I
A
E
S
I
n
t
e
r
n
at
io
n
al
Jou
r
n
al
of
A
r
t
if
ic
ia
l
I
n
t
e
ll
ig
e
n
c
e
(
I
J
-
AI
)
V
ol
.
10
, N
o.
1
,
Ma
r
ch
2021
, pp.
9
~
23
I
S
S
N
:
2252
-
8938
,
D
O
I
:
10.11591/
ij
a
i.
v
10
.i
1
.pp
9
-
23
9
Jou
r
n
al
h
om
e
page
:
ht
tp
:
//
ij
ai
.
ia
e
s
c
or
e
.c
om
D
e
si
gn
an
d
an
a
l
ysi
s of
a m
u
l
t
i
-
age
n
t
e
-
l
e
ar
n
i
n
g sys
t
e
m
u
si
n
g
p
r
om
e
t
h
e
u
s d
e
si
gn
t
ool
K
e
n
n
e
d
y E
. E
h
im
w
e
n
m
a,
S
u
j
at
h
a k
r
is
h
n
am
oor
t
h
y
Department of Compu
ter Science, College o
f Science and Technology,
Wenzhou
-
Kean University
,
China
A
r
t
ic
le
I
n
f
o
A
B
S
T
R
A
C
T
A
r
ti
c
le
h
is
to
r
y
:
R
e
c
e
iv
e
d
A
pr
5
,
2020
R
e
vi
s
e
d
D
e
c
31
,
2020
A
c
c
e
pt
e
d
J
a
n
6
,
2021
Agent
unified
modeling
languages
(AUML)
are
agent
-
oriented
approaches
that
supports
the
specification,
design,
visualization
and
documentatio
n
of
an
agent
-
based
system.
This
paper
presents
the
use
of
prometheus
AUML
approach
for
the
modelin
g
of
a
pre
-
asses
sment
system
of
five
inte
ractive
agents.
The
pre
-
assessmen
t
system
,
as
previously
reported,
is
a
multi
-
agent
-
based
e
-
learning
system
that
i
s
developed
to
support
the
assessment
of
prior
learning
skills
in
students
so
as
to
classify
their
skills
and
make
re
commendat
ion
for
their
learning.
This
paper
discuss
es
the
detailed
design
approach
of
the
system
in
a
step
-
by
-
step
manner;
and
dom
ain
kno
wledge
abstracti
on
and
organizati
on
in
the
system.
In
additio
n,
the
analysis
of
the
data
collated
and
models
of
predict
ion
for
future
pre
-
assessmen
t
res
ults
are
also pres
ented.
K
e
y
w
o
r
d
s
:
A
ge
nt
m
e
th
odol
ogy
C
om
put
a
ti
ona
l
e
duc
a
ti
on
F
ir
s
t
-
or
de
r
l
ogi
c
P
r
e
-
a
s
s
e
s
s
m
e
nt
c
la
s
s
if
ic
a
ti
on
R
e
qui
r
e
m
e
nt
s
e
ngi
ne
e
r
in
g
This is an
open
acce
ss artic
le unde
r
the
CC BY
-
SA
license.
C
or
r
e
s
pon
di
n
g A
u
th
or
:
K
e
nne
dy E
. E
hi
m
w
e
nm
a
D
e
pa
r
tm
e
nt
of
C
om
put
e
r
S
c
ie
nc
e
W
e
nz
hou
-
K
e
a
n U
ni
ve
r
s
it
y
88 D
a
xue
R
d, O
uha
i
D
i
s
tr
ic
t,
W
e
nz
hou, Z
he
ji
a
ng, C
hi
na
E
m
a
il
:
ke
hi
m
w
e
n@
ke
a
n.e
du
1.
I
N
T
R
O
D
U
C
T
I
O
N
A
n
a
ge
nt
s
of
twa
r
e
m
e
th
odol
ogy
is
a
s
e
t
of
gui
de
li
ne
s
th
a
t
c
o
ve
r
s
th
e
e
nt
ir
e
li
f
e
-
c
yc
le
of
a
m
ul
ti
-
a
ge
nt
de
ve
lo
pm
e
nt
pr
oc
e
s
s
.
A
m
ul
ti
-
a
ge
nt
s
ys
te
m
(
M
A
S
)
is
a
s
ys
te
m
of
in
te
r
a
c
ti
ve
a
ge
nt
s
or
a
ut
onomous
pr
ogr
a
m
m
odul
e
s
.
I
n
ge
ne
r
a
l,
a
uni
f
ie
d
m
ode
ll
in
g
la
ngua
g
e
(
U
M
L
)
a
s
s
is
t
s
s
of
twa
r
e
d
e
ve
lo
pe
r
s
to
s
pe
c
if
y,
de
s
ig
n,
vi
s
ua
li
z
e
a
nd
doc
um
e
nt
s
of
twa
r
e
e
ngi
ne
e
r
in
g
pr
oc
e
s
s
e
s
th
a
t
m
e
e
ts
a
ppl
ic
a
ti
on
r
e
qui
r
e
m
e
nt
s
[
1]
.
A
U
M
L
a
ll
ow
m
ode
ls
to
be
c
r
e
a
te
d,
c
ons
id
e
r
e
d,
d
e
ve
lo
pe
d,
a
nd
pr
oc
e
s
s
e
d
in
a
s
t
a
nda
r
d
w
a
y
f
r
om
th
e
in
it
ia
l
pha
s
e
of
a
na
ly
s
is
to
de
s
ig
n
a
nd
im
pl
e
m
e
nt
a
ti
on
[
2]
.
S
ys
te
m
s
i
m
pl
e
m
e
nt
a
ti
on
is
f
oc
us
e
d
on
us
e
r
s
’
ne
e
ds
a
s
w
e
ll
a
s
s
ys
te
m
f
unc
ti
ona
li
ty
w
it
h
r
e
qui
r
e
m
e
nt
s
s
pe
c
if
ic
a
ti
on
a
s
th
e
dr
iv
e
r
.
F
r
o
m
s
ta
r
t
to
f
in
is
h,
e
f
f
e
c
ti
ve
a
nd
e
f
f
ic
ie
nt
s
ys
te
m
e
vol
ve
s
f
r
om
us
e
r
in
te
r
a
c
ti
on
a
nd
th
e
in
c
r
e
m
e
nt
a
l
pr
in
c
ip
le
of
de
ve
lo
pm
e
nt
.
S
o
f
twa
r
e
de
ve
lo
pm
e
nt
s
ta
g
e
s
ha
ve
s
ha
r
e
d
a
bs
tr
a
c
ti
on
in
bot
h
obj
e
c
t
-
or
i
e
nt
e
d
pr
ogr
a
m
m
in
g
(
O
O
P
)
m
e
th
odol
ogy
a
nd
a
ge
nt
-
or
ie
nt
e
d
s
of
twa
r
e
e
ngi
ne
e
r
in
g
(
A
O
S
E
)
.
I
n
O
O
P
pa
r
a
di
gm
,
th
e
s
e
s
ta
g
e
s
a
r
e
:
r
e
qui
r
e
m
e
nt
s
ga
th
e
r
in
g,
a
na
ly
s
is
,
de
s
ig
n,
im
pl
e
m
e
nt
a
ti
on,
te
s
ti
ng
a
nd
m
a
in
te
na
nc
e
.
W
hi
ls
t
th
e
A
O
S
E
pr
oc
e
s
s
s
ubs
um
e
s
th
e
s
te
ps
in
OOP
m
e
th
odol
ogi
e
s
, t
he
c
onc
e
pt
s
f
or
de
ve
lo
pi
ng obje
c
ts
(
in
O
O
P
)
a
r
e
how
e
ve
r
di
f
f
e
r
e
nt
f
r
o
m
t
hos
e
i
n a
ge
nt
-
ba
s
e
d
s
y
s
te
m
s
.
F
or
in
s
ta
nc
e
,
obj
e
c
t
-
or
ie
nt
e
d
m
e
th
odol
ogi
e
s
c
ove
r
c
onc
e
pt
s
s
uc
h
a
s
obj
e
c
ts
,
c
la
s
s
e
s
a
nd
in
he
r
it
a
nc
e
.
B
ut
in
A
O
S
E
,
de
s
ig
n
c
onc
e
pt
s
a
r
e
te
r
m
s
th
a
t
vi
e
w
a
ge
nt
s
a
s
a
ut
onomou
s
,
s
it
ua
te
d,
r
e
a
c
ti
ve
,
a
nd
s
oc
ia
l.
T
hi
s
pa
pe
r
i
s
a
pr
e
s
e
nt
a
ti
on of
t
he
a
ppl
ic
a
ti
on of
pr
om
e
th
e
us
[
3, 4]
a
ge
nt
-
or
ie
nt
e
d m
e
th
odol
ogy f
or
t
he
s
ta
ti
c
a
nd
dyn
a
m
ic
de
s
ig
n
of
a
n
e
le
a
r
ni
ng
M
A
S
.
T
hough
th
e
r
e
a
r
e
s
e
ve
r
a
l
A
O
S
E
m
e
th
odol
ogi
e
s
f
or
de
s
ig
ni
ng
a
ge
nt
-
ba
s
e
d
s
y
s
te
m
s
,
th
e
c
hoi
c
e
of
pr
om
e
th
e
us
w
a
s
pr
e
di
c
a
te
d
on
it
s
s
tr
uc
tu
r
e
d
a
nd
de
ta
il
e
d
s
te
p
-
by
-
s
te
p
pr
oc
e
dur
e
th
a
t
s
uppor
ts
how
r
e
qui
r
e
m
e
nt
s
t
a
te
m
e
nt
s
c
a
n
be
a
c
qui
r
e
d. T
he
pur
pos
e
of
th
e
s
y
s
te
m
is
to
pr
e
-
a
s
s
e
s
s
s
tu
de
nt
s
’
pr
io
r
le
a
r
ni
ng,
c
la
s
s
if
y
th
e
ir
s
ki
ll
s
,
a
nd
m
a
ke
r
e
c
om
m
e
nda
ti
on
f
or
a
ppr
opr
ia
te
m
a
te
r
ia
l
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
10
, N
o.
1
,
M
a
r
c
h
20
2
1
:
9
–
23
10
s
ui
ta
bl
e
to
th
e
ir
ne
e
ds
.
T
hus
,
th
e
c
ont
r
ib
ut
io
n
of
th
is
pa
pe
r
is
:
i)
T
o
de
m
ons
tr
a
te
r
e
qui
r
e
m
e
nt
s
a
na
ly
s
is
a
nd
de
s
ig
n
s
p
e
c
if
ic
a
ti
ons
f
or
th
e
d
e
v
e
lo
pm
e
nt
of
a
n
e
-
le
a
r
ni
ng
pr
e
-
a
s
s
e
s
s
m
e
nt
s
y
s
te
m
u
s
in
g
M
A
S
.
ii
)
T
o
a
na
ly
s
e
th
e
de
s
c
r
ip
ti
ve
f
unc
ti
ons
a
nd r
ol
e
s
of
m
ul
ti
-
a
ge
nt
s
w
it
hi
n a
n e
-
l
e
a
r
ni
ng pr
e
-
a
s
s
e
s
s
m
e
nt
s
ys
te
m
. i
ii
)
T
o s
how
a
d
e
ta
il
e
d
m
ode
l
of
s
of
twa
r
e
e
ngi
n
e
e
r
in
g
w
it
h
a
g
e
nt
U
M
L
(
A
U
M
L
)
to
ol
f
or
te
a
c
hi
ng
a
nd
le
a
r
ni
ng.
iv
)
T
o
de
m
ons
tr
a
te
in
te
r
-
a
ge
nt
c
om
m
uni
c
a
ti
on
f
or
th
e
a
s
s
e
s
s
m
e
nt
s
a
nd
c
la
s
s
if
ic
a
ti
on
of
s
tu
de
nt
s
'
pr
io
r
s
ki
ll
-
s
e
t.
v)
T
o
a
na
ly
s
e
th
e
da
t
a
c
ol
la
te
d
f
r
om
th
e
s
ys
te
m
us
in
g
r
e
gr
e
s
s
io
n
m
ode
ls
of
pr
e
di
c
ti
on.
T
hi
s
pa
pe
r
c
ont
in
ue
s
w
it
h
th
e
b
a
c
kgr
ound
lo
gi
c
of
knowle
dge
e
ngi
ne
e
r
in
g
f
or
th
e
s
ys
te
m
in
w
hi
c
h
a
n
a
b
s
tr
a
c
t
m
ode
l
of
a
n
ont
ol
ogy tr
e
e
t
r
a
ve
r
s
a
l
is
di
s
c
us
s
e
d a
s
a
ppl
ic
a
bl
e
i
n t
he
M
A
S
i
m
pl
e
m
e
nt
a
ti
on. I
n s
e
c
ti
on 2, the
pa
pe
r
pr
e
s
e
nt
s
A
U
M
L
to
ol
s
a
nd
a
ge
nt
s
of
twa
r
e
de
v
e
lo
pm
e
nt
li
f
e
c
yc
le
(
A
S
D
L
C
)
.
I
n
s
e
c
ti
on
3,
m
ode
ls
of
a
na
ly
s
is
a
nd
de
s
ig
n
f
r
om
th
e
us
e
of
th
e
pr
om
e
th
e
us
de
s
ig
n
to
ol
(
P
D
T
)
is
p
r
e
s
e
nt
e
d.
S
e
c
ti
on
4
lo
oks
in
to
im
pl
e
m
e
nt
a
ti
on,
is
s
ue
s
a
t
e
xpe
r
im
e
nt
a
ti
on, da
ta
c
ol
le
c
ti
on a
nd a
na
ly
s
i
s
;
a
nd s
e
c
t
io
n 5 i
s
c
onc
lu
s
io
n.
L
ogi
c
pr
ogr
a
m
i
n de
c
is
io
n s
uppor
t
s
ys
te
m
L
ogi
c
f
or
m
ul
a
s
a
r
e
f
or
m
a
l
s
pe
c
if
ic
a
ti
ons
th
a
t
a
r
e
r
e
a
di
ly
us
e
d
to
r
e
pr
e
s
e
nt
f
a
c
ts
,
s
ta
te
m
e
nt
s
a
nd
pr
opos
it
io
ns
.
S
uc
h
f
or
m
a
li
s
m
e
.g.,
H
or
n
c
la
u
s
e
s
,
a
ns
w
e
r
s
e
t
pr
ogr
a
m
m
in
g,
f
ir
s
t
or
de
r
lo
gi
c
f
or
m
ul
a
s
a
nd
de
s
c
r
ip
ti
on
lo
gi
c
s
a
r
e
u
s
e
d
to
f
or
r
e
a
s
oni
ng
-
s
uppor
te
d
d
e
c
is
io
n
pr
oc
e
s
s
e
s
in
a
dyna
m
ic
s
ys
te
m
.
