NLP for CP
Addressing Constraint Programming with Natural Language Processing
Home
Resources
Publications
Correct
predictions are in
blue
. If we detect only a subset of a labelled sentence, we highlight the caught part as
blue
, the missing part
light blue.
False positives
are in
green
and
false negatives
are in
red
.
Problem Opencast_Mining — Constraint detection
A
company
has
obtained
permission
to
opencast
mine
within
a
square
plot
200
ft
times
200
ft.
The
angle
of
slip
of
the
soil
is
such
that
it
is
not
possible
for
the
sides
of
the
excavation
to
be
steeper
than
45
degrees
.
The
company
has
obtained
estimates
for
the
value
of
the
ore
in
various
places
at
various
depths
.
Bearing
in
mind
the
restrictions
imposed
by
the
angle
of
slip
,
the
company
decides
to
consider
the
problem
as
one
of
the
extracting
of
rectangular
blocks
.
Each
block
has
horizontal
dimensions
50
ft
times
50
ft
and
a
vertical
dimension
of
25
ft.
If
the
blocks
are
chosen
to
lie
above
one
another
,
as
illustrated
in
vertical
section
in
Figure
12.4
,
then
it
is
only
possible
to
excavate
blocks
forming
an
upturned
pyramid
.
If
the
estimates
of
ore
value
are
applied
to
give
values
-LRB-
in
percentage
of
pure
metal
-RRB-
for
each
block
in
the
maximum
pyramid
,
which
can
be
extracted
,
then
a
set
of
values
are
obtained
and
given
in
a
table
.
The
cost
of
extraction
increases
with
depth
.
At
successive
levels
,
the
cost
of
extracting
a
block
is
as
follows
:
3000
pounds
-LRB-
level
1
-RRB-
,
6000
pounds
-LRB-
level
2
-RRB-
,
8000
pounds
-LRB-
level
3
-RRB-
,
10000
pounds
-LRB-
level
4
-RRB-
.
The
revenue
obtained
from
a
``
100
%
value
block
''
would
be
200000
pounds
.
For
each
block
here
,
the
revenue
is
proportional
to
ore
value
.
Build
a
model
to
help
decide
the
best
blocks
to
extract
.
The
objective
is
to
maximise
revenue-cost
.
Problem Opencast_Mining — Detection of the decisions and objects to be modeled
A
company
has
obtained
permission
to
opencast
mine
within
a
square
plot
200
ft
times
200
ft.
The
angle
of
slip
of
the
soil
is
such
that
it
is
not
possible
for
the
sides
of
the
excavation
to
be
steeper
than
45
degrees
.
The
company
has
obtained
estimates
for
the
value
of
the
ore
in
various
places
at
various
depths
.
Bearing
in
mind
the
restrictions
imposed
by
the
angle
of
slip
,
the
company
decides
to
consider
the
problem
as
one
of
the
extracting
of
rectangular
blocks
.
Each
block
has
horizontal
dimensions
50
ft
times
50
ft
and
a
vertical
dimension
of
25
ft.
If
the
blocks
are
chosen
to
lie
above
one
another
,
as
illustrated
in
vertical
section
in
Figure
12.4
,
then
it
is
only
possible
to
excavate
blocks
forming
an
upturned
pyramid
.
If
the
estimates
of
ore
value
are
applied
to
give
values
-LRB-
in
percentage
of
pure
metal
-RRB-
for
each
block
in
the
maximum
pyramid
,
which
can
be
extracted
,
then
a
set
of
values
are
obtained
and
given
in
a
table
.
The
cost
of
extraction
increases
with
depth
.
At
successive
levels
,
the
cost
of
extracting
a
block
is
as
follows
:
3000
pounds
-LRB-
level
1
-RRB-
,
6000
pounds
-LRB-
level
2
-RRB-
,
8000
pounds
-LRB-
level
3
-RRB-
,
10000
pounds
-LRB-
level
4
-RRB-
.
The
revenue
obtained
from
a
``
100
%
value
block
''
would
be
200000
pounds
.
For
each
block
here
,
the
revenue
is
proportional
to
ore
value
.
Build
a
model
to
help
decide
the
best
blocks
to
extract
.
The
objective
is
to
maximise
revenue-cost
.
Back to list