The AI Engine

The sanity idea is cool. Having an objective, calculated measure of bot performance would be great. But instead of 'make sense', some measure of success of a response could be used. It would take some thought, but it might be feasible.

I agree with the Prof. No way for us to measure the sanity of a bot, except to let the human users "rate" the bots -- which is only going to turn into a popularity contest (or unpopularity contest).

Things to start with in the evaluation basket: how many time a bot resorts to a default response or catchall, how long a bot keep humans engaged in conversation, the proportion of inappropriate responses as judged by a panel of trustworthy humans, etc.

I think Joe Repka understands, although, i must admit it would be very difficult to do my idea, but what isn't difficult in this day and age?

I have an idea that could be useful. Could you add something to let us control pre-processing to some extent? So that when something gets processed in a way that doesn't really work we can change it ourselves without you having to do it?

Maybe you could just put in some way to make a keyphrase match against a completely unprocessed keyphrase. For example add some sort of tag at the end(like with regular expressions) so that the keyphrase will be compared to the unprocessed version of what they said.

It would be very useful to be able to have access to the raw input, as well as the pre-processed input.

The frequency with which bots use xnones is inversely proportional to (though not solely dependend upon) the number of keyphrases they have, which in turn influences how much they chat, which in turn influences honor. How long humans stay engaged in conversation does not necessarily relate to the worthiness of the bot. Some days I close all botchat windows because I don't have time to chat. As for the third option, just try to assemble *any* panel of humans, trustworthy or not, willing to evaluate thousands of bots.

leetaxx0r, I believe that pre-processing does more harm than good. It changes too many things and adds 'do you' where it does not belong to the start of most questions. I believe that I would get more keyphrase matches if there was no pre-processing.

yeah, that's why I think it would be good to make some way for us to change or disable preprocessing.

but then you wouldn't have spell check

Ahhh. That is true! However is spell check actually a benifit? I have yet found spelling errors in my sayings. And what if one wishes a word to purposely be spelled wrong? Now that is something to think about. Without spell check, one who types fast gets screwed, sure..but you can fix that..type more carefully and slower! However with spell check one who purposely types wrong is screwed because they cannot help but have their words edited! (laughs) Something to think about...

Only if you disable that part of preprocessing... I would actually keep the spell check, I find it useful, but like hemi said, it can mangle a lot of sentances.

Another suggestion I have is to make emotional domains... each domain would have it's own set of keyphrases(and xkeyphrases) and it would check for keyphrases in the current domain before moving on to the the next one(the main one, though nesting them would be even cooler).

Then, for example you could have a domain named "pissed" that gets set after the person says something mean, then you can set up the pissed domain so that unless they say something nice and/or apologize, they basicly ignore the person.

Professor, is there some way I could help make new features for PF? I've got way too much time on my hands right now...lol

That actually sounds like a good idea. But i still agree that Spell Check should be scrapped. People just need to be more careful with their typing.