W
it
h
lo
gi
c
pr
ogr
a
m
s
a
nd
it
s
di
ve
r
s
e
va
r
ia
nt
s
of
f
or
m
a
li
z
a
ti
on,
f
a
c
ts
a
nd
obj
e
c
ts
ha
ve
be
e
n
c
ol
le
c
te
d,
c
a
te
gor
iz
e
d,
a
nd
r
e
la
ti
ons
e
s
t
a
bl
is
he
d
in
-
be
twe
e
n
obj
e
c
ts
of
f
a
c
ts
;
a
nd
de
c
is
i
ons
ta
ke
n
e
.g
.,
[
5,
6]
.
A
ppr
oa
c
he
s
f
or
M
A
S
de
ve
lo
pm
e
nt
us
in
g
f
ir
s
t
or
de
r
lo
gi
c
(
F
O
L
)
ha
ve
a
ls
o
be
e
n
de
m
ons
tr
a
te
d
in
[
7,
8]
.
I
n
a
hyb
r
id
di
s
tr
ib
ut
e
d
s
ys
te
m
in
w
hi
c
h
a
s
ym
pt
ot
ic
c
ons
e
ns
u
s
r
e
s
ul
t
w
a
s
obt
a
in
e
d,
[
7
]
pr
e
s
e
nt
e
d
a
le
a
d
e
r
-
f
ol
lo
w
e
r
c
ons
e
ns
us
M
A
S
in
w
hi
c
h
th
e
le
a
de
r
s
ys
te
m
s
ha
r
e
d
knowle
dge
w
it
h
th
e
f
ol
lo
w
e
r
s
ys
te
m
w
hos
e
de
s
c
r
ip
ti
on
w
a
s
gi
ve
n
in
F
O
L
.
S
im
il
a
r
ly
,
r
e
c
e
nt
s
tu
di
e
s
in
M
A
S
e
.g.
[8
-
11]
ha
ve
e
m
pha
s
iz
e
d
th
e
n
e
e
d
f
or
a
da
pt
iv
e
e
l
e
a
r
ni
ng
s
ys
te
m
s
th
a
t
c
a
n
pe
r
s
ona
li
z
e
le
a
r
ni
ng
s
o
a
s
to
m
e
e
t
in
di
vi
dua
l
le
a
r
ne
r
ne
e
d
s
.
T
hi
s
is
b
e
c
a
u
s
e
w
ha
t
a
le
a
r
ne
r
w
a
nt
s
,
m
a
y
a
c
tu
a
ll
y,
be
di
f
f
e
r
e
nt
f
r
om
w
ha
t
he
n
e
e
ds
to
le
a
r
n.
T
hi
s
r
e
s
e
a
r
c
h
a
ddr
e
s
s
e
s
th
is
ga
p
in
th
e
de
ve
lo
pm
e
nt
of
el
e
a
r
ni
ng s
ys
te
m
s
.
B
a
c
kgr
ound logi
c
of
knowle
dge
e
ngi
ne
e
r
in
g f
or
t
he
pr
e
-
a
s
s
e
s
s
m
e
nt
s
ys
te
m
L
e
t
be
a
pr
e
di
c
a
te
.
A
bi
na
r
y
r
e
la
ti
on
be
twe
e
n
obj
e
c
ts
1
a
nd
2
c
a
n
be
gi
ve
n
s
ym
bol
ic
a
ll
y
a
s
(
1
,
2
)
.
A
ls
o,
le
t
ⅅ
b
e
a
dom
a
in
of
di
r
e
c
te
d
ont
ol
ogi
c
a
l
node
s
,
a
nd
i,
j,
a
nd
k
=
1,
2,
3,
…
,
n,
n
+
1,
r
e
s
pe
c
ti
ve
ly
.
I
f
i
r
e
pr
e
s
e
nt
s
th
e
le
ve
l
of
hi
e
r
a
r
c
hi
e
s
of
no
de
s
on
th
e
hor
iz
ont
a
l
t
r
a
ve
r
s
a
l,
a
nd
j
th
e
a
r
r
a
nge
m
e
nt
s
of
node
s
on
th
e
v
e
r
ti
c
a
l
tr
a
ve
r
s
a
l
a
nd
,
s
om
e
r
a
nd
om
pr
e
di
c
a
te
s
or
pr
ope
r
ty
s
u
c
h
th
a
t
,
is
a
pa
r
e
nt
node
,
+
1
,
+
1
a
pe
r
qui
s
it
e
pa
r
e
nt
node
n
e
xt
to
,
,
a
nd
,
a
le
a
f
node
;
th
e
n
th
e
f
ol
lo
w
in
g
a
bs
tr
a
c
ti
on
hol
ds
:
Ɐ
x ϵ
ⅅ
,
,
(
,
)
, w
hi
c
h s
ta
te
s
t
h
a
t,
e
ve
r
y node
ha
s
a
p
r
ope
r
ty
.
Ɐ
x
ϵ
ⅅ
,
,
(
,
,
+
1
,
±
1
)
,
w
hi
c
h
m
e
a
ns
a
par
e
nt
to
pr
e
r
e
qui
s
it
e
par
e
nt
node
r
e
la
ti
on.
A
ny
pa
r
e
nt
node
is
a
de
s
ir
e
d t
opi
c
t
o be
l
e
a
r
ne
d by s
tu
de
nt
s
.
Ɐ
x ϵ
ⅅ
,
+
1
,
(
,
,
,
)
, t
ha
t
th
e
r
e
i
s
a
di
r
e
c
t
r
e
la
ti
on of
a
par
e
nt
node
to
i
ts
o
w
n l
e
af
node
s
.
Ɐ
xⱯ
z
ϵ
ⅅ
,
,
(
,
,
+
1
,
±
1
)
ꓥ
+
1
,
±
1
(
+
1
,
±
1
,
±
1
,
)
→
+
1
,
±
1
(
,
,
±
1
,
)
, w
hi
c
h i
s
a
t
r
a
ns
it
iv
e
r
e
la
ti
on f
or
na
vi
ga
ti
ng
le
af
node
s
c
onne
c
te
d t
o
pr
e
r
e
qui
s
it
e
pa
r
e
nt
node
s
, or
Ɐ
xⱯ
z
ϵ
ⅅ
,
,
(
,
,
+
1
,
±
1
)
ꓥ
+
1
,
±
1
(
+
1
,
±
1
,
±
1
,
)
→
,
±
1
(
,
,
±
1
,
)
, w
hi
c
h a
r
e
t
r
a
ns
it
iv
e
r
e
la
ti
ons
f
or
t
r
a
ve
r
s
a
ls
of
le
af
node
s
c
onne
c
t
e
d t
o
par
e
nt
node
s
.
W
it
h a
t
r
e
e
di
a
gr
a
m
a
s
s
how
n i
n
F
ig
ur
e
1
, t
he
a
bove
s
ta
te
d s
ym
bol
ic
r
e
la
ti
ons
a
r
e
f
u
r
th
e
r
de
duc
e
d a
s
f
ol
lo
w
s
:
T
ha
t
th
e
pa
r
e
nt
node
s
or
th
e
ob
je
c
ts
of
le
a
r
ni
ng
a
r
e
th
e
c
1,
c
2,
c
3,
c
4,
c
5,
a
nd
c
6.
A
lo
ng
th
e
hor
iz
ont
a
l
tr
a
ve
r
s
a
l,
L
e
v
e
l
1
to
L
e
v
e
l
4
,
th
e
pr
e
c
e
di
ng
node
e
.g.,
c
1
is
a
pa
r
e
nt
to
c
2
a
nd
c
3;
a
nd
c
2
a
nd
c
3
a
r
e
in
tu
r
n
pe
r
qui
s
it
e
s
node
s
t
o c
1. T
he
n, c
2 a
nd c
3 a
r
e
pa
r
e
nt
node
s
t
o
c
4
a
nd c
5, r
e
s
pe
c
ti
ve
ly
;
a
nd c
4 a
nd c
5 a
r
e
i
n t
u
r
n
pr
e
r
e
qui
s
it
e
s
to
c
2
a
nd
c
3,
r
e
s
pe
c
ti
ve
ly
.
F
ur
th
e
r
dow
n,
c
4
is
a
pa
r
e
nt
to
c
6,
a
nd
c
6
in
tu
r
n
i
s
a
pr
e
r
e
qui
s
it
e
to
c
4. I
n t
he
t
r
e
e
, t
he
s
e
t
of
∈
C
ha
ve
t
he
ir
r
e
s
pe
c
ti
ve
l
e
a
f
node
s
la
be
ll
e
d a
s
N
1
, N
2
, N
3
, …
, N
11
a
r
e
l
e
a
f
node
s
∈
.
E
s
ta
bl
is
hi
ng na
vi
ga
ti
ona
l
r
e
la
ti
ons
be
twe
e
n node
s
O
n
th
e
ba
s
is
of
th
e
tr
e
e
a
s
s
how
n
in
F
ig
ur
e
1,
w
e
now
s
how
t
he
r
e
la
ti
ons
hi
p
be
twe
e
n
nod
e
s
in
th
e
tr
e
e
a
nd
th
e
s
ym
bol
ic
a
xi
om
s
s
ta
te
d
e
a
r
li
e
r
a
s
im
pl
e
m
e
nt
e
d
o
n
th
e
m
ul
ti
-
a
ge
nt
pr
e
-
a
s
s
e
s
s
m
e
nt
s
ys
te
m
th
a
t:
,
(
,
)
de
s
c
r
ib
e
s
a
una
r
y
pr
e
di
c
a
te
,
a
nd a
n
e
xa
m
pl
e
is
de
s
i
r
e
dC
onc
e
pt
(
c1
)
;
a
nd
,
(
,
,
+
1
,
±
1
)
de
s
c
r
ib
e
s
a
bi
na
r
y
r
e
la
ti
on
w
hi
c
h
s
ta
te
s
th
a
t
a
pa
r
e
nt
ha
s
a
na
m
e
d
p
r
e
r
e
qui
s
it
e
,
a
nd
a
n
e
x
a
m
pl
e
of
th
e
f
or
m
is
has
P
r
e
r
e
qui
s
it
e
(
c
1,
c
2
)
;
a
nd
+
1
,
(
+
1
,
,
,
)
w
hi
c
h
s
ta
te
s
th
a
t
a
pa
r
e
nt
ha
s
a
na
m
e
d
le
a
f
node
,
a
nd
a
n
e
xa
m
pl
e
is
f
or
m
ul
a
has
K
B
(
c
6,
n
11
)
.
T
he
n
f
or
a
ll
pr
e
-
a
s
s
e
s
s
m
e
nt
a
nd
r
e
c
om
m
e
nda
ti
on
of
a
ny
fa
il
e
d
le
a
r
ni
ng
uni
t
i.
e
., t
he
l
e
a
f
node
s
N
, w
e
s
ta
te
t
ha
t
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
D
e
s
ig
n and analys
i
s
of
a m
ul
ti
-
age
nt
e
-
le
ar
ni
ng s
y
s
te
m
us
in
g
p
r
om
e
th
e
us
…
(
K
e
nne
d
y
E
. E
hi
m
w
e
nm
a
)
11
,
(
,
,
+
1
,
±
1
)
ꓥ
+
1
,
±
1
(
+
1
,
±
1
,
±
1
,
)
→
+
1
,
±
1
(
,
,
±
1
,
)
.
ax
io
m
1
A
x
io
m
1
s
ta
te
s
th
a
t
if
a
pa
r
e
nt
node
,
ha
s
a
na
m
e
d
pr
e
r
e
qui
s
it
e
+
1
,
±
1
(
one
le
ve
l
be
lo
w
th
e
hi
e
r
a
r
c
hy)
of
,
e
it
he
r
on
it
s
r
ig
ht
ha
nd
or
le
f
t
-
ha
nd
s
id
e
w
hi
c
h
is
de
not
e
d
by
±
1
(
+
f
or
r
ig
h
t,
a
nd
–
f
or
le
f
t)
,
a
nd
th
e
na
m
e
d
pr
e
r
e
qui
s
it
e
ha
s
a
na
m
e
d
le
a
f
node
±
1
,
,
th
e
n
th
e
pa
r
e
nt
nod
e
ha
s
a
di
r
e
c
t
r
e
la
ti
on
w
it
h
th
e
le
a
f
node
li
ke
th
e
pe
r
qui
s
it
e
ha
s
.
A
n
e
xa
m
pl
e
of
th
is
tr
a
ns
it
iv
e
c
lo
s
ur
e
i
s
has
P
r
e
r
e
qui
s
it
e
(
c
4,
c
6)
ꓥ
has
K
B
(
c
6,
n
11
)
→
has
K
B
(
c
4,
n
11
)
th
a
t
s
a
ti
s
f
ie
s
th
e
pr
ope
r
ty
of
tr
an
s
it
iv
it
y
.
I
n
a
ddi
ti
on,
th
e
ax
io
m
1
c
onve
ys
th
e
le
a
f
node
s
±
1
,
th
a
t
a
r
e
:
i)
pr
e
-
a
s
s
e
s
s
e
d
upon,
a
nd
ii)
th
e
node
s
th
a
t
a
r
e
r
e
c
om
m
e
nde
d
w
he
n
a
ny
l
e
a
f
node
N
th
a
t
is
c
onn
e
c
te
d
to
t
he
pr
e
r
e
qui
s
it
e
node
+
1
,
±
1
a
r
e
fa
il
e
d
. O
n t
he
ot
he
r
ha
nd, t
he
c
ount
e
r
pa
r
t
ax
io
m
2
,
(
,
,
+
1
,
±
1
)
ꓥ
+
1
,
±
1
(
+
1
,
±
1
,
±
1
,
)
→
+
1
,
(
,
,
±
1
,
)
ax
io
m
2
is
th
e
ax
io
m
th
a
t
a
ls
o
s
a
ti
s
f
ie
s
th
e
pr
ope
r
ty
o
f
tr
ans
it
iv
it
y
.
I
n
th
is
c
a
s
e
,
it
is
f
or
th
e
r
e
c
o
m
m
e
nda
ti
on
o
f
le
a
f
node
s
±
1
,
th
a
t
ha
s
di
r
e
c
t
r
e
la
ti
ons
to
th
e
de
s
ir
e
d
to
pi
c
’
s
,
gi
ve
n
th
a
t
a
n
e
pi
s
ode
of
pr
e
-
a
s
s
e
s
s
m
e
nt
on
th
e
pe
r
qui
s
it
e
±
1
,
c
onne
c
te
d
to
+
1
,
±
1
ha
ve
a
ll
be
e
n
a
tt
e
m
pt
e
d
a
nd
a
r
e
a
ll
pas
s
e
d
.
