rows which contain that substring.

One other possibility would be the "StringSimilarity" node which we provide via the Palladian nodes. It currently offers n-gram overlap, Levenshtein and Jaro-Winkler similarity.