AI Zone Admin Forum Add your forum

NEWS: Chatbots.org survey on 3000 US and UK consumers shows it is time for chatbot integration in customer service!read more..

New to aiml chatbots
 
 

  Hello everyone, I have been using ultra Hal from Zabaware since 2004. My version of ultra Hal supports using Haptek characters. In the past I had been able to get the character to change clothing by using trigger words but in the latest version I have been having problems with the Haptek character. So I have decided to do some research on other types of chat bot’s.
  I have been able to download an older but free version of VirtualFem which works in Windows 7.
I’ve also put Aimlpad which contains a chat bot and Rebecca which I was able to download. At this point in time I have removed all of the Aiml files and replaced them with Rebecca’s files except for the dictionary which was used with VirtualFem.
  At this point in time I have a character window with a looping film, a chat bot that works and is able to handle a somewhat normal conversation. Unfortunately the dictionary which I recently added seems to create as well as solve problems. An example would be : he is disabled and needs help to get things done.  Bot >disabled…  Ah, disabled, people who are crippled or otherwise physically handicapped.
  I’ll be adding some questions to the forum later.
Carl

 

 
  [ # 1 ]

Hi, Carl, and welcome to chatbots.org! smile

From what you’ve described regarding the dictionary, I wonder if it may be that there is either only one definition per term, or if there’s a method of selecting the correct definition that isn’t being utilized. Without further info, it’s hard to do more than guess, though. smile

 

 
  [ # 2 ]

  Here is a sample of a conversation that seems to have gone wrong but serves as a good example:
Courtney>greeting…  Ah, greeting, an acknowledgment or expression of good will (especially on meeting).
Carl>so we are greeting each other?
Courtney>greeting…  Ah, greeting, an acknowledgment or expression of good will (especially on meeting).
Carl>I will say hi as my greeting.
Courtney>greeting…  Ah, greeting, an acknowledgment or expression of good will (especially on meeting).
Carl>what would you like to say?
Courtney>Let’s talk about movies.
Carl>what type of movie do you like?
Courtney><sad>What difference would it make? t wouldn’t be any fun.
Carl>of course it makes a difference to me.
Courtney>difference…  Ah, difference, the quality of being unlike or dissimilar.
  As the dictionary I originally put in 26 aiml (listed d2-a to d2-z) files which did not work as a dictionary, I then added two additional aiml files !xdef and !xdefbrain-n2 and the dictionary function works.  I’d just opened the !xdefbrain-n2 aiml and found the definition for disabled is the same as above and the definition in the d2-d is a little different.  So it seems that the d2 files are not needed and can be removed which I will try next.  For conversational purposes I feel that the dictionary hinders rather than helps the conversation at this time.
  At this time I would like to say thank you for responding to this post so quickly, I believe you have moved me forward in the right direction.
Carl

 

 
  [ # 3 ]

I can see where you would be a bit dissatisfied by that conversation, Carl. It’s not very “conversational”, is it?

As a bit of advice, to make posts of that nature a bit easier to read, I would like to suggest that you surround those conversation snippets with a [ quote ] tag (no spaces, though). The result of that would look like this:

Courtney>greeting…  Ah, greeting, an acknowledgment or expression of good will (especially on meeting).
Carl>so we are greeting each other?
Courtney>greeting…  Ah, greeting, an acknowledgment or expression of good will (especially on meeting).
Carl>I will say hi as my greeting.
Courtney>greeting…  Ah, greeting, an acknowledgment or expression of good will (especially on meeting).
Carl>what would you like to say?
Courtney>Let’s talk about movies.
Carl>what type of movie do you like?
Courtney><sad>What difference would it make? t wouldn’t be any fun.
Carl>of course it makes a difference to me.
Courtney>difference…  Ah, difference, the quality of being unlike or dissimilar.

As you can see, the separation of the conversation snippet from the rest of the post helps greatly with readability. smile

Just a thought.

Please let us know how it goes when you remove the original dictionary files. I, for one, am interested in finding out if it helps.

 

 
  [ # 4 ]

It sounds like you are defining each word as the conversation develops. Just a wild guess but you aren’t misusing the <srai> tags like this are you:

<category>
<
pattern>HI THERE</pattern>
<
template><srai>HELLO</srai></template>
</
category

Could you post a sample category you are using which is causing problems and we can take a closer look.

 

 
  [ # 5 ]

  At this time I believe I have found the solution for the problem I was having, I had found that the !xdef aiml is needed to start the search, putting in the d2-a to d2-z files does not cause a problem and is where the search is conducted. The problem seems to come from the !defbrain2-nr file which contains XDEF and standard pattern with templates. 
  The <srai> at this time is over my head, haven’t done a thing with them but would like to learn more about them for future use.  I also have a learning aiml that I don’t think is working.
Carl2

 

 
  login or register to react