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 045 (Covering arrays) — Constraint detection
The
covering
array
problem
is
formulated
as
follows
.
A
covering
array
$
CA
-LRB-
t
,
k
,
g
-RRB-
$
of
size
$
b
$
and
strength
$
t
$
,
is
a
$
k
x
b
$
array
$
A
=
-LRB-
a
_
-LCB-
ij
-RCB-
-RRB-
$
over
$
Z_g
=
-LCB-
0,1,2
,
-
,
g-1
-RCB-
$
with
the
property
that
for
any
t
distinct
rows
$
1
>
=
r_1
>
=
r_2
>
=
-
>
=
r_t
>
=
k
$
,
and
any
member
$
-LRB-
x_1
,
x_2
,
-
,
x_t
-RRB-
$
of
$
Z
^
-LCB-
t_g
-RCB-
$
there
exists
at
least
one
column
$
c
$
such
that
$
x_i
$
equals
the
$
-LRB-
r_i
,
c
-RRB-
-
th
$
element
of
$
A$
for
all
$
1
>
=
i
>
=
t
$
.
Informally
,
any
t
distinct
rows
of
the
covering
array
must
encode
column-wise
all
numbers
from
$
0
$
to
$
g
^
-LCB-
t-1
-RCB-
$
-LRB-
repititions
are
allowed
-RRB-
.
Problem 045 (Covering arrays) — Detection of the decisions and objects to be modeled
The
covering
array
problem
is
formulated
as
follows
.
A
covering
array
$
CA
-LRB-
t
,
k
,
g
-RRB-
$
of
size
$
b
$
and
strength
$
t
$
,
is
a
$
k
x
b
$
array
$
A
=
-LRB-
a
_
-LCB-
ij
-RCB-
-RRB-
$
over
$
Z_g
=
-LCB-
0,1,2
,
-
,
g-1
-RCB-
$
with
the
property
that
for
any
t
distinct
rows
$
1
>
=
r_1
>
=
r_2
>
=
-
>
=
r_t
>
=
k
$
,
and
any
member
$
-LRB-
x_1
,
x_2
,
-
,
x_t
-RRB-
$
of
$
Z
^
-LCB-
t_g
-RCB-
$
there
exists
at
least
one
column
$
c
$
such
that
$
x_i
$
equals
the
$
-LRB-
r_i
,
c
-RRB-
-
th
$
element
of
$
A$
for
all
$
1
>
=
i
>
=
t
$
.
Informally
,
any
t
distinct
rows
of
the
covering
array
must
encode
column-wise
all
numbers
from
$
0
$
to
$
g
^
-LCB-
t-1
-RCB-
$
-LRB-
repititions
are
allowed
-RRB-
.
Back to list