T
E
L
KO
M
N
I
KA
T
e
lec
om
m
u
n
icat
ion
,
Com
p
u
t
i
n
g,
E
lec
t
r
on
ics
an
d
Cont
r
ol
Vol.
18
,
No.
2
,
Apr
il
2020
,
pp
.
8
90
~
89
8
I
S
S
N:
1693
-
6930,
a
c
c
r
e
dit
e
d
F
ir
s
t
G
r
a
de
by
Ke
me
nr
is
tekdikti
,
De
c
r
e
e
No:
21/E
/KP
T
/2018
DO
I
:
10.
12928/
T
E
L
KO
M
NI
KA
.
v18i2.
14866
890
Jou
r
n
al
h
omepage
:
ht
tp:
//
jour
nal.
uad
.
ac
.
id/
index
.
php/T
E
L
K
OM
N
I
K
A
U
se
r
st
o
r
ie
s c
ol
le
c
t
io
n
vi
a i
n
t
e
r
ac
t
iv
e
c
h
at
b
o
t
t
o
su
p
p
o
r
t
r
e
q
u
i
r
e
m
e
n
t
s gat
h
e
r
i
n
g
F
e
r
li
an
a
Dw
it
a
m
a
,
And
r
e
Rus
li
D
ep
ar
t
men
t
o
f
In
f
o
rmat
i
cs
,
U
n
i
v
ers
i
t
a
s
Mu
l
t
i
me
d
i
a
N
u
s
an
t
ara,
In
d
o
n
es
i
a
Ar
t
icle
I
n
f
o
AB
S
T
RA
CT
A
r
ti
c
le
h
is
tor
y
:
R
e
c
e
ived
J
ul
7
,
2019
R
e
vis
e
d
J
a
n
7
,
2020
Ac
c
e
pted
F
e
b
20
,
2020
N
o
w
ad
a
y
s
,
s
o
ft
w
are
p
r
o
d
u
ct
s
h
av
e
b
eco
me
a
n
es
s
e
n
t
i
al
p
ar
t
o
f
h
u
ma
n
l
i
fe.
T
o
b
u
i
l
d
s
o
f
t
w
are,
d
ev
e
l
o
p
ers
mu
s
t
h
av
e
a
g
o
o
d
u
n
d
ers
t
an
d
i
n
g
o
f
t
h
e
req
u
i
reme
n
t
s
o
f
t
h
e
s
o
f
t
w
ar
e.
H
o
w
e
v
er,
s
o
ft
w
are
d
ev
el
o
p
er
s
t
en
d
t
o
j
u
mp
s
t
art
s
y
s
t
em
co
n
s
t
r
u
ct
i
o
n
w
i
t
h
o
u
t
h
a
v
i
n
g
a
cl
ear
an
d
d
e
t
ai
l
ed
u
n
d
ers
t
an
d
i
n
g
o
f
t
h
e
re
q
u
i
remen
t
s
.
T
h
e
u
s
er
s
t
o
r
y
co
n
ce
p
t
i
s
o
n
e
o
f
t
h
e
p
rac
t
i
ce
s
o
f
t
h
e
req
u
i
remen
t
s
el
i
ci
t
at
i
o
n
.
T
h
i
s
p
a
p
er
ai
ms
t
o
p
re
s
en
t
t
h
e
w
o
rk
co
n
d
u
ct
e
d
t
o
d
e
v
el
o
p
a
n
A
n
d
ro
i
d
ch
a
t
b
o
t
a
p
p
l
i
ca
t
i
o
n
t
o
s
u
p
p
o
rt
t
h
e
req
u
i
r
emen
t
s
el
i
ci
t
at
i
o
n
act
i
v
i
t
y
i
n
s
o
ft
w
are
en
g
i
n
eer
i
n
g
,
mak
i
n
g
t
h
e
w
o
r
k
l
es
s
t
i
me
-
c
o
n
s
u
m
i
n
g
an
d
s
t
r
u
ct
u
red
ev
e
n
fo
r
u
s
er
s
n
o
t
accu
s
t
o
med
t
o
req
u
i
reme
n
t
s
en
g
i
n
eeri
n
g
.
