search

Babzsak / duke

Automatically exported from code.google.com/p/duke
0 stars 0 forks source link

Implement weighted Levenshtein without full matrix #85

Closed GoogleCodeExporter closed 8 years ago

GoogleCodeExporter commented 8 years ago 
We can speed up Weighted Levenshtein, too, by using the same trick as for 
ordinary Levenshtein.

Original issue reported on code.google.com by lar...@gmail.com on 5 Jan 2013 at 9:54

GoogleCodeExporter commented 8 years ago 
Well, it's implemented, but it turned out not to be any faster, after all. Not 
sure why. Thinking that probably Ukkonen and similar approaches will be more 
fruitful for weighted Levenshtein.

Original comment by lar...@gmail.com on 7 Jan 2013 at 8:47

GoogleCodeExporter commented 8 years ago 
Closing this, since it didn't work.

Original comment by lar...@gmail.com on 17 Nov 2013 at 9:04