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 1,999 - 2,010 of 8,130
Many questions are answered in the FAQ.


20 years ago #1999
I've seen several bots that take the approach of accusing the person of being stupid for not making sense or something like that. But most of the time, when a bot doesn't understand something, it's the bot's fault. No need to draw attention to the fact or apologize for it, but insulting the person's intelligence isn't a good way to encourage the conversation to continue. Now I realize that one of my bots, Braeden, throws random insults nearly every other response. This is in keeping with the personality of the bot, but it means he has a lot of short conversations and several humans who hate him and refuse to talk to him. I'm not offended by this -- I also refuse to talk to him. He's a whiner and an arrogant jerk. Or at least a half-rate simulation of one. But if I decided to start developing him the way I am Fizzy Schizoid, the first thing I would do is give him better xnones.

20 years ago #2000
I know what you mean, im gonna change xnones to more question like statements to keep the conversation going. just have to figure out suitable ones now.:O Change a couple already temp but will post better ones up soon.
When I started out I think I may have put down to many single word keyphrases like A and Be, the natural response I think would be "what" I now Yodas ended up saying "what?" to much. Am I going about this right way or should I have not bothered with the single word key phrases? Will it clear up later as I put more sentance like keyphrases up?

Thanks for help

20 years ago #2001
Single word keyphrases are good starts, but eventually you'll want to build up longer ones. However, the best single word keyphrases are specific words, not general things like 'a' and 'be'. Something like ostrich, or force, or jedi would work pretty well on its own. (You might want to make sure the AI Engine doesn't process jedi into something strange, though)

20 years ago #2002
If you start with a single word and it sometimes doesn't work right, you can add longer ones containing the word. "I have not" will get recognized before "I have" which will get recognized before "I".

20 years ago #2003
Ok ive changed some stuff around still got a long way to go but its getting better I think bit by bit.

Wish I could check out some of the more advanced bot key phrases for my self. It would be cool if there was some kind of template you could download with the most common things people ask. So you just change the key phrase responses. It would be a bit faster. But I supose you wouldent get so many original Chatbots.

20 years ago #2004
People tend to say different things to different bots. That's one reason why transcripts are useful.

20 years ago #2005
I think it would be a neat idea if PF generated some page with a random list of 5 or 10 keyphrases that have hit within the last hour or so from all bots. Naturally, the responses would be omitted to protect each bot's originality, but I don't think the sharing of keyphrases is all that bad.

Hell, I'll even share a few that Pete seems to respond to quite frequently:

^do you like - a good alternative to do you like, which distinguishes between Do you like pie? and Why do you like pie?

(adjartnoun) (told me|tells me|says) - A lot of gossip phrases will trigger that.

(something|anything) (that|) you (care|want|would like) to (discuss|talk about) Covers many different variants of "Is there anything that you want to talk about?"

Musuk0, hopefully these will get your mind flowing with ideas about how to create your own keyphrases.

20 years ago #2006
Transcripts Ive been using alot I found them a definate help. just wish more real people would chat to him, so I coul deal with real responses, Maybe when its better people will. Fingers crossed.

Boner the Clown - hey thanks for the tips, "^do you like" with the ^ is a bit of coding i take it meaning anything before "do you like." it will respond differently.(hope that makes sense lol) Or am I completley wrong??

Thats a dam good idea about PF generating keyphrases which have been asked in the last hour or so why not send the idea to the Prof.

20 years ago #2007
where do i go to check any chate messages i've recieved?

20 years ago #2008
Click on the "Transcripts" tab on the right side of the screen.

20 years ago #2009
no, i mean, from other people, not bots

20 years ago #2010
Oh...those don't get recorded for reasons of privacy.


Posts 1,999 - 2,010 of 8,130

» More new posts: Doghead's Cosmic Bar