But at some point you must supervise a child's learning a language. It may be possible for a child to learn to speak a language correctly after exposure, but would they be using correct grammar? For the most part probably, but chances are they'd be corrected by others. I doubt it would be possible for an algorithm without intervention or some form of real AI. They would have to know, at some point, correct uses of grammar to abide by it, wouldn't they ?
If it does develop decent grammar rules, which by the sounds of it it does, they'd be just that, it's own. Sure, it may be similar to what the language would define as good grammar, but it wouldn't necessarily be correct.
If I'm not mistaken, there is also a different between good grammar, and correct grammar.
Now, I don't purport to be a language or programming expert, but in English there's exceptions to lots of different rules which I don't think it would be possible to learn without being told so (ie, intervention). Again, the difference between good and correct grammer could come into play here.
If fed with a heap of decent grammar, what happens when it's fed with bad grammar and spelling? Will it learn, and incorporate, the tripe or reject it?
That's the sort of problem with natural language apps, it's quite hard to sort the good from the bad when it's learning. Take the megahal library http://megahal.alioth.debian.org/> for example. Although possibly not as complex, it does a decent job at learning, but when fed with rubbish it will output rubbish.
I don't think it's the learning that will be that hard part, but rather the recognition of the good vs. the bad that will prove how good the system is.
Slashdot Mirror
So, how many people outside of the USA and over 15 actually use AIM ?
I'd like to see the market statistics not including Americans.
If it does develop decent grammar rules, which by the sounds of it it does, they'd be just that, it's own. Sure, it may be similar to what the language would define as good grammar, but it wouldn't necessarily be correct.
If I'm not mistaken, there is also a different between good grammar, and correct grammar.
Now, I don't purport to be a language or programming expert, but in English there's exceptions to lots of different rules which I don't think it would be possible to learn without being told so (ie, intervention). Again, the difference between good and correct grammer could come into play here.
If fed with a heap of decent grammar, what happens when it's fed with bad grammar and spelling? Will it learn, and incorporate, the tripe or reject it? That's the sort of problem with natural language apps, it's quite hard to sort the good from the bad when it's learning. Take the megahal library http://megahal.alioth.debian.org/> for example. Although possibly not as complex, it does a decent job at learning, but when fed with rubbish it will output rubbish. I don't think it's the learning that will be that hard part, but rather the recognition of the good vs. the bad that will prove how good the system is.