<s>
A	O
blackboard	B-General_Concept
system	I-General_Concept
is	O
an	O
artificial	B-Application
intelligence	I-Application
approach	O
based	O
on	O
the	O
blackboard	B-General_Concept
architectural	O
model	O
,	O
where	O
a	O
common	O
knowledge	O
base	O
,	O
the	O
"	O
blackboard	B-General_Concept
"	O
,	O
is	O
iteratively	O
updated	O
by	O
a	O
diverse	O
group	O
of	O
specialist	O
knowledge	O
sources	O
,	O
starting	O
with	O
a	O
problem	O
specification	O
and	O
ending	O
with	O
a	O
solution	O
.	O
</s>
<s>
Each	O
knowledge	O
source	O
updates	O
the	O
blackboard	B-General_Concept
with	O
a	O
partial	O
solution	O
when	O
its	O
internal	O
constraints	O
match	O
the	O
blackboard	B-General_Concept
state	O
.	O
</s>
<s>
The	O
blackboard	B-General_Concept
model	I-General_Concept
was	O
originally	O
designed	O
as	O
a	O
way	O
to	O
handle	O
complex	O
,	O
ill-defined	O
problems	O
,	O
where	O
the	O
solution	O
is	O
the	O
sum	O
of	O
its	O
parts	O
.	O
</s>
<s>
The	O
following	O
scenario	O
provides	O
a	O
simple	O
metaphor	O
that	O
gives	O
some	O
insight	O
into	O
how	O
a	O
blackboard	B-General_Concept
functions	O
:	O
</s>
<s>
A	O
group	O
of	O
specialists	O
are	O
seated	O
in	O
a	O
room	O
with	O
a	O
large	O
blackboard	B-General_Concept
.	O
</s>
<s>
They	O
work	O
as	O
a	O
team	O
to	O
brainstorm	O
a	O
solution	O
to	O
a	O
problem	O
,	O
using	O
the	O
blackboard	B-General_Concept
as	O
the	O
workplace	O
for	O
cooperatively	O
developing	O
the	O
solution	O
.	O
</s>
<s>
The	O
session	O
begins	O
when	O
the	O
problem	O
specifications	O
are	O
written	O
onto	O
the	O
blackboard	B-General_Concept
.	O
</s>
<s>
The	O
specialists	O
all	O
watch	O
the	O
blackboard	B-General_Concept
,	O
looking	O
for	O
an	O
opportunity	O
to	O
apply	O
their	O
expertise	O
to	O
the	O
developing	O
solution	O
.	O
</s>
<s>
When	O
someone	O
writes	O
something	O
on	O
the	O
blackboard	B-General_Concept
that	O
allows	O
another	O
specialist	O
to	O
apply	O
their	O
expertise	O
,	O
the	O
second	O
specialist	O
records	O
their	O
contribution	O
on	O
the	O
blackboard	B-General_Concept
,	O
hopefully	O
enabling	O
other	O
specialists	O
to	O
then	O
apply	O
their	O
expertise	O
.	O
</s>
<s>
This	O
process	O
of	O
adding	O
contributions	O
to	O
the	O
blackboard	B-General_Concept
continues	O
until	O
the	O
problem	O
has	O
been	O
solved	O
.	O
</s>
<s>
Like	O
the	O
human	O
experts	O
at	O
a	O
blackboard	B-General_Concept
,	O
each	O
knowledge	O
source	O
provides	O
specific	O
expertise	O
needed	O
by	O
the	O
application	O
.	O
</s>
<s>
The	O
blackboard	B-General_Concept
,	O
a	O
shared	O
repository	O
of	O
problems	O
,	O
partial	O
solutions	O
,	O
suggestions	O
,	O
and	O
contributed	O
information	O
.	O
</s>
<s>
The	O
blackboard	B-General_Concept
can	O
be	O
thought	O
of	O
as	O
a	O
dynamic	O
"	O
library	O
"	O
of	O
contributions	O
to	O
the	O
current	O
problem	O
that	O
have	O
been	O
recently	O
"	O
published	O
"	O
by	O
other	O
knowledge	O
sources	O
.	O
</s>
<s>
In	O
a	O
blackboard	B-General_Concept
system	I-General_Concept
,	O
this	O
is	O
provided	O
by	O
the	O
control	O
shell	O
.	O
</s>
<s>
A	O