A
n
e
x
a
m
pl
e
of
th
is
lo
gi
c
a
l
a
xi
om
2 i
s
has
P
r
e
r
e
qui
s
it
e
(
c
1, c
3)
ꓥ
has
K
B
(
c
3, {n4,n5, n6})
→
has
K
B
(
c
1, n1)
.
F
ig
ur
e
1. A
knowle
dge
gr
a
ph of
m
ul
ti
pl
e
hor
iz
ont
a
l
a
nd ve
r
ti
c
a
l
tr
a
ve
r
s
a
l
I
n
our
a
ge
nt
-
ba
s
e
d
pr
e
-
a
s
s
e
s
s
m
e
nt
s
ys
t
e
m
,
a
ge
nt
s
n
e
e
d
to
c
om
m
uni
c
a
te
th
e
gr
ound
f
a
c
t
r
e
pr
e
s
e
nt
a
ti
on
of
th
is
lo
gi
c
a
l
a
xi
om
s
.
F
or
in
s
ta
nc
e
,
f
or
a
n
a
ge
nt
to
r
e
s
ol
ve
th
e
r
e
le
va
nt
pl
a
ns
f
or
th
e
ir
ne
xt
a
c
ti
on,
th
is
gr
oup
of
a
ge
nt
s
m
u
s
t
in
te
r
-
c
om
m
uni
c
a
te
th
e
de
s
i
r
e
d
to
pi
c
s
,
pa
s
s
e
d
le
af
node
s
a
nd/
or
fa
il
e
d
le
af
node
s
us
in
g
th
e
f
ol
lo
w
in
g
pr
e
di
c
a
te
lo
gi
c
f
or
m
pas
s
e
d(
+
1
,
)
a
nd
fa
il
e
d(
±
1
,
)
.
T
he
p
r
e
d
ic
a
te
s
w
hi
c
h
a
r
e
th
e
a
c
ti
ons
ta
ke
n
by t
he
m
ul
ti
-
a
ge
nt
s
ba
s
e
d on s
tu
de
nt
s
’
r
e
s
pons
e
t
o que
s
ti
ons
, f
or
m
t
he
ba
s
is
of
th
e
f
a
c
ts
a
bout
t
he
out
c
om
e
o
f
a
s
tu
de
nt
u
s
in
g
lo
gi
c
pr
ogr
a
m
m
in
g.
T
hi
s
i
s
be
c
a
us
e
a
ny
obj
e
c
t
ha
s
a
pr
ope
r
ty
th
a
t
it
s
a
ti
s
f
ie
s
or
th
a
t
a
ny
obj
e
c
t
is
c
onne
c
te
d
by
s
om
e
r
e
la
ti
on
to
a
not
he
r
obj
e
c
t.
F
r
om
t
he
f
or
e
goi
ng,
th
e
e
xpl
ic
it
ly
s
ta
te
d
lo
gi
c
-
ba
s
e
d
f
or
m
u
la
s
a
r
e
th
e
pr
e
m
is
e
s
on w
hi
c
h
th
e
m
ul
ti
-
a
ge
nt
of
th
e
pr
e
-
a
s
s
e
s
s
m
e
nt
s
ys
te
m
in
te
r
a
c
ts
,
s
e
le
c
t
ont
ol
ogi
c
a
l
node
s
,
s
e
le
c
t
que
s
ti
ons
a
s
s
oc
ia
te
d
w
it
h
le
a
f
node
s
,
a
s
s
e
s
s
u
s
e
r
s
,
c
la
s
s
if
y
us
e
r
s
ki
ll
s
a
nd
r
e
c
om
m
e
nd
le
a
r
ni
ng
m
a
te
r
ia
ls
.
2.
R
E
L
A
T
E
D
WO
R
K
O
N
A
U
M
L
A
N
D
E
-
LE
A
R
N
I
N
G
S
Y
S
T
E
M
S
A
U
M
L
a
nd
A
U
M
L
di
a
gr
a
m
s
a
r
e
a
r
c
hi
te
c
tu
r
a
l
m
od
e
l
f
or
in
f
or
m
a
ti
on
s
ys
te
m
s
de
ve
lo
pm
e
nt
.
A
U
M
L
in
pa
r
ti
c
ul
a
r
,
ha
s
be
e
n
us
e
d
in
th
e
de
s
ig
n
a
nd
s
pe
c
if
ic
a
ti
on
of
s
ys
te
m
s
in
th
e
a
r
e
a
s
of
w
e
a
th
e
r
f
or
e
c
a
s
ti
ng,
bus
in
e
s
s
tr
a
di
ng,
pe
tr
i
-
ne
t,
m
in
e
r
r
obot
;
a
nd
a
ge
nt
ne
got
ia
t
io
n
be
f
or
e
im
pl
e
m
e
nt
a
ti
on.
I
n
th
e
f
ie
ld
of
e
duc
a
ti
ona
l
s
y
s
te
m
s
,
a
ge
nt
-
ba
s
e
d
s
y
s
te
m
de
ve
lo
pm
e
nt
r
e
s
e
a
r
c
h
on
in
te
ll
ig
e
nt
a
ge
nt
m
ode
ls
f
or
e
le
a
r
ni
ng
ha
ve
r
e
c
e
iv
e
d
li
tt
le
a
tt
e
nt
io
n.
A
m
ongs
t
th
is
f
e
w
a
r
e
th
e
w
or
ks
of
[
12
]
a
nd
[
13]
.
I
n
[
12]
,
G
a
ia
m
e
th
odol
ogy
w
a
s
a
ppl
ie
d
in
th
e
a
na
ly
s
is
a
nd
d
e
s
ig
n
of
a
n
e
le
a
r
ni
ng
ba
s
e
d
M
A
S
.
T
he
s
y
s
te
m
w
hi
c
h
w
a
s
pr
opos
e
d
f
or
im
pl
e
m
e
nt
a
ti
on
on
J
A
D
E
f
r
a
m
e
w
or
k
is
a
s
e
c
ur
it
y
-
ba
s
e
d
M
A
S
th
a
t
w
a
s
m
e
a
nt
to
de
te
c
t
th
r
e
a
ts
a
nd
pr
ovi
de
pr
ot
e
c
ti
on
on
w
e
b
-
ba
s
e
d
le
a
r
ni
ng
m
a
na
g
e
m
e
nt
s
y
s
te
m
s
(
L
M
S
)
s
uc
h
a
s
M
oodl
e
.
I
n
[
13]
a
n
a
na
ly
s
i
s
a
nd
de
s
ig
n
of
a
w
e
b
-
ba
s
e
d
s
e
r
vi
c
e
s
f
or
vi
r
tu
a
l
le
a
r
ni
ng
e
nvi
r
on
m
e
nt
(
V
L
E
)
w
a
s
a
ls
o
di
s
c
us
s
e
d.
T
he
pa
pe
r
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
10
, N
o.
1
,
M
a
r
c
h
20
2
1
:
9
–
23
12
c
onc
e
pt
ua
li
z
e
d
a
V
L
E
a
ppl
ic
a
ti
on
on
m
obi
le
a
ge
nt
te
c
hnol
og
y
f
or
th
e
a
s
s
e
s
s
m
e
nt
of
s
tu
de
nt
s
’
knowle
dg
e
,
a
nd
de
s
c
r
ib
e
d
a
ge
nt
r
ol
e
a
nd
a
ge
nt
in
te
r
a
c
ti
on
us
in
g
a
U
M
L
to
ol
,
a
nd
f
in
a
ll
y
to
im
pl
e
m
e
nt
a
ti
on
us
in
g
J
A
D
E
.
T
he
I
nf
oS
ta
ti
on
s
ys
te
m
[
2]
is
a
pr
oj
e
c
t
of
di
s
tr
ib
ut
e
d
e
le
a
r
ni
ng
c
e
nt
r
e
(
D
e
L
C
)
,
a
ls
o
us
e
d
m
ul
ti
-
a
ge
nt
te
c
hnol
ogy
w
it
h
pr
opos
e
d
im
pl
e
m
e
nt
a
ti
on
on
J
A
D
E
[
14
]
.
W
it
h
a
U
M
L
,
[
2
]
de
s
c
r
ib
e
d
th
e
I
n
f
oS
ta
ti
on
s
ys
te
m
a
s
a
s
y
s
te
m
of
in
te
r
a
c
ti
ve
a
ge
nt
s
w
ho
s
e
f
unc
ti
ons
in
c
lu
de
d
de
s
ig
na
te
d
e
-
s
e
r
vi
c
e
s
.
A
ls
o,
in
[
15]
th
e
A
G
I
L
E
-
P
A
S
S
I
m
e
th
od
ol
ogy
w
a
s
r
e
por
te
d
a
s
th
e
d
e
ve
lo
pm
e
nt
to
ol
f
or
a
m
e
di
c
a
l
e
du
c
a
ti
ona
l
ga
m
e
c
a
ll
e
d
M
E
D
E
D
U
C
f
or
th
e
pur
pos
e
of
im
pr
ovi
ng
le
a
r
ni
ng
in
m
e
di
c
a
l
e
duc
a
ti
on
a
nd
c
li
ni
c
a
l
p
e
r
f
or
m
a
nc
e
.
A
s
a
,
M
E
D
E
D
U
C
a
ll
ow
e
d
s
tu
de
nt
s
to
a
ns
w
e
r
que
s
ti
ons
a
t
di
f
f
e
r
e
nt
le
ve
l
of
di
f
f
ic
ul
t
y
on
m
ul
ti
m
e
di
a
pr
e
s
e
nt
a
ti
on.
W
hi
le
m
a
ny
a
ppl
ic
a
ti
ons
on
a
g
e
nt
-
ba
s
e
d
te
c
hnol
ogy
a
r
e
de
ve
l
ope
d
in
f
ie
ld
s
s
uc
h
a
s
c
om
m
e
r
c
e
a
nd
s
e
c
ur
it
y,
or
a
da
pt
iv
e
dyna
m
ic
pr
ogr
a
m
m
in
g
[
16]
ve
r
y
li
m
it
e
d
a
tt
e
nt
io
n
h
a
s
be
e
n
gi
ve
n
to
a
ge
nt
-
ba
s
e
d
de
ve
lo
pm
e
nt
f
or
s
tu
de
nt
le
a
r
ni
ng.
A
m
ong
th
e
a
f
o
r
e
m
e
nt
io
ne
d
f
e
w
,
none
ha
d
th
e
c
om
bi
ne
d
s
ys
te
m
goa
l
of
s
ki
ll
s
c
la
s
s
if
ic
a
ti
on
a
nd r
e
c
om
m
e
nda
ti
on of
l
e
a
r
ni
ng ma
te
r
ia
ls
t
ha
t
w
e
a
r
e
pr
e
s
e
nt
in
g i
n t
hi
s
pa
pe
r
.
2.1. Age
n
t
-
or
ie
n
t
e
d
m
e
t
h
od
ol
ogy
M
e
th
odol
ogi
e
s
a
s
a
pr
oc
e
s
s
of
e
ngi
ne
e
r
in
g
a
s
of
twa
r
e
e
na
bl
e
s
de
ve
lo
pe
r
s
to
c
onc
r
e
ti
s
e
th
e
va
r
io
us
in
te
r
a
c
ti
on c
om
pone
nt
s
of
a
s
ys
t
e
m
a
nd t
he
f
unc
ti
ons
ne
e
d
e
d a
m
ongs
t
th
e
va
r
io
us
c
om
pone
nt
s
f
or
a
s
y
s
te
m
t
o
be
r
e
a
li
s
e
d.
T
he
w
or
k
of
[
17]
a
c
knowle
dge
d
th
a
t
s
e
ve
r
a
l
A
O
S
E
m
e
th
odol
ogi
e
s
e
xi
s
ts
f
or
th
e
a
na
ly
s
i
s
a
nd
de
s
ig
n
of
M
A
S
but
th
e
r
e
a
l
s
o
e
xi
s
ts
di
f
f
ic
ul
ty
in
th
e
c
hoi
c
e
of
th
e
a
ppr
opr
ia
te
m
e
th
odol
ogy
f
or
s
of
twa
r
e
s
ol
ut
io
ns
f
r
om
dom
a
in
to
dom
a
in
.
F
or
in
s
ta
nc
e
,
a
m
ongs
t
ga
ia
[
18
-
21]
,
tr
opos
[
22,
23]
,
M
a
S
E
[
24
]
,
P
A
S
S
I
[
25, 26]
a
nd p
r
om
e
th
e
us
[
27]
m
e
th
odol
ogi
e
s
w
ha
t
f
a
c
to
r
s
houl
d
i
nf
or
m
th
e
de
ve
lo
pe
r
’
s
c
hoi
c
e
?
T
hough the
s
e
m
e
th
odol
ogi
e
s
s
how
s
im
il
a
r
it
ie
s
in
th
e
ir
de
s
ig
n
pr
oc
e
s
s
,
th
e
r
e
a
r
e
how
e
ve
r
a
va
r
yi
ng
de
gr
e
e
of
di
f
f
e
r
e
nc
e
s
:
F
r
om
r
e
qui
r
e
m
e
nt
s
a
na
ly
s
is
,
th
r
ough
to
f
unc
ti
ona
li
ty
m
ode
l
li
ng
f
or
a
ge
nt
s
,
a
nd
im
pl
e
m
e
nt
a
ti
on.
I
n
th
e
f
ol
lo
w
in
g
s
e
c
ti
on,
P
r
o
m
e
th
ous
is
pr
e
s
e
nt
e
d;
a
nd
in
T
a
bl
e
1
is
a
c
om
pa
r
a
ti
ve
s
um
m
a
r
y
of
G
a
ia
,
T
r
opos
a
nd
P
r
om
e
th
e
us
m
e
th
odol
ogi
e
s
w
it
h
r
e
ga
r
ds
to
th
e
ir
s
im
il
a
r
it
ie
s
a
nd
di
f
f
e
r
e
nc
e
s
,
a
nd
th
e
ba
s
i
s
upon
w
hi
c
h
th
e
P
r
om
e
th
e
us
m
e
th
odol
ogy
w
a
s
c
hoo
s
in
g
f
or
th
e
de
s
ig
n
of
th
e
pr
e
-
a
s
s
e
s
s
m
e
nt
s
ys
te
m
.
A
s
s
how
n
in
T
a
bl
e
1,
one
c
om
m
on
s
im
il
a
r
it
y
is
th
e
c
us
to
m
is
e
d
to
ol
a
s
s
o
c
ia
te
d
w
it
h
e
a
c
h
m
e
th
odol
ogy
to
s
uppor
t
th
e
ir
e
ngi
ne
e
r
in
g
pr
oc
e
s
s
.