T
h
e
ch
at
b
o
t
u
s
es
N
az
i
ef
&
A
d
ri
a
n
i
s
t
emmi
n
g
al
g
o
ri
t
h
m
t
o
p
re
-
p
ro
ce
s
s
t
h
e
n
a
t
u
ra
l
l
a
n
g
u
ag
e
i
t
rece
i
v
e
s
fro
m
t
h
e
u
s
ers
an
d
art
i
f
i
ci
a
l
mark
-
u
p
l
an
g
u
a
g
e
(A
IML
)
as
t
h
e
k
n
o
w
l
e
d
g
e
b
a
s
e
t
o
p
r
o
ces
s
t
h
e
b
o
t
’
s
res
p
o
n
s
es
.
A
p
rel
i
mi
n
ary
accep
t
an
ce
t
es
t
b
as
ed
o
n
t
h
e
t
ec
h
n
o
l
o
g
y
accep
t
a
n
ce
mo
d
el
re
s
u
l
t
s
i
n
a
n
8
3
.
0
3
%
s
co
re
f
o
r
u
s
ers
’
b
eh
av
i
o
r
al
i
n
t
e
n
t
i
o
n
t
o
u
s
e.
K
e
y
w
o
r
d
s
:
AI
M
L
C
ha
tbot
Na
z
ief
&
Adr
iani
a
lgor
it
hm
R
e
quir
e
ments
e
li
c
it
a
ti
on
U
s
e
r
s
tor
ies
Th
i
s
i
s
a
n
o
p
en
a
c
ces
s
a
r
t
i
c
l
e
u
n
d
e
r
t
h
e
CC
B
Y
-
SA
l
i
ce
n
s
e
.
C
or
r
e
s
pon
din
g
A
u
th
or
:
F
e
r
li
a
na
Dw
it
a
ma
,
De
pa
r
tm
e
nt
of
I
nf
or
mat
ics
,
Unive
r
s
it
a
s
M
ult
im
e
dia
Nus
a
ntar
a
,
S
c
ientia
B
ouleva
r
d
S
t.
,
Ga
ding
,
S
e
r
pong,
T
a
nge
r
a
ng,
B
a
nten
15227,
I
ndone
s
ia.
E
mail:
f
e
r
li
a
na
.
dwitama
@s
tudent.
umn.
a
c
.
id
1.
I
NT
RODU
C
T
I
ON
S
of
twa
r
e
is
a
n
im
por
tant
a
s
pe
c
t
that
ha
s
a
n
im
p
a
c
t
on
human
li
f
e
,
s
uc
h
a
s
wor
k,
da
il
y
a
c
ti
vit
y,
f
inanc
e
,
a
nd
other
e
s
s
e
nti
a
l
f
ields
.
T
he
r
e
a
r
e
s
ome
s
teps
to
buil
d
s
of
twa
r
e
,
a
nd
de
ve
loper
s
mus
t
ha
v
e
a
c
lea
r
unde
r
s
tanding
of
the
r
e
quir
e
ments
of
the
s
of
twa
r
e
.
How
e
ve
r
,
the
de
ve
loper
tends
to
buil
d
s
of
twa
r
e
without
ha
ving
a
c
lea
r
unde
r
s
tanding
o
f
the
r
e
quir
e
ments
in
de
tail
be
c
a
us
e
f
or
them
ha
ving
unde
r
s
tood
r
e
quir
e
ments
in
de
tail
is
a
wa
s
te
of
ti
me,
in
whic
h
r
e
quir
e
ments
a
r
e
c
ons
tantly
c
ha
nging
[
1]
.
T
he
s
e
thi
ngs
lea
d
to
s
of
twa
r
e
f
a
il
ur
e
be
c
a
us
e
whe
n
the
de
ve
loper
doe
s
not
ha
ve
a
c
lea
r
unde
r
s
tandi
ng
of
the
s
ys
tem,
a
n
e
r
r
or
c
a
n
oc
c
ur
in
s
ys
tem
de
s
ign
a
nd
the
ne
xt
s
teps
[
2]
.
