Newcomers
This is a forum for newcomers to the Personality Forge. Many questions can be answered by reading the Book of AI and the FAQ under the "My Bots" link in the upper corner.
Posts 5,255 - 5,266 of 8,130
Salve, ma parlate tutti italiano?
No, soprattutto Ingelese. Ma parlo un piccolo Italiano anche.
But my tourist Italian is not good enough to answer this
:
I do something like that:
Keyphrase: what about you,* about you
These 2 are synonymous - the wildcard "*" will also match "what", so there is no need to specify "what about you". It is also redundant, since "* about you" is synonymous with "about you" (the wildcard is soft, so will match anything or nothing in the string it is analyzing.)
...
I tryed with kyphrase:
(What|*) about you
* about you (not allowed)
These are also synonymous, and would logically match anything that "about you" matches.
You never need to use a wildcard at the beginning or end of a keyphrase, since these are always covered by (preykey) and (postkey).
This should also solve your (animal23) plugin question - use "do you like" instead of "do you like *", and then he can respond "no, I do not like (postkey)" etc.
"do you like (animal23)" would always take precedence if (postkey) is an animal listed in the plugin, because it is a longer keyphrase than "do you like".
Posts 5,255 - 5,266 of 8,130
Many questions are answered in the FAQ.
I just created a shared plug-in (under "phrases") called
(told_#_sentence)
which contains a few words analogous to "told" in that they fit in contexts like this:
I told him they would not agree.
or,
I told him that they would not agree.
That is, they are followed optionally by "that", followed by a noun phrase, followed by a sentential clause. For example, you could use a keyphrase,
(adjartnounprep) (told_#_sentence) (that|) (adjartnounprep) (verb)
I doubt that it contains all such verbs, but as I think fo more, I will add them.
(told_#_sentence)
which contains a few words analogous to "told" in that they fit in contexts like this:
I told him they would not agree.
or,
I told him that they would not agree.
That is, they are followed optionally by "that", followed by a noun phrase, followed by a sentential clause. For example, you could use a keyphrase,
(adjartnounprep) (told_#_sentence) (that|) (adjartnounprep) (verb)
I doubt that it contains all such verbs, but as I think fo more, I will add them.
No, soprattutto Ingelese. Ma parlo un piccolo Italiano anche.
But my tourist Italian is not good enough to answer this
:Keyphrase: what about you,* about you
These 2 are synonymous - the wildcard "*" will also match "what", so there is no need to specify "what about you". It is also redundant, since "* about you" is synonymous with "about you" (the wildcard is soft, so will match anything or nothing in the string it is analyzing.)
I tryed with kyphrase:
(What|*) about you
* about you (not allowed)
These are also synonymous, and would logically match anything that "about you" matches.
You never need to use a wildcard at the beginning or end of a keyphrase, since these are always covered by (preykey) and (postkey).
This should also solve your (animal23) plugin question - use "do you like" instead of "do you like *", and then he can respond "no, I do not like (postkey)" etc.
"do you like (animal23)" would always take precedence if (postkey) is an animal listed in the plugin, because it is a longer keyphrase than "do you like".
marco3b:
If you want your bot to respond half the time with negative answers, then I would suggest not using two separate keyphrases, but giving two responses to the same keyphrase.
For example:
do you like (animal23) [0,0]
yes
no
The AIengine will alternate between "yes" and "no" as replies.
If you want your bot to respond half the time with negative answers, then I would suggest not using two separate keyphrases, but giving two responses to the same keyphrase.
For example:
do you like (animal23) [0,0]
yes
no
The AIengine will alternate between "yes" and "no" as replies.
Hallo, I'm still bothering with the same problem:
it is strange for me, but the kyphrase: "about you" is correctly matched if the question is "tell me something about you?", "what can you say about you?" "about you?", but if the question is "what about you?" I have an xnone!
Do you understand why?
it is strange for me, but the kyphrase: "about you" is correctly matched if the question is "tell me something about you?", "what can you say about you?" "about you?", but if the question is "what about you?" I have an xnone!
Do you understand why?
"what about you" is one of the many phrases that get labeled "blab" by the AI engine. Nothing you can do. It's irritating but you just have to move on.
marco3b:
You might be able to get around the "blab" by writing something like
What (about|hysteresis) you [0,0]
This would in practice be equivalent to
What about you [0,0]
because no one is ever going to say,
What hysteresis you?
But perhaps the AIengine won't see that. I've never tried this; please let me know how it comes out!
You might be able to get around the "blab" by writing something like
What (about|hysteresis) you [0,0]
This would in practice be equivalent to
What about you [0,0]
because no one is ever going to say,
What hysteresis you?
But perhaps the AIengine won't see that. I've never tried this; please let me know how it comes out!
Yeah, I haven't found a way around blab--Psi or Prob, would "about you (re)" work in raw? or the thing with all the annoying brackets?
» More new posts: Doghead's Cosmic Bar