T
h
e
di
f
f
e
r
e
nc
e
in
th
e
ir
r
e
s
pe
c
ti
ve
e
ngi
ne
e
r
in
g
pr
oc
e
s
s
c
a
n
b
e
f
ound
in
th
e
ir
de
s
ig
n
s
te
ps
.
F
or
in
s
ta
nc
e
,
th
e
tr
opo
s
c
onc
e
pt
of
s
of
tg
oal
s
w
hi
c
h
is
e
qui
va
le
nt
t
o
s
ubgoals
in
pr
om
e
th
e
us
is
a
br
e
a
kdown
of
har
dgoals
a
nd
in
it
ia
l
goal
of
a
ge
nt
s
(
or
a
c
to
r
s
)
f
unc
ti
ona
li
ti
e
s
,
r
e
s
pe
c
ti
ve
ly
.
T
hi
s
,
[
28]
r
e
f
e
r
r
e
d
to
a
s
r
ol
e
de
c
om
pos
it
io
n
w
hi
c
h
r
e
duc
e
s
th
e
c
om
pl
e
xi
ty
in
M
A
S
e
ngi
ne
e
r
in
g.
T
he
ba
s
i
s
f
or
pr
om
e
th
e
us
is
th
e
de
s
ig
n
s
te
p f
or
de
r
iv
in
g
in
it
ia
l
goal
s
.
T
a
bl
e
1.
A
c
om
pa
r
a
ti
ve
s
um
m
a
r
y of
ga
ia
, t
r
opos
a
nd pr
om
e
th
e
us
M
e
t
hodol
ogi
e
s
P
ha
s
e
s
C
om
pa
r
i
s
on
G
a
i
a
* St
at
e
m
e
nt
of
r
e
qui
r
e
m
e
nt
* A
nal
y
s
i
s
*
D
e
s
i
gn
* L
a
c
k de
t
a
i
l
e
d
s
t
e
p
-
by
-
s
t
e
p br
e
a
kdow
n.
* N
o de
t
a
i
l
s
on how
r
e
qui
r
e
m
e
nt
s
t
a
t
e
m
e
nt
s
m
a
y be
a
c
qui
r
e
d.
* V
i
e
w
a
ge
nt
s
y
s
t
e
m
a
s
a
n or
ga
ni
s
a
t
i
ona
l
m
ode
l
.
* R
ol
e
s
a
r
e
s
i
m
i
l
a
r
t
o f
unc
t
i
ona
l
i
t
i
e
s
i
n P
r
om
e
t
he
us
.
* E
di
t
or
t
ool
G
a
i
a
4E
s
uppor
t
s
de
s
i
gn.
T
r
opos
*E
ar
l
y
r
e
qui
r
e
m
e
nt
phas
e
* L
at
e
r
r
e
qui
r
e
m
e
nt
phas
e
* A
r
c
hi
t
e
c
t
ur
al
de
s
i
gn
*D
e
t
ai
l
e
d de
s
i
gn
* I
m
pl
e
m
e
nt
at
i
on
* E
m
pha
s
i
s
e
s
t
he
E
ar
l
y
R
e
qui
r
e
m
e
nt
A
nal
y
s
i
s
, t
he
n t
he
L
at
e
r
R
e
qui
r
e
m
e
nt
P
ha
s
e
.
* S
pe
c
i
a
l
i
s
a
t
i
on of
G
oa
l
s
i
nt
o s
ubc
l
a
s
s
e
s
of
H
ar
dgoal
, a
nd
S
of
t
goal
s
f
or
a
c
t
or
s
of
s
ys
t
e
m
.
* N
o ge
ne
r
a
l
a
r
c
hi
t
e
c
t
ur
e
c
ont
a
i
ni
ng a
l
l
t
he
pha
s
e
s
of
de
s
i
gn a
s
i
n G
a
i
a
, M
a
S
E
, or
P
r
om
e
t
he
us
.
* H
a
s
a
de
s
i
gn s
uppor
t
t
ool
c
a
l
l
e
d T
a
om
4E
.
P
r
om
e
t
he
us
* Sy
s
t
e
m
s
p
e
c
i
f
i
c
at
i
on
* A
r
c
hi
t
e
c
t
ur
al
de
s
i
gn
* D
e
t
ai
l
e
d de
s
i
gn
phas
e
* N
o E
a
r
l
y R
e
qui
r
e
m
e
nt
pha
s
e
a
s
i
n T
r
opos
. B
ut
t
hi
s
c
a
n be
a
da
pt
e
d.
* U
s
e
s
I
ni
t
i
al
goal
s
, t
ha
t
a
r
e
r
e
f
i
ne
d or
br
oke
n dow
n i
nt
o
Subgoal
s
f
or
a
ge
nt
s
.
* V
e
r
y de
t
a
i
l
e
d de
s
i
gn a
c
t
i
vi
t
y f
r
om
S
ys
t
e
m
S
pe
c
i
f
i
c
a
t
i
on pha
s
e
t
o ot
he
r
pha
s
e
s
.
* R
e
l
i
a
nc
e
on e
xpe
r
t
know
l
e
dge
on dom
a
i
n
s
ubj
e
c
t
f
or
r
e
qui
r
e
m
e
nt
a
c
qui
s
i
t
i
on.
* H
a
s
a
c
u
s
t
om
i
s
e
d P
D
T
, a
A
U
M
L
t
ool
t
ha
t
s
uppor
t
s
de
s
i
gn pr
oc
e
s
s
.
2.2.
P
r
om
e
t
h
e
u
s
P
r
om
e
th
e
us
[
27]
is
a
m
e
th
odol
ogy
de
s
ig
ne
d
f
or
th
e
r
e
a
li
s
a
ti
on
of
B
D
I
a
ge
nt
s
ys
te
m
s
w
it
h
th
e
us
e
of
goa
ls
a
nd pla
ns
. I
t
s
uppor
ts
de
ve
lo
pm
e
nt
a
c
ti
vi
ti
e
s
f
r
om
r
e
qui
r
e
m
e
nt
s
s
pe
c
if
ic
a
ti
on t
hr
ough to deta
il
e
d de
s
ig
n
f
or
i
m
pl
e
m
e
nt
a
ti
on
.
P
r
om
e
th
e
us
de
s
ig
n t
ool
(
P
D
T
)
[
29, 30
]
is
a
gr
a
phi
c
a
l
e
di
to
r
t
ha
t
s
uppor
ts
t
he
P
r
om
e
th
e
us
m
e
th
odol
ogy.
T
he
P
D
T
s
uppor
ts
th
e
de
ve
lo
pm
e
nt
a
nd
doc
um
e
nt
a
ti
on
of
a
ll
th
e
pha
s
e
s
of
th
e
P
r
om
e
th
e
us
m
e
th
odol
ogy
f
or
bui
ld
in
g
a
ge
nt
-
ba
s
e
d
s
ys
te
m
s
.
P
r
om
e
th
e
us
ha
s
th
r
e
e
in
te
r
-
c
onne
c
te
d
de
s
ig
n
pha
s
e
s
w
hi
c
h
a
r
e
s
y
s
te
m
s
pe
c
if
ic
at
io
n
,
a
r
c
hi
te
c
tu
r
al
d
e
s
ig
n,
a
nd t
he
de
ta
il
e
d
de
s
ig
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
D
e
s
ig
n and analys
i
s
of
a m
ul
ti
-
age
nt
e
-
le
ar
ni
ng s
y
s
te
m
us
in
g
p
r
om
e
th
e
us
…
(
K
e
nne
d
y
E
. E
hi
m
w
e
nm
a
)
13
3.
M
U
L
T
I
-
A
G
E
N
T
B
A
S
E
D
P
R
E
-
A
S
S
E
S
S
M
E
N
T
S
Y
S
T
E
M
D
E
S
I
G
N
M
E
T
H
O
D
O
L
O
G
Y
I
n
th
is
s
e
c
ti
on,
w
e
pr
e
s
e
nt
th
e
pr
om
e
th
e
us
de
s
ig
n
m
e
th
odol
ogy
a
nd
de
ta
il
e
d
a
na
ly
s
is
of
th
e
pr
e
-
a
s
s
e
s
s
m
e
nt
s
ys
te
m
.
I
n
T
a
bl
e
2
a
r
e
th
e
P
D
T
s
ym
bol
s
a
nd
de
s
c
r
ip
ti
on
of
th
e
ir
f
unc
ti
ons
in
th
e
de
s
ig
n
of
a
ge
nt
-
ba
s
e
d
s
ys
te
m
s
.
T
he
pr
e
-
a
s
s
e
s
s
m
e
nt
s
ys
te
m
is
a
f
or
m
a
ti
ve
a
s
s
e
s
s
m
e
nt
s
y
s
te
m
de
s
ig
ne
d
to
s
uppor
ts
th
e
le
a
r
ni
ng
of
S
Q
L
.
L
e
a
r
ni
ng
a
s
a
s
s
e
r
te
d
in
[
31]
is
a
n
e
f
f
or
t
in
te
n
s
iv
e
ta
s
k.
T
hu
s
,
de
s
ig
ni
ng
a
M
A
S
f
or
le
a
r
ni
ng
pur
pos
e
s
i
s
a
c
om
pl
e
x pr
oc
e
s
s
[
28]
.
3.1. S
ys
t
e
m
s
p
e
c
i
f
ic
at
io
n
T
he
s
ys
te
m
s
pe
c
if
ic
a
ti
on
pha
s
e
be
gi
n
s
w
it
h
a
hi
gh
-
le
ve
l
de
s
c
r
ip
ti
on
of
th
e
pr
obl
e
m
,
w
hi
c
h
le
a
d
s
to
th
e
i
de
nt
if
ic
a
ti
on of
in
it
ia
l
goal
s
de
c
om
pos
it
io
n a
s
s
how
n i
n F
i
gur
e
2.
F
ig
ur
e
2. H
ig
h l
e
ve
l
de
s
c
r
ip
ti
on of
pr
obl
e
m
i
nc
lu
di
ng i
ni
ti
a
l
go
a
l
a
nd ove
r
a
ll
s
ys
te
m
goa
l
s
pe
c
if
ic
a
ti
on
T
a
bl
e
2.
T
he
P
D
T
not
a
ti
on
s
ym
bol
a
nd me
a
ni
ng
[
3, 32, 33]
N
a
m
e
S
ym
bol
D
e
s
c
r
i
pt
i
on
A
ge
nt
T
he
a
ge
nt
s
ym
bol
s
.
A
c
t
i
on
T
hi
s
i
s
w
ha
t
t
he
a
ge
nt
doe
s
t
h
a
t
ha
s
e
f
f
e
c
t
on t
he
e
nvi
r
onm
e
nt
or
ot
he
r
a
ge
nt
s
.
R
ol
e
T
hi
s
s
ym
bol
i
z
e
s
r
ol
e
s
or
gr
oup of
r
ol
e
s
f
or
a
ge
nt
s
.
P
r
ot
oc
ol
P
r
ot
oc
ol
s
s
pe
c
i
f
i
e
s
i
nt
e
r
a
c
t
i
on be
t
w
e
e
n a
g
e
nt
s
. P
r
ot
oc
ol
s
a
r
e
s
p
e
c
i
f
i
e
d us
i
ng t
e
x
t
ua
l
not
a
t
i
ons
t
ha
t
m
a
ps
t
o A
U
M
L
2.
D
a
t
a
T
hi
s
i
s
us
e
d t
o r
e
pr
e
s
e
nt
t
he
be
l
i
e
f
(
i
nt
e
r
na
l
know
l
e
dge
m
ode
l
)
or
e
xt
e
r
na
l
da
t
a
. I
t
i
s
w
he
r
e
f
unc
t
i
ona
l
i
t
i
e
s
t
ha
t
t
r
a
ns
c
e
nds
t
o a
ge
nt
r
e
a
d or
w
r
i
t
e
da
t
a
or
i
nf
or
m
a
t
i
on.
M
e
s
s
a
ge
s
T
hi
s
i
s
us
e
d t
o s
ym
bol
i
z
e
a
m
e
s
s
a
ge
c
om
m
uni
c
a
t
i
on be
t
w
e
e
n a
ge
nt
s
.
B
D
I
M
e
s
s
a
ge
s
T
hi
s
s
ym
bol
i
s
us
e
d t
o
r
e
pr
e
s
e
nt
m
e
s
s
a
g
e
s
t
ha
t
upda
t
e
s
t
he
be
l
i
e
f
s
of
a
ge
nt
s
.
P
e
r
c
e
pt
R
e
pr
e
s
e
nt
s
t
h
e
i
nput
c
om
i
ng f
r
om
t
he
e
nvi
r
onm
e
nt
t
o t
he
a
ge
nt
.
S
c
e
na
r
i
o
T
hi
s
i
s
a
n a
bs
t
r
a
c
t
de
s
c
r
i
pt
i
on of
a
s
e
que
nc
e
of
s
t
e
p
s
t
a
ke
n i
n t
he
de
v
e
l
opm
e
nt
of
a
s
ys
t
e
m
.
I
t
i
s
us
ua
l
l
y t
he
i
ni
t
i
a
l
s
t
e
p t
ha
t
s
t
a
r
t
s
f
or
t
he
br
e
a
kdow
n of
t
he
“
s
t
a
t
e
m
e
nt
of
pr
obl
e
m
”
or
de
s
c
r
i
pt
i
on of
t
he
pr
obl
e
m
t
o s
ol
ve
.
G
oa
l
I
t
i
s
t
he
r
e
a
l
i
z
a
bl
e
t
a
r
ge
t
or
a
c
hi
e
ve
m
e
nt
s
e
t
f
or
a
n a
ge
nt
.
C
onne
c
t
i
on A
r
r
ow
s
T
he
y a
r
e
e
dge
s
t
ha
t
c
onne
c
t
s
e
nt
i
t
i
e
s
(
i
.e
.
s
ym
bol
s
)
t
oge
t
he
r
.
3.1.1. S
c
e
n
ar
io
ove
r
vi
e
w
S
c
e
na
r
io
s
a
nd
s
ys
te
m
goa
ls
a
r
e
c
om
pl
e
m
e
nt
a
r
y.
I
n
th
e
pr
oc
e
s
s
of
e
xt
r
a
c
ti
ng
th
e
m
ai
n
goal
s
f
r
om
th
e
pr
obl
e
m
de
s
c
r
ip
ti
on,
s
c
e
na
r
io
s
w
e
r
e
de
ve
lo
pe
d
a
s
s
how
n
in
F
ig
ur
e
3.
I
n
F
ig
ur
e
3
a
r
e
th
e
s
e
t
of
s
c
e
na
r
io
s
de
r
iv
e
d f
r
om
t
he
s
pe
c
if
ie
d goa
ls
us
in
g t
he
P
D
T
s
c
e
na
r
io
ov
e
r
v
i
e
w
di
a
gr
a
m
.
