Re: Spellchecking queries to a database
Re: Spellchecking queries to a database
- Subject: Re: Spellchecking queries to a database
- From: petite_abeille <email@hidden>
- Date: Fri, 24 Oct 2003 12:28:38 +0200
Hi Arturo,
On Friday, Oct 24, 2003, at 01:14 Europe/Amsterdam, Arturo Pirez wrote:
I think it's the only one on the web.
Well, there are several spellchecker available online.
Or maybe the only one attached to a search engine.
Or perhaps the only one that works that well :)
In any case WordNet is best suited for finding synonyms.
Well... this is one of its use, but you can also use its raw data as a
building block for a spellchecker.
Do you know of anything that can break a word up into phonemes?
That would be better.
Nope. But so far, I'm pretty happy with using WordNet as raw materials.
WordNet is a taxonomy. So taxonomies in general fit into the picture
if WordNet is of interest to you.
Hmmm... perhaps. But as far as spellchecker goes, I simply use WordNet
raw data.
We tried. Or rather some confused developers at my former place of
occupation did. It was a disaster from the get-go.
Yes. A traditional database is not well suited. I had more luck with
(ab)using Lucene for such devious purpose.
Memory isn't the problem.
But is helps :)
It's actually computationally expensive. Of course, you can always
trade off memory for computes. Off the cuff I'd say you'd need
something like 500GB. We had 120GB-150GB and it wasn't enough.
Hmmm... that's quiet a lot of memory... perhaps you guys are in a
league of your own... thanks god, a traditional spellcheck (e.g. aspell
or NSSpellChecker) is not that resource intensive...
Talking of NSSpellChecker, if you are serving your WOApp from Mac OS X,
you can access it directly for your own devious purposes...
Now there's a solution. Would they permit its use in this case?
Which case? You will need to check their usage policy. Or ask them
directly if you need something more. And keep in mind that this goes
through the wire, which is not recommended in all circumstances.
Statistical approaches bug me.
I learned to love statistics :)
But it would work if you had enough data and enough computes to
distill the raw data into a useful form.
Well, yes, the more the merrier. But even if you deal with one user,
keeping track of what suggestions work and which one doesn't helps.
Last time I tried something like this it kept an SGI Origin 2000 busy
for a month and did not complete.
Hmmm... oh, well... I guess you guys _are_ in a league of your own :)
And it only needed a working set of 24GB.
Are we still talking about a spellchecker here? 24GB is a _lot_ of
distinct words ;)
Cheers,
PA.
_______________________________________________
webobjects-dev mailing list | email@hidden
Help/Unsubscribe/Archives: http://www.lists.apple.com/mailman/listinfo/webobjects-dev
Do not post admin requests to the list. They will be ignored.