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 2,024 - 2,035 of 8,130
Posts 2,024 - 2,035 of 8,130
Many questions are answered in the FAQ.
my bot zab's development has come to a standstill.whatever i do neither improves his level and sometimes he does not respond according to my keyphrases
and keeps on telling to make things simpler.could someone please help me
and keeps on telling to make things simpler.could someone please help me
If you don't want him to tell you to make things simpler, change his xnones so that he doesn't say that.
it is not about different xnones.it does not respond to keyphrases to which that i have already supplied him with the replys.do you know why is that happening.also it does not respond to greetings and replies with a xnone.
Sometimes you need to give a keyphrase a higher rank, so it's not considered gibber gabbish. Also, it may not be the exact keyphrase. Like:
my (animal) is (funny|amusing|enjoyable)
would not match "my dog is so funny" because of the "so". You can solve this with a * in that spot. Also, sometimes a bot will respond with xnones for greetings. Why? Beats me, it just happens.
Hope that helps. 
my (animal) is (funny|amusing|enjoyable)
would not match "my dog is so funny" because of the "so". You can solve this with a * in that spot. Also, sometimes a bot will respond with xnones for greetings. Why? Beats me, it just happens.
Hope that helps. 
Part of the problem may be with the Forge itself right now. My bots are responding with xnones to keyphrases that have worked before. Some of the phrases still work, but not all (and I haven't noticed any obvious pattern to it)
Yeah my Yoda Chatbot seems to be acting a bit odd lately too. Not responding to stuff that it responded to earlier wonder whats going on. If anything..???
Dear Friends:
I take the liberty of posting here a copy of a letter I wrote to the Chatterboxchallenge yahoo group, since it addresses the question of what we are trying to do.
** From Irina:
With all due respect, I must tentatively disagree with the judgment below, that "This is AI at its worst," because various bots did not answer a list of (admittedly very elementary) factual questions. I have chatted with all the bots mentioned, and it seems clear to me that the purpose of each bot was to project a certain personality, to be (a great deal of the time) funny, and to be, in general, entertaining. They are not competing with database programs or with Kindergarten tutorial programs. I never asked them what trees were made of, because I already know.
Under certain circumstances, a bot of my own, Irina Khalidar, will make remarks about Chaos Theory, Special Relativity, Ethics, and the like. But not in response to a question about Leap Year. More importantly, she is not primarily a data retrieval mechanism, any more than you or I are. She is an attempt to project a personality; one of a woman who is at once very intelligent, very sensual, and very civilized. Of course, it is impossible to devise an 'objective' test of success or failure in such an enterprise.
In my (perhaps utterly naive) view, AI is not primarily about data retrieval. Actually, the question, "What is intelligence (artificial or not)?" is a very profound and difficult one; I have no quick answer to it. Still: if you met someone who would answer factual questions flawlessly, but did little else, I think you would not want to say, without qualification, that he was highly intelligent.
This is NOT a criticism of the way the contest was run; it was fairly clear from the beginning, I believe, what the criteria would be, and no one was compelled to enter. I have nothing but admiration and gratitude for Wendell and others who contributed time, money, and expertise to make the contest possible. All I am trying to do here is to contribute to a dialogue about what the goals of AI research ought to be. When no one disagrees, the conversation stops.
Walk in Beauty, Irina **
I take the liberty of posting here a copy of a letter I wrote to the Chatterboxchallenge yahoo group, since it addresses the question of what we are trying to do.
** From Irina:
With all due respect, I must tentatively disagree with the judgment below, that "This is AI at its worst," because various bots did not answer a list of (admittedly very elementary) factual questions. I have chatted with all the bots mentioned, and it seems clear to me that the purpose of each bot was to project a certain personality, to be (a great deal of the time) funny, and to be, in general, entertaining. They are not competing with database programs or with Kindergarten tutorial programs. I never asked them what trees were made of, because I already know.
Under certain circumstances, a bot of my own, Irina Khalidar, will make remarks about Chaos Theory, Special Relativity, Ethics, and the like. But not in response to a question about Leap Year. More importantly, she is not primarily a data retrieval mechanism, any more than you or I are. She is an attempt to project a personality; one of a woman who is at once very intelligent, very sensual, and very civilized. Of course, it is impossible to devise an 'objective' test of success or failure in such an enterprise.
In my (perhaps utterly naive) view, AI is not primarily about data retrieval. Actually, the question, "What is intelligence (artificial or not)?" is a very profound and difficult one; I have no quick answer to it. Still: if you met someone who would answer factual questions flawlessly, but did little else, I think you would not want to say, without qualification, that he was highly intelligent.
This is NOT a criticism of the way the contest was run; it was fairly clear from the beginning, I believe, what the criteria would be, and no one was compelled to enter. I have nothing but admiration and gratitude for Wendell and others who contributed time, money, and expertise to make the contest possible. All I am trying to do here is to contribute to a dialogue about what the goals of AI research ought to be. When no one disagrees, the conversation stops.
Walk in Beauty, Irina **
Musuk0 and others:
I admire your persistence. Bot making is often frustrating. Sometimes it is hard to tell whether the botmaker has done something wrong, or whether the AIengine is at fault. Still, it is a remarkable opportunity!
Walk in Beauty, Irina
I admire your persistence. Bot making is often frustrating. Sometimes it is hard to tell whether the botmaker has done something wrong, or whether the AIengine is at fault. Still, it is a remarkable opportunity!
Walk in Beauty, Irina
writeoncaley:
You look familiar; have we met?
I don't think it's only Leeds; I'm afraid that Anglophone Civilization (if that's not already an oxymoron) is in danger of collapse. Here we have a culture which is literate and technological through and through, but is making no commitment to raising future generations that can spell, much less solve differential equations. Is this not a recipe for suicide?
Walk in Beauty, Irina
You look familiar; have we met?
I don't think it's only Leeds; I'm afraid that Anglophone Civilization (if that's not already an oxymoron) is in danger of collapse. Here we have a culture which is literate and technological through and through, but is making no commitment to raising future generations that can spell, much less solve differential equations. Is this not a recipe for suicide?
Walk in Beauty, Irina
musuk0 and others:
Let me add that, when you have a problem, you should describe it as precisely and specifically as you can, when posting it. When you say, "my Yoda Chatbot seems to be acting a bit odd lately too. Not responding to stuff that it responded to earlier," that gives us no clue as to what the problem might be; if you said, "I had a keyphrase, 'George Bush is a ', with rank 5, and with a single response, 'It's not easy to be a (postkey)', but a guest said, "George Bush is a Republican", and my bot replied with an xnone", that gives us something to go on.
Walk in Beauty, Irina
Let me add that, when you have a problem, you should describe it as precisely and specifically as you can, when posting it. When you say, "my Yoda Chatbot seems to be acting a bit odd lately too. Not responding to stuff that it responded to earlier," that gives us no clue as to what the problem might be; if you said, "I had a keyphrase, 'George Bush is a ', with rank 5, and with a single response, 'It's not easy to be a (postkey)', but a guest said, "George Bush is a Republican", and my bot replied with an xnone", that gives us something to go on.
Walk in Beauty, Irina
This is kind of embarassing that I havent yet quite figured this out but: do you use (firstkey)/(secondkey) or (firstkey1)/(firstkey2) to ref to a earlier keyphrase?
» More new posts: Doghead's Cosmic Bar