3.1.2. S
ys
t
e
m
goal
d
ia
gr
am
T
he
P
D
T
s
ys
te
m
goal
ov
e
r
v
ie
w
di
a
gr
a
m
e
n
a
bl
e
s
th
e
br
e
a
k
-
dow
n
or
r
e
f
in
e
m
e
nt
of
th
e
s
e
t
of
de
r
iv
e
d
s
c
e
na
r
io
s
in
to
uni
ts
of
a
c
hi
e
va
bl
e
de
s
ig
n
s
te
p
s
.
T
he
F
ig
ur
e
4
is
th
e
s
ys
te
m
goal
a
nd
s
ubgoals
d
e
s
ig
n
a
nd
th
e
in
te
r
a
c
ti
ons
be
twe
e
n
th
e
m
.
T
he
A
N
D
is
a
c
onj
unc
ti
on
f
unc
ti
o
n
w
hi
c
h
in
di
c
a
te
s
th
a
t,
a
t
th
a
t
le
ve
l
o
f
de
s
ig
n,
th
e
a
ge
nt
m
us
t
c
om
m
uni
c
a
te
w
it
h
bot
h
c
la
s
s
if
y
a
nd
th
e
pe
r
s
is
te
nt
B
B
update
goa
ls
a
f
te
r
it
s
de
c
is
io
n
-
m
ak
in
g
f
unc
ti
on.
I
n
F
ig
ur
e
4,
a
t
th
e
us
e
r
in
te
r
f
a
c
e
,
pe
r
c
e
pt
goal
is
s
e
e
n
in
te
r
a
c
ti
ng
w
it
h
th
e
unde
r
s
ta
ndi
ng
of
pr
e
r
e
qui
s
it
e
goal
w
hi
c
h
c
on
ne
c
t
s
to
th
e
te
s
ti
ng
goal
.
T
h
e
n
to
th
e
m
ak
e
de
c
i
s
io
n
goal
th
a
t
is
li
nki
ng
bot
h
th
e
I
n
i
t
i
a
l
g
o
a
l
s
:
§
O
b
s
e
r
v
e
p
e
r
c
e
p
t
§
U
n
d
e
r
s
t
a
n
d
i
n
g
o
f
p
r
e
r
e
q
u
i
s
i
t
e
§
T
e
s
t
i
n
g
§
C
l
a
s
s
i
f
y
i
n
g
§
C
o
n
t
i
n
u
o
u
s
f
e
e
d
b
a
c
k
§
K
B
u
p
d
a
t
e
§
R
e
c
o
m
m
e
n
d
m
a
t
e
r
i
a
l
s
G
o
a
l
S
p
e
c
i
f
i
c
a
t
i
o
n
:
§
H
O
W
C
A
N
E
A
C
H
O
F
T
H
E
S
E
I
N
I
T
I
A
L
G
O
A
L
S
B
E
A
C
H
I
E
V
E
D
?
§
E
A
C
H
G
O
A
L
S
H
A
D
F
U
R
T
H
E
R
S
U
B
-
G
O
A
L
S
D
E
V
E
L
O
P
E
D
A
S
:
*
O
b
s
e
r
v
e
p
e
r
c
e
p
t
-
R
e
c
e
i
v
e
u
s
e
r
c
o
n
c
e
p
t
-
p
r
e
s
e
n
t
c
o
n
c
e
p
t
D
E
S
I
R
E
D
_
C
O
N
C
E
P
T
*
U
n
d
e
r
s
t
a
n
d
i
n
g
o
f
p
r
e
r
e
q
u
i
s
i
t
e
-
q
u
i
z
z
e
s
i
n
B
B
-
a
n
s
w
e
r
s
i
n
B
B
-
p
r
e
r
e
q
u
i
s
i
t
e
a
s
s
e
s
s
m
e
n
t
f
r
o
m
q
u
i
z
z
e
s
&
a
n
s
w
e
r
s
U
N
D
E
R
S
T
A
N
D
I
N
G
P
R
E
R
E
Q
U
I
S
I
T
E
*
T
e
s
t
i
n
g
-
s
e
a
r
c
h
B
B
f
o
r
q
u
i
z
z
e
s
-
f
e
t
c
h
(
s
u
b
-
c
o
n
c
e
p
t
s
o
r
)
p
r
e
r
e
q
u
i
s
i
t
e
q
u
i
z
z
e
s
-
r
e
c
e
i
v
e
a
n
s
w
e
r
-
f
e
t
c
h
B
B
a
n
s
w
e
r
a
n
d
c
o
m
p
a
r
e
w
i
t
h
s
t
u
d
e
n
t
s
’
-
m
a
k
e
a
s
s
e
s
s
m
e
n
t
d
e
c
i
s
i
o
n
T
E
S
T
I
N
G
U
S
E
R
*
C
l
a
s
s
i
f
y
i
n
g
-
a
g
g
r
e
g
a
t
e
l
e
a
r
n
i
n
g
a
c
t
i
v
i
t
y
-
u
s
e
p
r
e
d
i
c
a
t
e
s
t
a
t
e
m
e
n
t
r
u
l
e
s
-
c
l
a
s
s
i
f
y
s
t
u
d
e
n
t
s
b
a
s
e
d
o
n
r
u
l
e
s
m
a
t
c
h
C
L
A
S
S
I
F
I
C
A
T
I
O
N
*
R
e
c
o
m
m
e
n
d
m
a
t
e
r
i
a
l
s
-
c
o
n
c
e
p
t
o
n
t
o
l
o
g
y
i
n
B
B
-
s
e
a
r
c
h
o
n
t
o
l
o
g
i
c
a
l
r
e
l
a
t
i
o
n
-
f
e
t
c
h
U
R
L
l
i
n
k
-
p
r
e
s
e
n
t
t
o
u
s
e
r
R
E
C
O
M
M
E
N
D
A
T
I
O
N
*
K
B
u
p
d
a
t
i
n
g
-
s
t
o
r
e
u
s
e
r
l
e
a
r
n
i
n
g
a
c
t
i
v
i
t
y
p
e
r
s
i
s
t
e
n
t
l
y
P
E
R
S
I
S
T
E
N
T
B
E
L
I
E
F
S
T
O
R
E
*
C
o
n
t
i
n
u
o
u
s
u
s
e
r
f
e
e
d
b
a
c
k
-
u
s
e
r
f
r
i
e
n
d
l
y
i
n
t
e
r
a
c
t
i
o
n
f
r
o
m
a
s
s
e
s
s
m
e
n
t
s
-
w
e
l
c
o
m
e
a
n
d
i
n
t
r
o
d
u
c
t
i
o
n
t
o
s
y
s
t
e
m
U
S
E
R
I
N
T
E
R
A
C
T
I
O
N
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
10
, N
o.
1
,
M
a
r
c
h
20
2
1
:
9
–
23
14
c
la
s
s
if
y
a
nd
pe
r
s
is
te
nt
B
B
update
goal
s
a
f
te
r
it
s
de
c
is
io
n
-
m
a
ki
ng
f
unc
ti
on;
a
nd
th
e
c
la
s
s
if
y
goal
f
ur
th
e
r
c
onne
c
ts
t
he
r
e
c
om
m
e
nd m
at
e
r
ia
l
goal
.
F
ig
ur
e
3. S
ys
te
m
s
c
e
na
r
io
vi
e
w
F
ig
ur
e
4. S
ys
te
m
goa
ls
s
pe
c
if
ic
a
ti
on a
nd de
c
om
pos
it
io
n f
or
th
e
pr
e
-
a
s
s
e
s
s
m
e
nt
s
y
s
te
m
3.2
.
A
r
c
h
it
e
c
t
u
r
al
d
e
s
ig
n
A
t
th
is
pha
s
e
,
th
e
r
ol
e
of
th
e
s
y
s
te
m
f
or
th
e
pur
pos
e
of
pr
e
-
a
s
s
e
s
s
m
e
nt
ha
s
be
e
n
c
onc
e
pt
ua
li
z
e
d. T
he
ne
e
de
d
num
be
r
of
a
ge
nt
s
a
nd
th
e
ir
de
s
c
r
ip
ti
ve
na
m
e
s
h
a
ve
be
e
n
de
te
r
m
in
e
d
a
nd
in
c
lu
de
d
in
th
e
de
s
ig
n.
T
hi
s
pha
s
e
c
ove
r
s
th
e
s
y
s
te
m
ove
r
a
ll
(
s
ta
ti
c
)
s
tr
uc
tu
r
e
us
in
g
s
y
s
te
m
ov
e
r
v
ie
w
di
a
gr
a
m
,
a
nd
th
e
de
s
c
r
ip
ti
on
of
th
e
dy
nam
ic
be
ha
vi
our
of
th
e
s
ys
te
m
u
s
in
g
in
te
r
a
c
ti
on
di
a
gr
a
m
a
nd
in
te
r
a
c
ti
on
pr
ot
oc
ol
s
.
I
n
F
ig
ur
e
5
a
r
e
th
e
id
e
nt
if
ie
d
r
ol
e
s
th
a
t
a
r
e
ne
e
de
d
w
it
hi
n
th
e
r
ol
e
de
c
om
po
s
it
io
n
r
e
s
ul
ts
in
to
a
r
ol
e
hi
e
r
a
r
c
h
y
f
r
om
s
upe
r
-
r
ol
e
s
to
a
to
m
ic
r
ol
e
s
(
to
p
-
dow
n di
r
e
c
ti
on)
[
28]
.
I
n
th
is
s
te
p,
a
ll
th
e
a
ge
nt
s
,
th
e
ir
pe
r
c
e
pt
s
,
in
c
om
in
g
m
e
s
s
a
ge
s
,
a
c
ti
ons
dul
y
ta
ke
n
a
nd
in
te
r
a
c
ti
on
i
n
th
e
de
s
ig
n
a
r
e
pr
e
s
e
nt
e
d
in
F
ig
ur
e
6.
I
n
th
e
s
y
s
te
m
ov
e
r
v
ie
w
d
ia
gr
am
,
th
e
d
a
ta
(
a
ge
nt
kno
w
le
dge
)
th
a
t
i
s
e
xpe
c
te
d t
o us
e
d i
s
c
oupl
e
d w
it
h t
he
a
ge
nt
s
. I
n t
hi
s
de
s
ig
n, t
he
da
ta
a
r
e
qui
z
z
e
s
, a
ns
w
e
r
s
t
o quiz
z
e
s
, a
nd U
R
L
da
ta
li
nks
f
or
e
a
c
h
of
th
e
le
a
f
node
s
(
s
ub
-
to
pi
c
s
)
in
th
e
ont
ol
og
y.
T
he
d
a
ta
is
m
ode
ll
e
d
a
s
in
te
r
na
l
kn
ow
le
dg
e
or
be
li
e
f
s
in
th
e
a
ge
nt
s
.
F
ig
ur
e
6
a
ls
o
pr
e
s
e
nt
s
th
e
f
iv
e
w
or
ki
ng
a
ge
nt
s
of
th
e
s
y
s
te
m
w
ho
s
e
d
e
ta
il
de
s
ig
n
a
r
e
il
lu
s
tr
a
te
d i
n
age
nt
ov
e
r
v
ie
w
s
ta
ge
.
F
ig
ur
e
5. S
ys
te
m
r
ol
e
ove
r
vi
e
w
s
how
in
g s
tr
uc
tu
r
e
d
f
unc
ti
ona
li
ti
e
s
F
ig
ur
e
6. S
ys
te
m
ove
r
vi
e
w
di
a
gr
a
m
3.3. De
t
ai
le
d
d
e
s
ig
n
T
hi
s
pha
s
e
is
f
oc
us
e
d
on
th
e
de
s
c
r
ip
ti
on
of
r
e
s
pons
ib
il
it
ie
s
a
nd
c
a
pa
bi
li
ti
e
s
of
th
e
in
te
r
na
l
s
tr
uc
tu
r
e
of
th
e
in
di
vi
dua
l
a
ge
nt
,
a
nd
how
e
a
c
h
a
ge
nt
w
oul
d
a
c
hi
e
ve
th
e
ir
ta
s
k
w
it
hi
n
th
e
s
ys
te
m
.
D
ia
gr
a
m
m
a
ti
c
a
ll
y,
th
e
s
e
c
a
p
a
bi
li
ti
e
s
w
e
r
e
r
e
a
li
s
e
d a
t
th
e
age
nt
ov
e
r
v
ie
w
s
ta
ge
s
ho
w
n be
lo
w
.
3.3.1
.
A
ge
n
t
ove
r
vi
e
w
I
n
th
is
s
e
c
ti
on,
th
e
li
s
t
of
a
ge
nt
s
in
th
e
s
ys
te
m
a
nd
in
di
vi
dua
l
a
ge
nt
in
te
r
na
l
de
ta
il
s
a
r
e
pr
e
s
e
nt
e
d.
T
hi
s
in
c
lu
de
s
th
e
a
g
e
nt
s
’
pl
a
n
m
ode
l
s
,
not
a
ti
on
f
or
pe
r
c
e
pt
s
a
nd
tr
ig
ge
r
in
g_e
ve
nt
,
c
om
m
uni
c
a
ti
on
li
nks
a
nd
in
te
r
-
a
ge
nt
m
e
s
s
a
ge
d
e
s
c
r
ip
ti
on.
At
th
e
a
g
e
nt
ove
r
vi
e
w
s
ta
g
e
,
in
he
r
it
e
d
in
te
r
f
a
c
e
s
f
r
om
e
.g.,
th
e
s
y
s
te
m
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
D
e
s
ig
n and analys
i
s
of
a m
ul
ti
-
age
nt
e
-
le
ar
ni
ng s
y
s
te
m
us
in
g
p
r
om
e
th
e
us
…
(
K
e
nne
d
y
E
. E
hi
m
w
e
nm
a
)
15
ov
e
r
v
ie
w
pha
s
e
a
r
e
a
dopt
e
d
f
or
s
pe
c
if
yi
ng
a
ge
nt
s
’
de
ta
il
s
.
T
he
in
he
r
it
e
d
in
te
r
f
a
c
e
s
a
r
e
th
e
not
a
ti
on
s
ym
bol
s
th
a
t
a
ppe
a
r
gr
e
yi
s
h i
n c
ol
our
.
A
ge
n
t
agI
n
te
r
f
ac
e
:
I
n
F
ig
ur
e
7
is
a
m
uc
h
r
e
f
in
e
d
a
nd
de
t
a
il
e
d de
s
ig
n
w
he
r
e
C
A
r
tAgO
a
r
ti
f
a
c
t
is
th
e
m
e
di
um
th
a
t
w
a
s
us
e
d
to
ge
t
us
e
r
in
put
.