R
e
quir
e
ments
e
nginee
r
ing
is
a
n
a
c
ti
vit
y
that
h
a
s
the
a
im
to
know,
unde
r
s
tand,
a
na
lyze
,
a
nd
doc
umenting
wha
t
a
r
e
the
r
e
qui
r
e
ments
that
ne
e
de
d
by
s
take
holder
[
1
]
,
s
o
invol
ve
ment
of
s
take
h
older
s
is
c
r
uc
ial
in
s
of
twa
r
e
pr
ojec
t
e
s
pe
c
ially
in
r
e
qui
r
e
me
nts
e
nginee
r
ing
pr
oc
e
s
s
[
1,
3]
.
One
of
t
he
s
teps
r
e
quir
e
ments
e
nginee
r
ing
is
r
e
quir
e
ments
e
li
c
it
a
t
ion
whic
h
us
e
r
s
tor
y
is
one
o
f
i
ts
pr
a
c
ti
c
e
.
Us
e
r
s
tor
ies
int
r
oduc
e
d
the
f
i
r
s
t
ti
me
to
e
xt
r
e
me
pr
og
r
a
mi
ng
in
a
gil
e
de
ve
lopm
e
nt
methods
a
nd
s
tar
ted
us
e
d
in
other
methods
,
s
uc
h
a
s
S
c
r
um,
[
4]
.
Us
e
r
s
tor
ies
f
oc
us
on
int
e
r
a
c
ti
on
with
the
us
e
r
,
s
o
us
e
r
s
c
a
n
pa
r
ti
c
i
pa
te
a
nd
de
s
c
r
ibe
the
f
unc
ti
ons
that
c
a
n
be
us
e
f
ul
f
or
s
take
holder
s
.
A
c
ha
tbot
is
a
c
om
puter
a
ge
nt
that
c
a
n
int
e
r
a
c
t
with
the
us
e
r
a
nd
make
hu
man
-
bot
c
onve
r
s
a
ti
on
f
e
e
ls
li
ke
they
we
r
e
a
human
-
human
c
onve
r
s
a
ti
on.
B
a
s
e
d
on
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
Us
e
r
s
to
r
ies
c
oll
e
c
ti
on
v
ia
int
e
r
ac
ti
v
e
c
hatbot
to
s
u
ppor
t
r
e
quir
e
me
nt
s
gather
ing
(F
e
r
li
ana
Dw
it
am
)
891
pr
e
vious
r
e
s
e
a
r
c
h,
na
tur
a
l
langua
ge
pr
oc
e
s
s
ing
u
s
e
d
f
or
c
ha
tbot
c
a
n
int
e
r
a
c
t
a
nd
unde
r
s
tand
l
ike
a
human
[5
–
7]
.
B
a
s
e
d
on
a
s
ur
ve
y
by
O
r
a
c
le,
mo
r
e
than
7
0%
of
the
r
e
s
ponde
nts
s
a
id
that
pe
ople
o
r
bus
ine
s
s
pe
ople
a
lr
e
a
dy
us
e
d
c
ha
tbot
or
ha
ve
a
plan
to
us
e
a
c
ha
tbo
t
in
2020
[
8]
.
W
he
n
us
ing
a
c
ha
tbot
,
the
c
us
tom
e
r
doe
s
not
ne
e
d
to
wa
s
te
their
ti
me
to
mee
t
f
a
c
e
to
f
a
c
e
to
g
e
t
s
pe
c
i
f
ic
s
e
r
vice
whic
h
make
s
us
e
r
s
s
a
ve
the
c
os
t;
thes
e
thi
ngs
make
c
ha
tbot
ha
s
br
ought
pos
it
ive
im
pa
c
t
f
o
r
both
c
us
tom
e
r
s
a
nd
the
c
ompany
[
9]
.
S
e
nding
mes
s
a
ge
s
is
the
c
or
e
of
the
mob
il
e
e
xpe
r
i
e
nc
e
[
10]
,
a
nd
the
incr
e
a
s
e
in
in
ter
ne
t
a
nd
mobi
le
de
vice
s
in
the
p
r
e
s
e
nt
make
s
the
int
e
r
a
c
ti
on
be
twe
e
n
humans
e
a
s
ier
[
11]
.
