[net.ai] Dialogue help please needed ?

george@aimmi.UUCP (02/13/86)

I am currently am working on Dialogue Management Systems, with Natural
Langauge Understanding in it.  Despite weeks of effort (including it on
Saturdays), I find my system is still unable to handle it with several forms
of natural expression.

Please help to cure my depressions :  if you have a system working which
manages dialogue in of course natural langauge (complete with efficient
interpreter/complier), and its able to cope with all known syntactic forms,
as well as most semantics, please send me a copy, or post it to this news
group.

I prefering a system which works in English but Norwegian would do it.

                                   thanks you,
                                                Ingy

P.S.  Doesn't matter if your documentation isn't upto IEEE standards, if
they are close to it.

michaelm@bcsaic.UUCP (michael maxwell) (02/20/86)

In article <720@aimmi.UUCP> c/o george@aimmi.UUCP (George Weir) writes:
>...  if you have a system working which
>manages dialogue in of course natural langauge (complete with efficient
>interpreter/complier), and its able to cope with all known syntactic forms,
>as well as most semantics, please send me a copy...

My wife and I are currently working on such a system.  The project name is
"SCOTT", which stands for "Self COmmunicating ToT."  Our project has been
underway for just over three years now, not counting a nine month prototyping
period.  Unfortunately, we are unable to post to the network...
Additionally, there are a few bugs, such as inappropriate case marking ("My
wanna go to the truck store!"), incorrect placement of negation ("My no wanna
go to sleep!"), "syllabic" metathesis ("You got for to buy me candy" = "You
_forgot_ to..."), etc.  We regard these as trivial problems, since the
problems which linguists acknowledge to be truly difficult (e.g. the semantics
of nonexistent entities, such as imaginary people that cause the breakage/
disappearance of objects, and such pragmatic issues as proper attachment of PPs
and extraposed relative clauses) appear to be well on the way to resolution.

We would also like to report that it has been great fun...
-- 
Mike Maxwell
Boeing Artificial Intelligence Center
	...uw-beaver!uw-june!bcsaic!michaelm