T
he
in
te
r
f
a
c
e
a
ge
nt
f
ir
s
t
c
r
e
a
te
s
th
e
a
r
ti
f
a
c
t
in
or
de
r
to
obs
e
r
ve
it
.
T
he
obs
e
r
ve
d
in
put
s
a
r
e
c
om
m
uni
c
a
te
d
a
s
m
e
s
s
a
ge
s
in
a
ge
nt
pl
a
n
(
s
how
n
w
it
h
th
e
pl
a
n
di
a
gr
a
m
or
s
ym
bol
)
to
ot
he
r
a
ge
nt
s
e
.g., t
he
a
ge
nt
agSuppo
r
t
th
a
t
is
r
e
s
pon
s
ib
le
f
or
pr
e
-
a
s
s
e
s
s
in
g s
tu
de
nt
s
.
A
ge
n
t
agS
u
ppor
t:
T
hi
s
is
th
e
pr
e
-
te
s
t
a
ge
nt
th
a
t
is
s
a
ddl
e
d
w
it
h
th
e
ta
s
k
of
que
s
ti
oni
ng
a
u
s
e
r
’
s
s
ki
ll
s
be
f
or
e
m
a
ki
ng
r
e
c
om
m
e
nda
ti
on,
a
s
s
how
n
in
F
ig
ur
e
s
8
-
9.
T
he
a
ge
nt
agSuppor
t
us
e
s
it
s
ac
hi
e
v
e
m
e
nt
goal
s
f
or
na
vi
ga
ti
on,
f
r
om
le
a
f
node
,
to
le
a
f
node
,
+
1
in
th
e
hi
e
r
a
r
c
hy
of
c
onc
e
pt
s
to
r
e
tr
ie
ve
qui
z
z
e
s
w
hi
c
h
a
r
e
r
e
pr
e
s
e
nt
e
d i
n pr
e
di
c
a
te
l
ogi
c
i
n i
ts
B
B
t
o t
e
s
t
s
tu
de
nt
s
’
s
ki
ll
s
.
U
s
in
g t
he
a
ns
w
e
r
pe
r
c
e
pt
r
e
c
e
iv
e
d, i
t
c
om
pa
r
e
s
a
nd
m
a
tc
he
s
th
e
gi
ve
n
a
ns
w
e
r
in
put
w
it
h
th
e
pr
e
de
f
in
e
d
a
n
s
w
e
r
in
it
s
B
B
.
T
a
ki
ng
th
e
de
c
i
s
io
n
f
or
e
it
he
r
a
pas
s
e
d
or
a
fa
il
e
d
pr
e
di
c
a
te
on
e
ve
r
y
a
ns
w
e
r
r
e
c
e
iv
e
d,
th
is
a
ge
nt
a
ls
o
c
om
m
uni
c
a
t
e
a
ll
a
s
s
e
s
s
m
e
nt
a
c
ti
vi
ti
e
s
,
na
m
e
ly
:
th
e
de
c
is
io
n
r
e
a
c
he
d
pe
r
que
s
ti
on,
th
e
que
s
ti
ons
a
s
ke
d
,
a
nd
c
om
m
uni
c
a
ti
on
of
th
e
a
ns
w
e
r
s
r
e
c
e
iv
e
d
to
ot
he
r
a
ge
nt
s
i
n t
he
M
A
S
t
ha
t
ne
e
d
s
t
o know. T
hi
s
a
ge
nt
a
l
s
o
dat
e
a
nd
ti
m
e
s
ta
m
p
e
ve
r
y l
e
a
r
ni
ng a
c
ti
vi
ty
.
F
ig
ur
e
7. D
e
ta
il
e
d ove
r
vi
e
w
of
a
ge
nt
a
gI
nt
e
r
f
a
c
e
F
ig
ur
e
8. A
ge
nt
a
gS
uppor
t
r
e
c
e
iv
in
g t
he
de
s
ir
e
d_C
onc
e
pt
pe
r
c
e
pt
a
nd r
e
tr
ie
vi
ng
qui
z
z
e
s
us
in
g pl
a
ns
A
ge
n
t
agM
ode
ll
in
g:
T
hi
s
a
g
e
nt
ge
ts
m
e
s
s
a
ge
pe
r
c
e
pt
s
f
r
om
a
ge
nt
agSuppor
t
f
or
e
ve
r
y
le
a
f
node
(
que
s
ti
on
a
tt
a
c
he
d
to
a
uni
t
of
le
a
r
ni
ng)
in
th
e
ont
ol
ogy
w
hos
e
pr
e
-
a
s
s
e
s
s
m
e
nt
ha
s
be
e
n
c
om
pl
e
te
d. T
hi
s
a
g
e
nt
us
e
s
th
e
pe
r
c
e
pt
(
or
in
f
or
m
a
ti
on)
it
r
e
c
e
iv
e
s
to
m
a
tc
h
th
e
pr
e
-
c
ondi
ti
ons
in
it
s
pl
a
n
c
ont
e
x
t
,
a
nd
th
e
r
e
a
f
te
r
c
la
s
s
if
y
th
e
s
tu
de
nt
’
s
s
ki
ll
s
.
T
he
c
a
te
gor
y
of
in
f
or
m
a
ti
on
(
in
one
pl
a
n)
th
a
t
is
de
te
r
m
in
e
d
by
th
is
a
ge
nt
is
c
om
m
uni
c
a
te
d
to
th
e
ne
xt
r
e
c
e
iv
in
g
a
ge
nt
(
agM
a
te
r
ia
l
)
th
a
t
w
il
l
in
tu
r
n
s
e
nd
le
a
r
ni
ng
m
a
te
r
ia
l
to
th
e
s
tu
de
nt
,
a
s
s
how
n
in
F
ig
ur
e
10.
F
ig
ur
e
9. A
ge
nt
a
gS
uppor
t
ove
r
vi
e
w
F
ig
ur
e
10. T
he
a
ge
nt
a
gM
od
e
ll
in
g (
c
la
s
s
if
ie
r
a
ge
nt
)
A
ge
n
t
agM
at
e
r
ia
l:
F
ig
ur
e
11
is
a
ge
nt
agM
at
e
r
ia
l
th
a
t
ke
e
p
s
th
e
U
R
L
s
li
nks
of
le
a
r
ni
ng
m
a
te
r
ia
l
a
s
a
n
ont
ol
ogy.
T
he
pe
r
f
om
a
ti
ve
us
e
d
in
th
e
m
e
s
s
a
ge
to
th
is
a
g
e
nt
is
“
ac
hi
e
v
e
”
.
O
n
r
e
c
e
iv
in
g
th
e
“
ac
hi
e
v
e
”
pe
r
f
or
m
a
ti
ve
m
e
s
s
a
ge
f
r
om
th
e
c
la
s
s
if
ie
r
a
ge
nt
(
a
f
te
r
c
la
s
s
if
ic
a
ti
on)
,
th
e
a
ge
nt
agM
at
e
r
ia
l
th
e
n
r
e
le
a
s
e
s
le
a
r
ni
ng
m
a
te
r
ia
ls
f
or
s
tu
de
nt
s
to
le
a
r
n.
T
he
s
e
m
a
te
r
ia
l
s
a
r
e
d
e
pe
nde
nt
on
th
e
num
be
r
of
fa
il
e
d
a
nd
pas
s
e
d
pr
e
r
e
qui
s
it
e
a
s
s
e
s
s
m
e
nt
s
.
A
ge
n
t
agM
ode
l:
T
hi
s
a
ge
nt
us
e
s
th
e
J
a
va
T
e
x
tP
e
r
s
is
te
nt
B
B
c
la
s
s
to
s
to
r
e
a
ll
th
e
le
a
r
ni
ng
a
c
ti
vi
ti
e
s
in
th
e
s
ys
te
m
.
T
he
T
e
x
tP
e
r
s
is
te
nt
B
B
c
la
s
s
w
a
s
c
onf
ig
ur
e
d
in
th
e
M
A
S
a
t
th
e
poi
nt
of
de
c
la
r
a
ti
on
or
c
r
e
a
ti
on
of
th
e
m
ul
ti
-
a
ge
nt
s
pr
oj
e
c
t
w
it
h
th
e
M
as
2j
[
34]
e
xt
e
ns
io
n
a
t
th
e
le
ve
l
of
i
m
pl
e
m
e
nt
a
ti
on.
T
he
a
c
ti
vi
ti
e
s
s
to
r
e
d
a
r
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
10
, N
o.
1
,
M
a
r
c
h
20
2
1
:
9
–
23
16
m
e
s
s
a
ge
s
s
e
nt
t
o t
he
a
ge
nt
;
a
nd t
he
y i
nc
lu
de
s
tu
de
nt
s
’
de
s
ir
e
d t
opi
c
s
, a
nd a
ns
w
e
r
s
t
o que
s
ti
on (
bot
h c
or
r
e
c
t
or
in
c
or
r
e
c
t)
pe
r
c
e
pt
. A
s
s
how
n i
n F
ig
ur
e
12, the
pe
r
s
is
te
nt
be
li
e
f
s
a
r
e
pe
r
m
a
ne
nt
ly
s
to
r
e
d i
n t
he
s
ys
t
e
m
.
F
ig
ur
e
11. Age
nt
a
gM
a
te
r
ia
l:
T
he
l
e
a
r
ni
ng
m
a
te
r
ia
l
a
ge
nt
ove
r
vi
e
w
F
ig
ur
e
12. Age
nt
a
gM
ode
l
(
s
tu
de
nt
)
ove
r
vi
e
w
3.4
.
L
ogi
c
al
an
al
ys
is
of
p
r
e
-
as
s
e
s
s
m
e
n
t
P
r
e
vi
ous
r
e
por
ts
on
th
e
pr
e
-
a
s
s
e
s
s
m
e
nt
s
ys
te
m
pr
e
s
e
nt
e
d
a
pr
e
-
a
s
s
e
s
s
m
e
nt
m
e
c
ha
ni
s
m
[
35,
32]
a
nd
th
e
f
or
m
a
li
z
e
d
lo
gi
c
m
ode
l
[
36
]
f
o
r
pr
e
-
s
ki
ll
s
te
s
ti
ng,
de
c
is
io
n
m
a
ki
ng,
s
e
le
c
ti
on
a
nd
r
e
c
om
m
e
nda
ti
on
of
le
a
r
ni
ng
m
a
te
r
ia
ls
.
I
n
ge
ne
r
a
l,
on
a
ny
gi
v
e
n
knowle
dge
gr
a
ph
or
ont
ol
ogy,
th
e
f
ol
lo
w
in
g
s
ym
bol
ic
a
lg
or
it
hm
pr
e
s
e
nt
s
t
he
unde
r
ly
in
g r
e
a
s
oni
ng
[
32]
:
ⅅ
,
has
P
r
e
r
e
qui
s
it
e
(
ⅅ
,
,
+
1
)
ꓥ
has
K
B
(
+
1
,
±
1
,
±
1
,
)
[
:
Ǝ
ⅅ
ꓥ
pas
s
e
d(
±
1
,
)
=> l
e
ar
n(
,
)
ꓥ
has
K
B
(
ⅅ
,
,
,
)
e
ls
e
:
Ǝ
ⅅ
ꓥ
Ǝ
fa
il
e
d(
±
1
,
)
=> l
e
ar
n(
±
1
,
)
ꓥ
has
K
B
(
+
1
,
±
1
,
±
1
,
)
]
of
th
e
m
ul
ti
-
a
ge
nt
pr
e
-
a
s
s
e
s
s
m
e
nt
s
y
s
te
m
w
hos
e
s
of
twa
r
e
e
ngi
ne
e
r
in
g
de
s
ig
n
s
te
ps
w
e
ha
ve
pr
e
s
e
nt
e
d
in
th
e
pr
e
c
e
di
ng
s
e
c
ti
on
s
;
s
uc
h
th
a
t,
ⅅ
is
th
e
de
s
i
r
e
d
c
on
c
e
pt
(
a
ls
o
c
a
ll
e
d
th
e
de
s
ir
e
d
to
pi
c
)
th
a
t
s
ub
s
um
e
s
s
om
e
pr
e
r
e
qui
s
it
e
s
w
hi
c
h
f
ur
th
e
r
s
ub
s
um
e
s
s
om
e
le
a
f
node
s
,
.
I
n
de
s
c
r
ip
ti
on
lo
gi
c
not
a
ti
on,
it
s
ta
te
s
,
ⅅ
.
I
n
th
e
s
y
s
te
m
,
th
e
c
ont
e
nt
of
le
a
r
ni
ng
is
in
th
e
dom
a
in
of
S
Q
L
(
s
tr
uc
tu
r
e
d
que
r
ie
d
la
ngua
g
e
)
f
r
om
w
hi
c
h
to
pi
c
s
–
th
a
t
w
e
ha
ve
c
a
ll
e
d t
he
D
e
s
ir
e
dC
onc
e
pt
ⅅ
a
r
e
c
hos
e
n
a
nd s
tu
di
e
d by s
tu
de
nt
s
.
N
ow
,
le
t
ⅅ
=
{
∈
ⅅ
|
(
)
}
a
nd
N
=
{
∈
|
(
)
}
.
ⅅ
pr
e
c
e
de
s
in
th
e
hi
e
r
a
r
c
hy
of
c
on
c
e
pt
s
(
or
to
pi
c
s
)
of
le
a
r
ni
ng
s
uc
h
th
a
t
th
e
num
be
r
of
e
le
m
e
nt
s
in
D
=
C
+
1
.
T
he
n
th
e
s
e
t
of
to
pi
c
s
ot
he
r
w
is
e
known
a
s
e
le
m
e
nt
s
c
on
s
id
e
r
e
d i
n t
he
doma
in
ⅅ
is
gi
ve
n a
s
ⅅ
= {uni
on, j
oi
n, update
, de
le
te
, i
ns
e
r
t,
s
e
le
c
t};
a
nd t
he
s
e
t
of
a
ll
pr
e
r
e
qui
s
it
e
s
unde
r
ne
a
th
ⅅ
is
gi
ve
n a
s
= {j
oi
n, update
, de
le
te
, i
ns
e
r
t,
s
e
le
c
t};
a
nd t
he
s
e
t
of
a
ll
t
e
r
m
in
a
l
le
a
f
node
s
in
a
nd
ⅅ
, r
e
s
pe
c
ti
ve
ly
, gi
ve
n a
s
N
=
{uni
on,
uni
onA
ll
,
s
e
lf
J
oi
n,
fu
ll
O
ut
e
r
J
oi
n,
in
ne
r
J
oi
n,
upd
at
e
Se
le
c
t,
update
W
he
r
e
,
de
le
t
e
Se
le
c
t,
de
le
te
W
he
r
e
, i
ns
e
r
tS
e
le
c
t,
i
ns
e
r
tV
al
ue
, s
e
le
c
tO
r
de
r
B
y
,
s
e
le
c
tDi
s
ti
nc
t,
s
e
le
c
tW
he
r
e
, s
e
le
c
tA
ll
}.
