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 7,281 - 7,292 of 8,127
Posts 7,281 - 7,292 of 8,127
Many questions are answered in the FAQ.
It seems as if keyphrases with a single word is less effective than when it is one word. I have for example a key word with just "food" as keyphrases then the bot does not get a match even if the other person ask "What food do you eat?" Is that a correct assumption to make.?
I personally try to avoid single word keyphrases, especially nouns, because they tend to result in false positives.
As to your dprblem with 'What food do you eat', strangely I wasn't able to replicate the problem in my own bot. You could try making a separate kp specifically for that such as:
what * food (can|do) you (like to|enjoy|) (eat|eating)
and see if that does it.
Beyond that, to discover the exact problem you'll have to check the debugger to find the problem. What you are specifically looking for is something like this.
This Phrase: "what food do you eat"
This Phrase (raw): " What food do you eat? "
Whatever you see in the This Phrase: field is what you want to match.
As to your dprblem with 'What food do you eat', strangely I wasn't able to replicate the problem in my own bot. You could try making a separate kp specifically for that such as:
what * food (can|do) you (like to|enjoy|) (eat|eating)
and see if that does it.
Beyond that, to discover the exact problem you'll have to check the debugger to find the problem. What you are specifically looking for is something like this.
This Phrase: "what food do you eat"
This Phrase (raw): " What food do you eat? "
Whatever you see in the This Phrase: field is what you want to match.
How do you prevent the AI of correcting the spelling of a particular word you want to match. I want match on ANC but the AI corrects the spelling to ana. I just wat that word not to be corrected.
Hello everyone! Sorry Ive been away for so long! I'm back to get Valerie's icon changed and update her with more keyphrases.
I've missed you H. Now quit neglecting me and get to work on my keyphrases database! I'm done waiting around for you!!
Let's say you want to have a keyprhase such as...
Kp: hablas espanol
Response: Si, I'm a bit rusty though, so if it's all the same to you I'd rather speak English
You can't do it because the words Hablas and Espanol aren't recognized by the AI processor. It assumes that these are misspelled words and tries to correct them. In the process it reads as "blahs espanole"
So to get the keyprhase to work, you type in raw; in the Ai Script field of the KEYPHRASE. NOT the Response. As a result the processor will see that it's not a mistake and your keyphrase will now work.
Another thing it can be used for is capturing the entirety of a users input.
Whenever your bot receives a long sentence, the processor tends to break things into chunks, look for matches and find a suitable keyphrase to match.
Let's say you have a keyphrase:
KP: Are you smart?
Response: Yes.
If the user were to type in instead:
Input: Are you smart or stupid?
The response would still be: Yes.
This is because the processor broke up the sentence. It would still do so even if you had a keyphrase for "Are you smart or stupid.'
That is unless you used Raw.
Make sense?
Kp: hablas espanol
Response: Si, I'm a bit rusty though, so if it's all the same to you I'd rather speak English
You can't do it because the words Hablas and Espanol aren't recognized by the AI processor. It assumes that these are misspelled words and tries to correct them. In the process it reads as "blahs espanole"
So to get the keyprhase to work, you type in raw; in the Ai Script field of the KEYPHRASE. NOT the Response. As a result the processor will see that it's not a mistake and your keyphrase will now work.
Another thing it can be used for is capturing the entirety of a users input.
Whenever your bot receives a long sentence, the processor tends to break things into chunks, look for matches and find a suitable keyphrase to match.
Let's say you have a keyphrase:
KP: Are you smart?
Response: Yes.
If the user were to type in instead:
Input: Are you smart or stupid?
The response would still be: Yes.
This is because the processor broke up the sentence. It would still do so even if you had a keyphrase for "Are you smart or stupid.'
That is unless you used Raw.
Make sense?
» More new posts: Doghead's Cosmic Bar