Parsing inputs/outputs together

[michael.eves]

Hi guys, its been a busy day of posting I feel like you must be sick of reply to me now . I have another request.
Is it possible to parse inputs and outputs I.e.

If I have a rule _hello
Input: hello
Output: hi there

And a second rule _where you from?
Input: where you from?
Output: I am from Leeds, England. Where are you from?

Is there a way to parse the two inputs in such a way that the input could be….
Input: Hello again, where are you from?
Output: Hi there, I am from Leeds, England. Where are you from?

Presumably a comma or full stop would be need to separate the inputs so Verbot would search all inputs for a rule to govern “hello again” the comma, would tell verbot to search for a second rule to govern the latter half of the input “where are you from”

If verbot finds two such rules it would parse the two responses together using a comma to separate them. If verbot can only match one of the inputs it will respond to the recognised input and ignore what comes after the comma.

I think the ability to parse inputs/outputs in this way would give potential for an extremely powerful artificial intelligence. I am sure this question has probably been asked before so I would like to know what other people in the Verbot community think about it.

Thanks
Michael

[Aaron]

We've toyed with something like this before. Usually it resulted in some strange outputs. The best we could come up with was to split the user's input at sentence breaks (. ! ?) and feed each sentence as a separate input, then combine the outputs. Unfortunately, even this tended to cause more problems than it was worth. Still, I agree that it would be cool.

[michael.eves]

Is there any chance we could get our hands on the source code and or the compiled programme? Just to have a play around with it… please J

PS. is the source code written in c#?

[Aaron]

The source for the core engine is available here under the GPL:
http://verbots.sourceforge.net/

Yes, it's C#.