I
n
e
duc
a
ti
on,
te
a
c
hi
ng
-
le
a
r
ni
ng
is
c
hr
onol
ogi
c
a
l
a
nd
th
is
f
or
m
s
th
e
ba
s
is
f
or
th
e
c
onne
c
ti
on
of
a
pr
e
vi
ous
le
a
r
ni
ng
to
a
ne
w
or
ong
oi
ng
le
a
r
ni
ng.
L
e
t
th
e
r
e
la
ti
on
be
th
e
s
e
t
of
c
onn
e
c
ti
on
be
twe
e
n
node
s
i.
e
.
a
ne
w
to
pi
c
a
nd
a
pr
e
vi
ous
ly
le
a
r
ne
d
to
pi
c
.
T
he
n
w
e
s
ta
te
th
a
t
D
a
nd
C
,
a
nd
C
a
nd
N
be
lo
ng
to
s
om
e
r
e
la
ti
on
s
,
r
e
s
pe
c
ti
ve
ly
;
a
s
s
how
n i
n t
he
f
ol
lo
w
in
g s
e
t
A
a
nd
B
w
it
h r
e
ga
r
ds
t
o t
he
gi
ve
n ontol
ogi
c
a
l
node
r
e
la
ti
ons
hi
ps
A
=
D
x C
∈
R
a
nd, B
=
C
x N
∈
R
.
S
ym
bol
ic
a
ll
y, i
t
hol
ds
t
ha
t
∀
∈
ⅅ
∀
∈
∀
∈
,
(
ⅅ
,
)
∧
(
,
)
I
n f
ur
th
e
r
a
nc
e
, t
he
e
le
m
e
nt
s
of
t
he
s
e
ts
D
,
C
a
nd
N
th
us
s
a
ti
s
f
ie
s
t
he
f
ol
lo
w
in
g de
f
in
it
io
ns
a
nd t
he
ir
r
e
s
pe
c
ti
ve
pr
ope
r
ti
e
s
a
nd
in
pr
e
di
c
a
te
f
or
m
ul
a
s
(
d, c
)
=
{
∈
|
(
,
)
}
,
w
he
r
e
t
he
r
e
la
ti
on
=
has
P
r
e
r
e
qui
s
it
e
;
a
nd
(
c
, n
)
=
{
∈
|
(
,
)
}
, w
he
r
e
t
he
r
e
la
ti
on
=
has
K
B
.
S
ym
bol
ic
a
ll
y, t
he
c
onj
unc
ti
on of
t
he
a
bove
gi
ve
n r
e
la
ti
on i
s
∀
∈
ⅅ
∀
∈
∀
∈
, has
P
r
e
r
e
qui
s
it
e
(
d, c
)
∧
has
K
B
(
c
, n)
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
nt
J
A
r
ti
f
I
nt
e
ll
I
S
S
N
:
2252
-
8938
D
e
s
ig
n and analys
i
s
of
a m
ul
ti
-
age
nt
e
-
le
ar
ni
ng s
y
s
te
m
us
in
g
p
r
om
e
th
e
us
…
(
K
e
nne
d
y
E
. E
hi
m
w
e
nm
a
)
17
B
y t
he
pr
ope
r
ty
of
t
r
a
ns
it
iv
it
y,
has
P
r
e
r
e
qui
s
it
e
(
d, c
)
∧
has
K
B
(
c
, n)
→
has
K
B
(
d, n)
.
F
or
e
xa
m
pl
e
,
c
hoo
s
e
th
e
update
∈
ⅅ
.
I
ts
im
m
e
di
a
t
e
pr
e
r
e
qui
s
i
te
de
le
te
∈
.
T
he
nod
e
update
ha
s
it
s
le
a
dnode
s
n
=
{
update
Se
le
c
t,
update
W
he
r
e
}
,
a
nd
th
e
node
de
le
te
a
ls
o
ha
s
it
s
le
a
f
node
s
n
=
{
de
le
te
Se
le
c
t,
de
le
te
W
he
r
e
}
. F
r
om
t
he
f
or
e
goi
ng,
has
P
r
e
r
e
qui
s
it
e
(
update
,
de
le
te
)
∧
has
K
B
(
de
le
te
,
{
de
le
te
Se
le
c
t
,
de
le
te
W
he
r
e
})
→
has
K
B
(
update
,
{
de
le
te
Se
le
c
t,
de
le
te
W
he
r
e
})
.
T
he
n,
th
e
c
onc
lu
s
io
n
be
c
om
e
s
th
e
uni
ts
to
be
pr
e
-
te
s
te
d
on,
a
nd
if
a
ny
is
f
a
il
e
d;
s
tu
de
nt
s
a
r
e
r
e
c
om
m
e
nde
d
ma
te
r
ia
ls
f
or
th
e
f
a
il
e
d
uni
ts
to
le
a
r
n.
A
s
s
ta
te
d
e
a
r
li
e
r
,
if
no
pr
e
-
te
s
t
is
f
a
il
e
d,
i.
e
.
a
ll
pr
e
-
a
s
s
e
s
s
m
e
nt
s
a
r
e
pa
s
s
e
d;
t
he
n t
h
e
c
onc
lu
s
io
n a
s
i
n t
he
f
ol
lo
w
in
g t
hus
hol
d
has
P
r
e
r
e
qui
s
it
e
(
update
,
de
le
te
)
∧
has
K
B
(
de
le
te
,
{
de
le
te
Se
le
c
t
,
de
le
te
W
he
r
e
})
→
has
K
B
(
update
,
{
update
Se
le
c
t,
update
W
he
r
e
})
w
hi
c
h
a
r
e
th
e
le
a
f
node
s
of
th
e
c
hos
e
n
to
pi
c
.
T
he
s
ta
t
e
d
lo
gi
c
-
ba
s
e
d
r
e
pr
e
s
e
nt
a
ti
on,
f
or
m
e
d
th
e
gr
ound
f
a
c
t
in
te
r
na
l
knowle
dge
m
ode
l
of
t
he
m
ul
ti
-
a
ge
nt
pr
e
-
s
s
e
s
s
m
e
nt
s
ys
t
e
m
de
s
ig
ne
d i
n t
hi
s
s
tu
dy.
4.
D
I
S
C
U
S
S
I
O
N
T
he
pa
p
e
r
ha
s
pr
e
s
e
nt
e
d
th
e
pr
om
e
th
e
us
A
U
M
L
d
e
s
ig
n
to
ol
f
or
th
e
de
s
ig
n
a
nd
a
na
ly
s
is
of
th
e
pr
e
-
a
s
s
e
s
s
m
e
nt
s
y
s
te
m
,
a
nd
it
s
im
pl
e
m
e
nt
a
ti
on
w
it
h
J
a
s
on
–
a
J
a
va
-
ba
s
e
d
in
te
r
pr
e
te
r
a
nd
de
c
la
r
a
ti
ve
la
ngua
ge
.
T
he
c
hoi
c
e
of
P
r
om
e
th
e
us
m
e
th
odol
ogy
e
ns
ur
e
d
th
a
t
e
ve
r
y
r
e
qui
r
e
m
e
nt
a
nd
de
ta
il
e
d
d
e
s
ig
n
a
c
ti
vi
ty
w
e
r
e
c
a
pt
ur
e
d
w
it
h
th
e
a
ppr
op
r
ia
te
s
ym
bol
.
T
hi
s
w
e
ha
ve
de
pi
c
te
d
f
r
om
in
it
ia
l
goal
s
pe
c
if
ic
at
io
ns
,
to
s
ubgoals
,
to
a
ge
nt
r
ol
e
s
a
nd
in
te
r
a
c
ti
on
us
in
g
di
s
ti
nc
ti
ve
di
a
gr
a
m
s
.
F
r
om
c
r
it
ic
a
l
a
na
ly
s
is
,
P
r
om
e
th
e
us
pr
ovi
de
s
s
uppor
t
on
how
r
e
qui
r
e
m
e
nt
s
ta
te
m
e
nt
s
m
a
y
be
a
c
qui
r
e
d
--
s
ta
r
ti
ng
w
it
h
in
ti
al
goal
s
s
pe
c
if
ic
a
ti
on
--
a
s
w
e
ll
a
s
a
ge
ne
r
a
l
s
ys
te
m
a
r
c
hi
te
c
tu
r
e
a
s
a
ga
in
s
t
s
om
e
ot
he
r
A
U
M
L
to
ol
s
.
T
he
s
e
s
te
ps
a
r
e
vi
ta
l
a
s
a
ny
le
f
t
-
out
f
unc
ti
ona
li
ty
w
oul
d
c
a
us
e
a
voi
d
in
th
e
s
ys
te
m
:
A
voi
d
th
a
t
m
a
y
r
e
qui
r
e
th
e
r
e
-
e
ngi
ne
e
r
in
g
of
th
e
w
hol
e
s
ys
te
m
.
I
n
a
de
c
la
r
a
ti
ve
la
ngua
ge
, a
ge
nt
s
c
om
m
uni
c
a
t
e
vi
a
m
e
s
s
a
ge
pa
s
s
in
g
in
pr
e
di
c
a
te
lo
gi
c
f
or
m
.
T
hus
,
in
li
ne
w
it
h
th
e
r
e
por
te
d
m
e
c
ha
ni
s
m
of
pr
e
-
a
s
s
e
s
s
m
e
nt
&
r
e
c
om
m
e
nda
ti
on
a
nd
f
or
m
a
li
z
e
d
(
F
O
L
ba
s
e
d)
pr
e
-
a
s
s
e
s
s
m
e
nt
r
ul
e
s
[
35
-
37]
in
w
hi
c
h
th
e
M
A
S
m
a
de
a
c
c
ur
a
te
r
e
c
om
m
e
nda
ti
on
a
f
te
r
pr
e
-
a
s
s
e
s
s
m
e
nt
,
F
ig
ur
e
13
he
r
e
by
pr
e
s
e
nt
s
th
e
ps
e
udoc
ode
of
th
e
op
e
r
a
ti
on
of
th
e
s
y
s
te
m
a
nd
how
th
e
p
e
r
c
e
iv
e
d
knowle
dg
e
by
a
ge
nt
s
a
r
e
us
e
d:
f
r
om
pe
r
c
e
pt
a
c
qui
s
it
io
n
a
t
th
e
in
te
r
f
a
c
e
(
li
ne
7)
,
th
r
ough
to
ot
he
r
a
ge
nt
s
vi
a
th
e
.s
e
nd(
)
in
te
r
nal
ac
ti
on
[
4]
(
on
li
ne
s
9,
11,
18,
23,
27;
a
s
s
how
n
in
F
ig
ur
e
13)
w
hi
c
h
c
le
a
r
ly
s
how
s
th
e
num
be
r
of
in
te
r
a
c
ti
ve
a
ge
nt
s
i
n t
he
s
y
s
te
m
. B
e
tw
e
e
n e
a
c
h
in
te
r
nal
ac
ti
on
, i
s
t
he
a
c
ti
o
n de
s
ig
na
te
d f
or
a
r
e
c
e
iv
in
g a
ge
nt
t
o e
xe
c
ut
e
.
F
ig
ur
e
13. P
s
e
udo
-
a
lg
or
it
hm
of
t
he
pr
e
-
a
s
s
e
s
s
m
e
nt
pr
oc
e
s
s
t
ha
t
de
pe
nds
on t
he
numbe
r
of
l
e
a
f
no
de
s
N
c
ons
id
e
r
e
d unde
r
a
de
s
ir
e
dC
onc
e
pt
P
seudoco
de of p
re
-
ass
es
s
m
ent
a
nd inte
racti
on in the
multia
gent s
y
stem
1.
initial beliefs: predi
cate(Class, C
lass)
2.
i
ni
tial beliefs: predicate(Class, L
eafnode)
3.
i
nitial beliefs: predicate(L
eafnode, URL)
4.
i
nitial beliefs: quiz(PrerequisiteLeafnode)
5.
Given
a desired concept that has
N
leafn
odes
prerequisite
6.
IF
7.
Percept ← desiredConcept
8.
THEN
9.
.send(rec
eiver, tell, desiredConcept)
10.
fetch
the
next quiz(
Prerequisite_Leafnode
)
11.
.send(receiver, tell, quiz(
Prerequisite_Leafnode
)
12.
o
utput quiz(
Prerequisite_Leafnode
)
13.
Percept
← answer(X)
14.
IF
15.
a
nswer(X) == answer(Prerequisite_Leafnode)
16.
THEN
17.
passed(Prerequisite_Leafnode) decision
18.
.send(receiver, tell, passed(
Prerequisite_Leafnode
)
19.
IF
20.
answer(X)
\
== answer(Prerequisite_Leafnode)
21.
THEN
22.
failed(Prerequisite_Leafnode) decision
23.
.send(receiver, tell, failed(
Prerequisi
te_Leafnode
)
24.
IF
25.
N number of
leafnodes
have been
pre
-
assessed on
26.
THEN
27.
.send(receiver, achieve, recommendMaterial)
28.
Else
29.
repeat 10 to 27
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
2252
-
8938
I
nt
J
A
r
ti
f
I
nt
e
ll
,
V
ol
.
10
, N
o.
1
,
M
a
r
c
h
20
2
1
:
9
–
23
18
4.1. Re
s
ol
vi
n
g
is
s
u
e
s
of
d
e
ve
lo
p
m
e
n
t
A
s
a
pr
e
-
a
s
s
e
s
s
m
e
nt
s
ys
te
m
f
or
S
Q
L
pr
e
-
s
ki
ll
s
te
s
t
a
nd
s
ki
ll
s
c
la
s
s
if
ic
a
ti
on
in
to
two
[
1,
0]
bi
na
r
y
s
ta
te
s
,
th
e
s
ys
te
m
r
e
c
e
iv
e
s
ope
n
-
e
nde
d
S
Q
L
que
r
y
in
put
s
th
a
t
m
a
ybe
c
or
r
e
c
t
or
in
c
or
r
e
c
t
a
ns
w
e
r
f
or
a
pa
r
ti
c
ul
a
r
que
r
y
que
s
ti
on.
F
or
a
g
iv
e
n
que
r
y,
th
e
in
f
lo
w
of
in
c
o
r
r
e
c
t
a
ns
w
e
r
s
to
th
e
s
ys
te
m
is
not
de
f
in
it
e
nor
pr
e
de
te
r
m
in
e
d
c
om
pa
r
e
d
to
c
or
r
e
c
t
que
r
y
a
ns
w
e
r
s
th
a
t
a
r
e
known
a
nd
pr
e
de
f
in
e
d
in
t
he
s
ys
te
m
.