blackboard	B-General_Concept
system	I-General_Concept
is	O
the	O
central	O
space	O
in	O
a	O
multi-agent	O
system	O
.	O
</s>
<s>
It	O
's	O
used	O
for	O
describing	O
the	O
world	O
as	O
a	O
communication	O
platform	O
for	O
agents	B-General_Concept
.	O
</s>
<s>
To	O
realize	O
a	O
blackboard	B-General_Concept
in	O
a	O
computer	O
program	O
,	O
a	O
machine	B-General_Concept
readable	I-General_Concept
notation	O
is	O
needed	O
in	O
which	O
facts	O
can	O
be	O
stored	O
.	O
</s>
<s>
One	O
attempt	O
in	O
doing	O
so	O
is	O
a	O
SQL	B-Language
database	I-Language
,	O
another	O
option	O
is	O
the	O
Learnable	O
Task	O
Modeling	O
Language	O
(	O
LTML	O
)	O
.	O
</s>
<s>
The	O
syntax	O
of	O
the	O
LTML	O
planning	O
language	O
is	O
similar	O
to	O
PDDL	B-Application
,	O
but	O
adds	O
extra	O
features	O
like	O
control	O
structures	O
and	O
OWL-S	O
models	O
.	O
</s>
<s>
LTML	O
was	O
developed	O
in	O
2007	O
as	O
part	O
of	O
a	O
much	O
larger	O
project	O
called	O
POIROT	O
(	O
Plan	O
Order	O
Induction	O
by	O
Reasoning	O
from	O
One	O
Trial	O
)	O
,	O
which	O
is	O
a	O
Learning	B-Application
from	I-Application
demonstrations	I-Application
framework	O
for	O
process	B-Algorithm
mining	I-Algorithm
.	O
</s>
<s>
Here	O
is	O
a	O
small	O
example	O
:	O
A	O
human	O
user	O
is	O
executing	O
a	O
workflow	B-Operating_System
in	O
a	O
computer	O
game	O
.	O
</s>
<s>
That	O
means	O
the	O
user	O
's	O
actions	O
are	O
stored	O
in	O
a	O
logfile	B-Application
.	O
</s>
<s>
The	O
logfile	B-Application
gets	O
transformed	O
into	O
a	O
machine	B-General_Concept
readable	I-General_Concept
notation	O
which	O
is	O
enriched	O
by	O
semantic	O
attributes	B-General_Concept
.	O
</s>
<s>
The	O
result	O
is	O
a	O
textfile	B-General_Concept
in	O
the	O
LTML	O
syntax	O
which	O
is	O
put	O
on	O
the	O
blackboard	B-General_Concept
.	O
</s>
<s>
Agents	B-General_Concept
(	O
software	O
programs	O
in	O
the	O
blackboard	B-General_Concept
system	I-General_Concept
)	O
are	O
able	O
to	O
parse	O
the	O
LTML	O
syntax	O
.	O
</s>
<s>
We	O
start	O
by	O
discussing	O
two	O
well	O
known	O
early	O
blackboard	B-General_Concept
systems	I-General_Concept
,	O
BB1	O
and	O
GBB	O
,	O
below	O
and	O
then	O
discuss	O
more	O
recent	O
implementations	O
and	O
applications	O
.	O
</s>
<s>
The	O
BB1	O
blackboard	B-General_Concept
architecture	O
was	O
originally	O
inspired	O
by	O
studies	O
of	O
how	O
humans	O
plan	O
to	O
perform	O
multiple	O
tasks	O
in	O
a	O
trip	O
,	O
used	O
task-planning	O
as	O
a	O
simplified	O
example	O
of	O
tactical	O
planning	O
for	O
the	O
Office	O
of	O
Naval	O
Research	O
.	O
</s>
<s>
A	O
key	O
innovation	O
of	O
BB1	O
was	O
that	O
it	O
applied	O
this	O
opportunistic	O
planning	O
model	O
to	O
its	O
own	O
control	O
,	O
using	O
the	O
same	O
blackboard	B-General_Concept
model	I-General_Concept
of	O
incremental	O
,	O
opportunistic	O
,	O
problem-solving	O
that	O
was	O
applied	O
to	O
solve	O
domain	O
problems	O
.	O
</s>
<s>
GBB	O
improves	O
efficiency	O
by	O
allowing	O
blackboards	B-General_Concept
to	O
be	O
multi-dimensional	O
,	O
where	O
dimensions	O
can	O
be	O
either	O
ordered	O
or	O
not	O
,	O
and	O
