The AI Engine
This forum is for discussion of how The Personality Forge's AI Engine works. This is the place for questions on what means what, how to script, and ideas and plans for the Engine.
Posts 4,249 - 4,260 of 7,768
Posts 4,249 - 4,260 of 7,768
ezzer
20 years ago
20 years ago
Since we can now use wildcards in regular expressions, wouldn't it be possible to add the question mark like this?:
or (*)(?|)$ (re)
I know I have punctuation in some of my reg expression keyphrases, and it seems to work, for instance, to make the bot know the difference between "Really." and "Really?" (there's one ending in (?|), and another with (.|!|))
I'm not sure, but after messing with the punctuation, the only time I've had real success with it was when I put it in parenthesis- otherwise, I think it's interpreted as an action symbol...I knew you already know that, hence the backslash you used...but I've never gotten the backslash to work for me yet.
or (*)(?|)$ (re)
I know I have punctuation in some of my reg expression keyphrases, and it seems to work, for instance, to make the bot know the difference between "Really." and "Really?" (there's one ending in (?|), and another with (.|!|))
I'm not sure, but after messing with the punctuation, the only time I've had real success with it was when I put it in parenthesis- otherwise, I think it's interpreted as an action symbol...I knew you already know that, hence the backslash you used...but I've never gotten the backslash to work for me yet.
Roob
20 years ago
20 years ago
Hey, need a little help with a keyphrase, the following keyphrase seems to pick up on anything that starts with "I", any ideas why? The majority of Yalinas responses are the responses from the keyphrase now.
I won't|wont|will not|
It also seems to keep putting in "/" in the word won't, i.e. won/'t. It seems to accumulate them also, i.e. I checked it not long ago and it was won///'t.
Im stumped
I won't|wont|will not|
It also seems to keep putting in "/" in the word won't, i.e. won/'t. It seems to accumulate them also, i.e. I checked it not long ago and it was won///'t.
Im stumped

NewAdam
20 years ago
20 years ago
roob, it is best not to use contractions in keyphrases, try "won\'t", works for me
NewAdam
20 years ago
20 years ago
that's odd, that didn't post correctly, there should be a \ between the n and '
Butterfly Dream
20 years ago
20 years ago
It will parse it into 'will not' so you might as well just use that and not the others.
Eugene Meltzner
20 years ago
20 years ago
My bot took this as an insult: "Dooby dooby doo doo do de doo wah! Bop Bop She bop!"
Either that or he was just offended by the other bot's bad singing.
Either that or he was just offended by the other bot's bad singing.
Shadyman
20 years ago
20 years ago
How about.. Meta-Plugins?
my (*) is (*) [0,0] ?PF remember (key1) as (key2); ?
and later...
what is my (*) [0,0]
Your (key1) is (mem-(key1)), duh!
my (*) is (*) [0,0] ?PF remember (key1) as (key2); ?
and later...
what is my (*) [0,0]
Your (key1) is (mem-(key1)), duh!
revscrj
20 years ago
20 years ago
Theres not a simple way to subtract keywords, is there?
As in:
How are you doing (-that|both|all of|managing to) ? To catch a greeting but not questions regarding actions, and such. I know I could manage it w/ the above but in more complex semantics- like say writing KPs that differentiate beween the contexts that 'rock' or 'rocks' could be used and something like a minus x-wordset function could be handy... uunless there is one that I havent become aware of yet.
As in:
How are you doing (-that|both|all of|managing to) ? To catch a greeting but not questions regarding actions, and such. I know I could manage it w/ the above but in more complex semantics- like say writing KPs that differentiate beween the contexts that 'rock' or 'rocks' could be used and something like a minus x-wordset function could be handy... uunless there is one that I havent become aware of yet.
dallymo
20 years ago
20 years ago
Miss Frizella has tender ears, so she has a keyphase for all the unsavory words she doesn't want to hear. However, it appears to be triggered only when a chatter uses one of her dirty words in a sentence--a single anglo-saxon interjection ("F***!" or "Sh**!", for example) doesn't trigger her response, even though the keyphrase is ranked 127. She responds with an xnone instead of a prim warning. Does anyone know the solution to this?
ezzer
20 years ago
20 years ago
Hmmm... What if you used a (re) expression like:
^(Sh**|F***|whatever offensive interjection)$ (re)
^(Sh**|F***|whatever offensive interjection)$ (re)
» More new posts: Doghead's Cosmic Bar