<s>
An	O
intelligent	B-General_Concept
decision	I-General_Concept
support	I-General_Concept
system	I-General_Concept
(	O
IDSS	O
)	O
is	O
a	O
decision	B-Application
support	I-Application
system	I-Application
that	O
makes	O
extensive	O
use	O
of	O
artificial	B-Application
intelligence	I-Application
(	O
AI	B-Application
)	O
techniques	O
.	O
</s>
<s>
Use	O
of	O
AI	B-Application
techniques	O
in	O
management	B-Application
information	I-Application
systems	I-Application
has	O
a	O
long	O
history	O
–	O
indeed	O
terms	O
such	O
as	O
"	O
Knowledge-based	B-General_Concept
systems	I-General_Concept
"	O
(	O
KBS	O
)	O
and	O
"	O
intelligent	O
systems	O
"	O
have	O
been	O
used	O
since	O
the	O
early	O
1980s	O
to	O
describe	O
components	O
of	O
management	O
systems	O
,	O
but	O
the	O
term	O
"	O
Intelligent	B-General_Concept
decision	I-General_Concept
support	I-General_Concept
system	I-General_Concept
"	O
is	O
thought	O
to	O
originate	O
with	O
Clyde	O
Holsapple	O
and	O
Andrew	O
Whinston	O
in	O
the	O
late	O
1970s	O
.	O
</s>
<s>
Examples	O
of	O
specialized	O
intelligent	B-General_Concept
decision	I-General_Concept
support	I-General_Concept
systems	I-General_Concept
include	O
Flexible	O
manufacturing	O
systems	O
(	O
FMS	O
)	O
,	O
intelligent	O
marketing	O
decision	B-Application
support	I-Application
systems	I-Application
and	O
medical	O
diagnosis	O
systems	O
.	O
</s>
<s>
Ideally	O
,	O
an	O
intelligent	B-General_Concept
decision	I-General_Concept
support	I-General_Concept
system	I-General_Concept
should	O
behave	O
like	O
a	O
human	O
consultant	O
:	O
supporting	B-Application
decision	I-Application
makers	I-Application
by	O
gathering	O
and	O
analysing	O
evidence	O
,	O
identifying	O
and	O
diagnosing	O
problems	O
,	O
proposing	O
possible	O
courses	O
of	O
action	O
and	O
evaluating	O
such	O
proposed	O
actions	O
.	O
</s>
<s>
The	O
aim	O
of	O
the	O
AI	B-Application
techniques	O
embedded	O
in	O
an	O
intelligent	B-General_Concept
decision	I-General_Concept
support	I-General_Concept
system	I-General_Concept
is	O
to	O
enable	O
these	O
tasks	O
to	O
be	O
performed	O
by	O
a	O
computer	O
,	O
while	O
emulating	O
human	O
capabilities	O
as	O
closely	O
as	O
possible	O
.	O
</s>
<s>
Many	O
IDSS	O
implementations	O
are	O
based	O
on	O
expert	B-Application
systems	I-Application
,	O
a	O
well	O
established	O
type	O
of	O
KBS	O
that	O
encode	O
knowledge	O
and	O
emulate	O
the	O
cognitive	O
behaviours	O
of	O
human	O
experts	O
using	O
predicate	O
logic	O
rules	O
,	O
and	O
have	O
been	O
shown	O
to	O
perform	O
better	O
than	O
the	O
original	O
human	O
experts	O
in	O
some	O
circumstances	O
.	O
</s>
<s>
Expert	B-Application
systems	I-Application
emerged	O
as	O
practical	O
applications	O
in	O
the	O
1980s	O
based	O
on	O
research	O
in	O
artificial	B-Application
intelligence	I-Application
performed	O
during	O
the	O
late	O
1960s	O
and	O
early	O
1970s	O
.	O
</s>
<s>
Research	O
in	O
AI	B-Application
focused	O
on	O
enabling	O
systems	O
to	O
respond	O
to	O
novelty	O
and	O
uncertainty	O
in	O
more	O
flexible	O
ways	O
is	O
starting	O
to	O
be	O
used	O
in	O
IDSS	O
.	O
</s>
<s>
For	O
example	O
,	O
intelligent	B-General_Concept
agents	I-General_Concept
that	O
perform	O
complex	O
cognitive	O
tasks	O
without	O
any	O
need	O
for	O
human	O
intervention	O
have	O
been	O
used	O
in	O
a	O
range	O
of	O
decision	B-Application
support	I-Application
applications	O
.	O
</s>
<s>
Capabilities	O
of	O
these	O
intelligent	B-General_Concept
agents	I-General_Concept
include	O
knowledge	B-Language
sharing	I-Language
,	O
machine	O
learning	O
,	O
data	B-Application
mining	I-Application
,	O
and	O
automated	O
inference	O
.	O
</s>
<s>
A	O
range	O
of	O
AI	B-Application
techniques	O
such	O
as	O
case	O
based	O
reasoning	O
,	O
rough	O
sets	O
and	O
fuzzy	O
logic	O
have	O
also	O
been	O
used	O
to	O
enable	O
decision	B-Application
support	I-Application
systems	I-Application
to	O
perform	O
better	O
in	O
uncertain	O
conditions	O
.	O
</s>