B
a
s
e
d
on
a
s
ur
ve
y
in
2017
,
mobi
le
de
vice
us
e
r
s
in
I
ndone
s
ia
a
lr
e
a
dy
r
e
a
c
h
341,
4
mi
ll
ion
[
12]
.
Ar
ti
f
icia
l
mar
k
-
up
langua
ge
(
A
I
M
L
)
is
a
n
XM
L
-
ba
s
e
d
mar
k
-
up
langua
ge
that
us
e
d
f
o
r
b
r
a
in
or
knowle
dge
of
the
c
ha
tbot
whic
h
us
e
d
the
f
i
r
s
t
ti
me
in
AL
I
C
E
bot
[
13,
14]
.
I
n
other
r
e
s
e
a
r
c
he
s
,
us
ing
AI
M
L
f
or
the
br
a
in
of
the
c
ha
tbot
is
a
s
uc
c
e
s
s
in
c
r
e
a
ti
ng
int
e
r
a
c
ti
on
be
twe
e
n
bot
a
nd
us
e
r
.
T
he
Na
z
ief
&
Adr
iani
a
lgor
i
thm
is
a
s
temmi
ng
a
lgor
it
hm
c
omm
only
us
e
d
f
or
B
a
ha
s
a
I
ndone
s
ia,
whic
h
is
c
ons
ider
e
d
a
s
the
qua
li
f
ief
one
,
c
ompar
e
d
to
o
ther
s
temmi
ng
a
lgor
it
hms
[
15]
.
T
ha
t
a
lgor
it
hm
us
e
d
be
c
a
us
e
c
ha
tbot
us
ing
B
a
ha
s
a
I
ndone
s
ia
a
nd
Na
z
ief
&
Adr
iani
a
lgor
it
h
m
ha
ve
h
igher
a
c
c
u
r
a
c
y
f
or
the
I
ndone
s
ian
langua
ge
s
temmer
than
o
ther
a
lgo
r
it
hms
s
uc
h
a
s
por
ter
[
16]
.
Othe
r
wo
r
ks
ha
ve
a
ls
o
s
hown
the
potential
of
Na
z
ief
&
Ad
r
iani
a
lgor
it
hm
in
s
temmi
ng
na
tur
a
l
langua
ge
s
in
B
a
ha
s
a
I
ndone
s
ia
[
17
–
19]
.
F
ur
ther
mor
e
,
c
ha
tbot
f
or
r
e
late
d
wor
ks
of
li
ter
a
t
u
r
e
c
onc
e
r
ning
the
de
ve
lopm
e
nt
c
ha
tbot
in
B
a
ha
s
a
I
ndone
s
ia
ha
s
a
ls
o
s
hown
the
us
e
of
c
ha
tbot
in
va
r
ious
f
i
e
lds
[
20
–
22]
;
howe
ve
r
,
the
us
e
o
f
c
ha
tbot
a
nd
Na
z
ief
&
Adr
iani
a
lgor
it
hm
in
the
r
e
quir
e
ments
e
nginee
r
ing
f
ield
is
s
ti
ll
r
a
r
e
to
be
f
ound.
T
his
r
e
s
e
a
r
c
h
a
im
s
to
br
idge
the
ga
p
be
twe
e
n
the
s
take
holder
s
a
nd
the
s
of
twa
r
e
e
nginee
r
s
in
te
r
ms
of
r
e
quir
e
ments
ga
ther
ing
by
de
ve
lopi
ng
a
c
ha
tb
ot
f
o
r
the
s
take
holder
s
to
c
ha
t
with
whe
n
s
u
bmi
tt
ing
pr
e
li
mi
na
r
y
r
e
quir
e
ments
a
nd/o
r
us
e
r
f
e
e
dba
c
ks
r
e
ga
r
ding
a
s
of
twa
r
e
p
r
oduc
t.
T
he
c
ha
tbot
us
e
s
AI
M
L
a
s
it
s
knowle
dge
ba
s
e
,
Na
z
ief
&
Adr
iani
s
temmi
ng
a
l
gor
it
hm,
a
nd
f
oll
ows
the
Us
e
r
S
tor
ies
f
o
r
mat
whic
h
will
r
e
quir
e
the
us
e
r
s
to
pr
ov
ide
inf
or
mation
r
e
ga
r
di
ng
his
/her
r
ole,
ne
e
ds
,
a
nd
bus
ines
s
va
lue
in
a
na
tur
a
l
c
onve
r
s
a
ti
on.
