The AI Engine

Indeed, I don't think humans have genders, so I won't work for botmasters -or- guests.
I guess the only way to use it properly would be to have the bot ask the person their sex earlier on...

Request for the Prof:

Can we please get a mapping of "whatcha" to "what are you"? At the moment, it's being translated to the command "watch."

Thanks!

Professor,

I just had a fantastic idea that would add to humanization of responses, that hopefully wouldn't be very hard to implement.

I love the randomly substituted synonyms in the responses, and it gave me the idea that it would be really fun if you could also manually insert synonyms for matched patterns, such as (s_key1) means the output should return a synonym for key1, instead of key1 itself. Why would this be cool? Well, try these examples on for size:

Keyphrase="i have a (noun)" Response="There was a B-movie on last night called 'Day of the (S_key1)' that had a giant (key1) in it. What a hoot!"
Guest: I have a dog
Bot: There was a B-movie on last night called 'Day of the Canine' that had a giant dog in it. What a hoot!

Keyphrase="did you (verb) the (adjnoun)" Response="I didn't so much (key1) (key2) as (s_key1) it."
Guest: Did you eat the sandwich?
Bot: I didn't so much eat the sandwich as devour it.

Keyphrase="was it a (noun)" Response="I'm not sure, but it did seem like it might be some sort of (s_key1)."
Guest: Was it a mammal?
Bot: I'm not sure, but it did seem like it might be some sort of animal.

Turing's Dad: Not sure what you mean about the human users not having genders, because many profiles have a gender marked. Do you mean that the AI Engine doesn't register the human genders in a human-bot convo?

Whoa skysaw! nice!

That would be rad, Skysaw -- and you're right, it really might not be hard to implement.

Ok, all I can say is HOLY COW! I think I made a very interesting discovery! See if you can guess how this PF bot does this:

You: which actor do you like?
Bot: Any actor that is a scene-stealer gets my vote.
You: which computer do you like?
Bot: Oh! Do you remember the computer that was the number cruncher in that movie called 'The Guinness,' or something like that? Now THAT was a computer!
You: You: which computer do you like?
Bot: Any number cruncher that's not too good-for-nothing or brittle. I'm not too picky. Other than the 'good-for-nothing or brittle' part.
You: which vegetable do you prefer?
Bot: Any vegetable that is a spaghetti squash gets my vote.
You: which stone do you like?
Bot: The only good kind of stone is a pebble.
You: which airplane do you prefer?
Bot: I can't remember the name... do you recall that airplane that Time Magazine named 'biplane of the Year' in 2002?

Hint: no xfavorites were used in these examples.
Answer coming soon...

Awrrrsome!

weeeeeeeeeeeeeeee

You'd better tell us quick or Shadyman might run out of breath...

*listens* eeeeeeeeeeeee

Here's the answer... nested plugins!! I just took a chance to see if it would work, and it does in this particular case (with a few wobbles here and there). Here is Little Mu's test case:

Keyphrase: (which|what) (noun) do you (prefer|like)
Response: The only good kind of (key2) is a (typeof-(key2)-n).

What's happening is that (key2) first gets expanded to the noun that was input, then it gets tacked onto typeof-. Then the whole thing is looked up in wordnet. Some words work better than others, of course.

The main problem is with synonym substitution, which occurs before the lookup. If the noun is replaced by more than one word, a space appears in the typeof- lookup, wordnet gets confused, and it sort of breaks. Most of the time, it works.

Feel free to use the concept, just please use your own wording in your responses.

I'd be interested to see any similar experiments anyone else comes up with.