Many thanks for your effort towards a better user interface for Drupal.

The chosen algorithm could be accompanied by other algorithms:
like http://en.wikipedia.org/wiki/Levenshtein_distance

also the lucene search engine has some string stemming and similar stuff in it.

Please see this just as a suggestion or an invitation to discuss this topic.

Comments

xano’s picture

xano
he/they
Primary language English
Location Southampton
commented

I have tried to use Levenshtein, but I decided not to use it for reasons I cannot recall right now. The problem is that similarity review is slow and needs a fast but effective algorithm. The current algorithm may be relatively slow, but it's proven to be rather accurate so far, because it is based on words and not on characters.

Basically we need concrete reasons for implementing new algorithms. Benchmarks are greatly appreciated.

xano’s picture

xano
he/they
Primary language English
Location Southampton
commented
Status: Active » Closed (won't fix)

No new posts, so settings status to won't fix.