As I understand it, trigram matching is already possible in PostgreSQL but not part of the standard fuzzy-matching procedures of TinyTM as yet. Has anyone integrated this already or is working on this?
Cheers,
Arne.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
It is a part of PostgreSQL. I have tried it in naive way by some tutorial. I constructed trigram index and used search function. It was approximatly 8times quicker than levenstein. But i think that this use was not optimised. If anyone knows how to use trigram indexes optimaly, let us know.
I thougth that trigrams should be quicker than 8x times levenstein. I tried it on DB containing roughly 600k segments.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi there,
As I understand it, trigram matching is already possible in PostgreSQL but not part of the standard fuzzy-matching procedures of TinyTM as yet. Has anyone integrated this already or is working on this?
Cheers,
Arne.
It is a part of PostgreSQL. I have tried it in naive way by some tutorial. I constructed trigram index and used search function. It was approximatly 8times quicker than levenstein. But i think that this use was not optimised. If anyone knows how to use trigram indexes optimaly, let us know.
I thougth that trigrams should be quicker than 8x times levenstein. I tried it on DB containing roughly 600k segments.