then	O
by	O
increasing	O
the	O
efficiency	O
of	O
pattern	O
matching	O
.	O
</s>
<s>
Other	O
well-known	O
of	O
early	O
academic	O
blackboard	B-General_Concept
systems	I-General_Concept
are	O
the	O
Hearsay	O
II	O
speech	O
recognition	O
system	O
and	O
Douglas	O
Hofstadter	O
's	O
Copycat	B-Language
and	O
Numbo	O
projects	O
.	O
</s>
<s>
The	O
GTXImage	O
CAD	O
software	O
by	O
was	O
developed	O
in	O
the	O
early	O
1990s	O
using	O
a	O
set	O
of	O
rulebases	O
and	O
neural	O
networks	O
as	O
specialists	O
operating	O
on	O
a	O
blackboard	B-General_Concept
system	I-General_Concept
.	O
</s>
<s>
Adobe	O
Acrobat	O
Capture	O
(	O
now	O
discontinued	O
)	O
,	O
as	O
it	O
used	O
a	O
blackboard	B-General_Concept
system	I-General_Concept
to	O
decompose	O
and	O
recognize	O
image	O
pages	O
to	O
understand	O
the	O
objects	O
,	O
text	O
,	O
and	O
fonts	O
on	O
the	O
page	O
.	O
</s>
<s>
Details	O
of	O
a	O
similar	O
OCR	O
blackboard	B-General_Concept
for	O
Farsi	O
text	O
are	O
in	O
the	O
public	O
domain	O
.	O
</s>
<s>
Blackboard	B-General_Concept
systems	I-General_Concept
are	O
used	O
routinely	O
in	O
many	O
military	O
C4ISTAR	O
systems	O
for	O
detecting	O
and	O
tracking	O
objects	O
.	O
</s>
<s>
Another	O
example	O
of	O
current	O
use	O
is	O
in	O
Game	B-Application
AI	I-Application
,	O
where	O
they	O
are	O
considered	O
a	O
standard	O
AI	B-Application
tool	O
to	O
help	O
with	O
adding	O
AI	B-Application
to	O
video	O
games	O
.	O
</s>
<s>
Blackboard-like	O
systems	O
have	O
been	O
constructed	O
within	O
modern	O
Bayesian	O
machine	O
learning	O
settings	O
,	O
using	O
agents	B-General_Concept
to	O
add	O
and	O
remove	O
Bayesian	O
network	O
nodes	O
.	O
</s>
<s>
In	O
these	O
'	O
Bayesian	O
Blackboard	B-General_Concept
 '	O
systems	O
,	O
the	O
heuristics	O
can	O
acquire	O
more	O
rigorous	O
probabilistic	O
meanings	O
as	O
proposal	O
and	O
acceptances	O
in	O
Metropolis	B-Algorithm
Hastings	I-Algorithm
sampling	I-Algorithm
though	O
the	O
space	O
of	O
possible	O
structures	O
.	O
</s>
<s>
Conversely	O
,	O
using	O
these	O
mappings	O
,	O
existing	O
Metropolis-Hastings	B-Algorithm
samplers	O
over	O
structural	O
spaces	O
may	O
now	O
thus	O
be	O
viewed	O
as	O
forms	O
of	O
blackboard	B-General_Concept
systems	I-General_Concept
even	O
when	O
not	O
named	O
as	O
such	O
by	O
the	O
authors	O
.	O
</s>
<s>
Blackboard	B-General_Concept
systems	I-General_Concept
have	O
also	O
been	O
used	O
to	O
build	O
large-scale	O
intelligent	O
systems	O
for	O
the	O
annotation	O
of	O
media	O
content	O
,	O
automating	O
parts	O
of	O
traditional	O
social	O
science	O
research	O
.	O
</s>
<s>
In	O
this	O
domain	O
,	O
the	O
problem	O
of	O
integrating	O
various	O
AI	B-Application
algorithms	O
into	O
a	O
single	O
intelligent	O
system	O
arises	O
spontaneously	O
,	O
with	O
blackboards	B-General_Concept
providing	O
a	O
way	O
for	O
a	O
collection	O
of	O
distributed	O
,	O
modular	O
natural	B-Language
language	I-Language
processing	I-Language
algorithms	O
to	O
each	O
annotate	O
the	O
data	O
in	O
a	O
central	O
space	O
,	O
without	O
needing	O
to	O
coordinate	O
their	O
behavior	O
.	O
</s>