T
hus
,
pr
ogr
a
m
m
in
g
a
M
A
S
f
or
th
e
r
e
c
ogni
ti
on
of
ne
gat
i
v
e
fa
c
t
s
(
i.
e
.
in
c
or
r
e
c
t
a
ns
w
e
r
s
)
c
a
n
pos
e
s
om
e
di
f
f
ic
ul
ty
f
or
a
ge
nt
pl
a
n
s
e
le
c
ti
on
a
nd
e
xe
c
ut
io
n
of
a
ge
nt
goal
s
w
he
n
t
he
e
xpe
c
te
d
in
put
s
m
a
y
va
r
y
de
pe
ndi
ng
on
a
s
tu
de
nt
unde
r
s
ta
ndi
ng
a
nd
que
r
y
c
om
pe
te
nc
e
s
.
I
n
s
uc
h
c
a
s
e
s
,
in
put
s
be
c
om
e
di
ve
r
s
e
,
unbounde
d
a
nd
s
ubj
e
c
ti
ve
to
th
e
s
tu
de
nt
s
.
O
n
th
e
ha
nd,
th
e
c
or
r
e
c
t
S
Q
L
q
ue
r
ie
s
w
hi
c
h
a
r
e
th
e
pos
it
iv
e
fa
c
ts
a
r
e
qui
te
s
tr
a
ig
ht
f
or
w
a
r
d
to
pr
ogr
a
m
be
c
a
us
e
e
ve
r
y
a
n
s
w
e
r
to
th
e
qu
e
s
ti
ons
a
s
k
e
d
on
th
e
s
ys
te
m
i
s
pr
e
de
te
r
m
in
e
d
ba
s
e
d
on
s
ta
nda
r
d
S
Q
L
que
r
ie
s
.
T
ypi
c
a
ll
y,
th
e
s
ynt
a
x
of
J
a
s
on
a
ge
nt
pl
a
n
c
om
pr
is
e
s
th
r
e
e
pa
r
ts
,
a
nd
th
e
s
tr
uc
tu
r
e
gi
ve
n
a
s
:
tr
ig
ge
r
in
g_e
v
e
nt
,
c
ont
e
x
t
<
--
body
[
4]
.
T
he
c
ont
e
x
t
is
th
e
pa
r
t
of
th
e
pl
a
n
w
hi
c
h
s
ta
te
s
th
e
pr
e
-
c
ondi
ti
on
th
a
t
a
c
ti
va
te
s
a
pl
a
n
f
or
e
x
e
c
ut
io
n.
B
y
d
e
f
a
ul
t,
a
bl
a
nk
pl
a
n
c
ont
e
x
t
is
tr
ue
f
or
a
ll
b
e
li
e
f
s
in
th
e
a
ge
nt
.
O
th
e
r
w
is
e
,
a
pr
e
di
c
a
te
f
or
m
pr
e
-
c
ondi
ti
on
m
us
t
be
s
ta
te
d
to
c
ont
r
ol
w
ha
t
pl
a
n
is
r
ig
ht
f
or
a
gi
ve
n
tr
ig
ge
r
in
g_e
v
e
nt
a
nd
be
li
e
f
s
.
T
o
de
te
r
m
in
e
w
he
th
e
r
a
que
r
y
in
put
is
r
ig
ht
o
r
w
r
ong,
th
e
pr
e
de
f
in
e
d
pos
it
iv
e
fa
c
ts
w
e
r
e
r
e
pr
e
s
e
nt
e
d
in
th
e
pr
e
-
a
s
s
e
s
s
m
e
nt
a
ge
nt
in
f
ir
s
t
or
de
r
lo
gi
c
(
F
O
L
)
pr
e
di
c
a
te
f
or
m
–
f
or
a
de
c
la
r
a
ti
ve
l
a
ngua
ge
.
W
he
n
a
n
a
ge
nt
ge
ts
pe
r
c
e
pt
(
a
ls
o
in
F
O
L
f
or
m
)
,
th
e
a
ge
nt
m
a
tc
he
s
th
a
t
pe
r
c
e
pt
(
now
a
be
li
e
f
)
a
ga
in
s
t
a
ll
pl
a
ns
, t
he
r
e
le
va
nt
pl
a
n i
s
s
e
le
c
te
d a
nd a
c
ti
ons
i
n t
he
body
of
t
he
pl
an
is
e
xe
c
ut
e
d
-
-
-
in
O
ne
v
s
. A
l
l
a
ppr
oa
c
h.
T
hi
s
i
s
be
c
a
us
e
pos
it
iv
e
fa
c
ts
a
r
e
in
f
or
m
a
ti
on
w
hos
e
r
e
pr
e
s
e
nt
a
ti
on
a
r
e
known
a
nd
c
a
n
be
r
e
pr
e
s
e
nt
e
d
or
gi
ve
n
to
th
e
a
ge
nt
(
i.
e
.
r
e
s
pon
s
ib
le
f
or
ha
ndl
in
g
pr
e
-
s
ki
ll
s
a
s
s
e
s
s
m
e
nt
)
f
or
c
om
pa
r
is
on
w
it
h
in
c
om
in
g
pe
r
c
e
pt
s
.
S
o,
th
e
c
or
r
e
c
t
S
Q
L
que
r
ie
s
w
e
r
e
in
it
ia
li
z
e
d
in
th
e
a
ge
nt
’
s
be
li
e
f
ba
s
e
a
nd
w
e
r
e
u
s
e
d
by
th
e
a
ge
nt
to
m
a
tc
h
a
nd
tr
ig
ge
r
r
e
le
va
nt
pl
a
ns
a
nd
a
ge
nt
goal
s
a
s
ne
e
de
d.
B
ut
ne
gat
iv
e
fa
c
ts
a
r
e
unknown
a
nd
a
s
s
u
c
h
c
a
nnot
be
pr
e
-
de
te
r
m
in
e
d
f
or
r
e
pr
e
s
e
nt
a
ti
on
a
s
m
e
nt
io
ne
d
e
a
r
li
e
r
.
S
o,
to
a
ddr
e
s
s
in
c
or
r
e
c
t
S
Q
L
que
r
y
in
put
s
,
J
a
s
on
di
ff
e
r
e
nt
\
==
ope
r
a
to
r
[
4]
w
a
s
us
e
d
a
s
th
e
c
om
pa
r
is
on
ope
r
a
to
r
in
th
e
a
ge
nt
pl
a
n
c
ont
e
x
t
.
I
n
a
na
lo
gy,
th
e
ope
r
a
to
r
m
e
a
n
s
“
n
ot
e
qu
al
”
or
“
f
al
s
e
”
.
B
ut
th
e
us
e
of
th
is
ope
r
a
to
r
w
a
s
not
w
it
hout
in
c
ons
is
te
nc
y
in
th
e
c
ol
le
c
ti
ve
m
ul
ti
-
a
ge
nt
s
b
e
ha
vi
our
dur
in
g
s
ys
te
m
c
odi
ng
a
nd
im
pl
e
m
e
nt
a
ti
on
pha
s
e
.
D
ur
in
g
c
odi
ng,
w
he
n
a
n
in
c
or
r
e
c
t
que
r
y
w
a
s
in
put
te
d
f
or
te
s
ti
ng,
th
e
\
=
=
ope
r
a
to
r
m
a
de
th
e
pr
e
-
a
s
s
e
s
s
m
e
nt
a
ge
nt
to
m
is
s
-
s
e
le
c
t
pl
a
ns
f
r
om
it
s
pl
a
n
li
br
a
r
y.
F
or
e
xa
m
pl
e
,
by
s
ta
ti
ng
in
a
pl
a
n
th
a
t
w
a
s
e
x
pe
c
te
d
to
ha
ndl
e
a
n i
nc
or
r
e
c
t
S
Q
L
que
r
y, t
ha
t,
if
th
e
ans
w
e
r
pe
r
c
e
pt
c
om
in
g
in
to
th
e
s
y
s
te
m
doe
s
n
ot
m
at
c
h
s
om
e
in
it
ia
ll
y
pr
e
de
fi
ne
d
SQ
L
que
r
ie
s
th
e
n
in
fo
r
m
th
e
s
tu
de
nt
th
at
th
e
ans
w
e
r
gi
v
e
n
is
in
c
or
r
e
c
t
and
th
e
n
s
e
le
c
t
th
e
ne
x
t
le
af
node
que
s
ti
on and pr
e
s
e
nt
t
o t
he
s
tu
d
e
nt
.
L
it
e
r
a
ll
y, f
r
om
t
he
be
ha
vi
our
e
xhi
bi
te
d by
t
he
a
ge
nt
, t
he
a
ge
nt
’
s
i
nt
e
r
pr
e
ta
ti
on w
a
s
a
ny othe
r
pl
a
n
w
hos
e
pl
a
n
c
ont
e
x
t
ha
s
no
m
a
t
c
h
to
a
ny
a
lr
e
a
dy
known
knowle
dge
in
th
e
a
ge
nt
’
s
be
li
e
f
ba
s
e
.
T
h
e
m
is
s
-
s
e
le
c
ti
on
of
pl
a
ns
w
a
s
due
to
s
om
e
unc
e
r
ta
in
ty
in
th
e
a
ge
nt
a
bi
li
ty
to
m
a
p
a
n
in
c
or
r
e
c
t
que
r
y
pe
r
c
e
pt
to
be
li
e
f
s
.
T
hi
s
be
ha
vi
our
a
s
obs
e
r
ve
d
a
dve
r
s
e
ly
a
lt
e
r
e
d
th
e
or
de
r
of
s
ubs
e
que
nt
go
a
l/
que
s
ti
on
s
e
le
c
ti
on
of
a
pr
e
r
e
qui
s
it
e
’
s
le
a
f
node
N
,
in
c
ont
r
a
s
t
to
th
e
a
r
r
a
nge
m
e
nt
s
of
node
s
in
th
e
ont
ol
ogy
tr
e
e
.
T
hi
s
w
a
s
a
non
-
tr
iv
ia
l
pr
obl
e
m
.
A
t
th
e
im
pl
e
m
e
nt
a
ti
on
pha
s
e
,
one
of
th
e
ke
y
pr
in
c
ip
le
s
of
s
of
twa
r
e
m
e
th
odol
ogy
is
to
c
om
bi
ne
c
odi
ng
a
nd
te
s
ti
ng
[
38]
.
T
hi
s
pr
in
c
ip
le
w
hi
c
h
e
na
bl
e
s
a
s
y
s
te
m
to
be
in
ve
s
ti
ga
te
d
w
hi
le
it
is
s
ti
ll
be
in
g
de
ve
lo
pe
d
e
ns
ur
e
d
th
a
t
th
is
non
-
tr
iv
ia
l
pr
obl
e
m
w
a
s
c
he
c
ke
d be
f
or
e
t
he
s
y
s
te
m
w
a
s
c
o
m
pl
e
te
ly
bui
lt
.
T
o
e
na
bl
e
th
e
pr
e
-
a
s
s
e
s
s
m
e
nt
a
ge
nt
,
a
s
s
ho
w
n
in
F
ig
ur
e
s
8
-
9
to
a
c
c
ur
a
te
ly
s
e
le
c
t
r
e
le
va
nt
pl
a
n(
s
)
f
or
a
m
a
tc
h
of
it
s
pl
a
n
c
ont
e
xt
to
th
e
pe
r
c
e
pt
th
a
t
i
s
a
dopt
e
d
in
th
e
\
=
=
ope
r
a
to
r
;
a
nd
to
c
or
r
e
c
tl
y
de
te
r
m
in
e
th
e
ne
xt
a
ppr
opr
ia
te
a
ge
nt
goa
l
a
nd
a
c
c
ur
a
te
m
e
s
s
a
ge
pa
s
s
in
g
to
ot
he
r
a
ge
nt
s
,
w
e
h
a
d
to
in
tr
oduc
e
a
pr
oc
e
s
s
of
it
e
r
a
ti
on
th
a
t
c
oul
d
c
ount
pl
a
n
s
e
le
c
ti
on
in
th
e
a
ge
nt
f
or
e
ve
r
y
pa
r
e
nt
node
(
o
r
to
pi
c
)
a
nd
th
e
ir
c
onne
c
te
d
le
a
f
node
s
N
.
I
n
J
a
s
on
pr
e
di
c
a
te
lo
gi
c
f
or
m
,
a
n
e
xa
m
pl
e
s
ynt
a
x
of
th
is
it
e
r
a
ti
on
is
c
ount
F
or
D
e
le
te
P
r
e
(
X
)
,
w
hi
c
h
de
pi
c
ts
th
e
c
ount
e
r
f
or
th
e
de
le
te
node
w
he
r
e
X
is
a
pos
i
ti
ve
in
te
ge
r
.
I
n
a
ddi
ti
on,
th
e
ne
ga
ti
on
of
s
om
e
in
c
om
in
g
pe
r
c
e
pt
w
a
s
r
e
qui
r
e
d
to
s
to
p
uns
ol
ic
it
e
d
pl
a
n
tr
ig
ge
r
.
A
n
e
xa
m
pl
e
of
s
uc
h
ne
ga
ti
on
in
th
e
c
ont
e
x
t
pa
r
t
of
a
pl
a
n,
w
a
s
th
e
not
d
e
s
ir
e
dC
onc
e
pt
(
“
in
s
e
r
t
”
)
w
hi
c
h
w
a
s
us
e
d
to
bl
oc
k
-
of
f
th
e
de
s
ir
e
d
C
onc
e
pt
(
“
in
s
e
r
t”
)
in
a
pl
a
n
c
ont
e
x
t
s
o
a
s
not
tr
ig
ge
r
th
e
w
r
ong
pl
a
n
a
nd
w
r
ong
a
ge
nt
go
a
l
a
t
a
gi
ve
n
ti
m
e
.
A
s
th
e
s
y
s
te
m
ke
pt
e
xpa
ndi
ng
w
it
h
th
e
pr
ogr
a
m
m
in
g
of
m
or
e
pa
r
e
nt
node
s
D
a
nd
l
e
a
f
node
s
N
b
e
e
n
a
dde
d,
th
is
bl
oc
k
-
of
f
c
ont
in
ue
d
a
nd
it
us
e
d
to
m
it
ig
a
te
a
ge
nt
a
nom
a
ly
be
ha
vi
our
.
T
he
s
e
two
c
om
bi
ne
d
s
tr
a
te
gi
e
s
e
f
f
e
c
ti
ve
ly
c
ont
r
ol
le
d
a
g
e
nt
be
ha
vi
our
a
s
w
e
ll
a
s
th
e
e
nt
ir
e
m
ul
ti
-
a
ge
nt
s
to
w
a
r
d
ha
ndl
in
g
of
th
e
in
c
or
r
e
c
t
S
Q
L
que
r
y i
nput
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.