T
he
bot
’
s
pe
r
f
o
r
manc
e
wa
s
mea
s
ur
e
d
us
ing
s
e
ve
r
a
l
c
ha
tbot
met
r
ics
,
a
nd
a
pr
e
li
mi
n
a
r
y
us
e
r
a
c
c
e
ptanc
e
tes
t
f
oll
owing
the
tec
hnology
a
c
c
e
ptanc
e
model
wa
s
c
onduc
ted
f
or
e
va
luation.
2.
RE
S
E
AR
CH
M
E
T
HO
D
I
n
thi
s
r
e
s
e
a
r
c
h,
the
c
ha
tbot
a
ppli
c
a
ti
on
buil
t
us
i
ng
the
Na
z
i
e
f
&
Adr
iani
s
temmi
ng
a
lgo
r
it
hm
f
or
the
s
temmi
ng
pr
oc
e
s
s
a
nd
AI
M
L
is
us
e
d
f
or
knowle
dge
of
the
c
ha
tbot
.
C
ha
tbot
us
e
d
f
or
ga
ther
ing
s
of
twa
r
e
r
e
quir
e
ments
us
ing
us
e
r
s
tor
ies
.
2.
1.
Chat
b
ot
d
e
s
ign
I
n
buil
ding
the
Andr
oid
a
ppli
c
a
ti
on
with
the
c
ha
tbot
include
d,
the
objec
t
-
or
iente
d
pr
ogr
a
mm
ing
a
ppr
oa
c
h
is
us
e
d.
UM
L
diagr
a
mm
a
ti
c
notations
a
r
e
us
e
d
to
de
s
ign
the
a
ppli
c
a
ti
on
.
F
igu
r
e
1
s
hows
the
us
e
c
a
s
e
diagr
a
m
of
the
c
ha
tbot
.
T
he
r
e
a
r
e
two
k
ind
s
of
us
e
r
s
,
the
ge
ne
r
ic
us
e
r
,
a
nd
the
s
of
twa
r
e
de
ve
loper
.
T
he
ge
ne
r
ic
us
e
r
s
c
ould
pe
r
f
or
m
two
main
a
c
ti
v
it
ies
whic
h
a
r
e
to
view
the
li
s
t
of
c
ha
t
r
ooms
in
his
/her
a
ppli
c
a
ti
on,
a
nd
a
ls
o
to
give
us
e
r
s
tor
ies
by
c
o
mm
unica
ti
ng
with
the
c
ha
tbot
.
T
he
de
ve
loper
s
c
ould
do
a
nythi
ng
the
ge
ne
r
ic
us
e
r
c
a
n
do,
a
nd
a
ls
o,
he
/s
he
c
ou
ld
a
ls
o
view
the
s
ubmi
tt
e
d
us
e
r
s
tor
ies
a
s
the
r
e
quir
e
ments
ga
ther
e
d
f
or
the
c
or
r
e
s
ponding
a
p
pli
c
a
ti
on.
2.
2.
AI
M
L
Our
r
e
s
e
a
r
c
h
us
e
s
AI
M
L
(
Ar
ti
f
icia
l
I
n
telli
ge
nc
e
M
a
r
kup
L
a
ngua
ge
)
us
e
d
f
or
the
knowle
dge
(
br
a
in
)
of
the
c
ha
tbot
.
I
n
AI
M
L
,
f
or
one
tag
c
a
te
gor
y
us
u
a
ll
y,
i
t
c
ons
is
ts
of
tag
pa
tt
e
r
n
a
nd
tag
template
.
T
a
g
pa
tt
e
r
n
is
us
e
d
f
or
matc
hing
us
e
r
input
a
nd
tag
template
u
s
e
d
f
or
output
f
r
om
the
bot.
F
igu
r
e
2
s
hows
a
n
e
x
a
mpl
e
of
the
AI
M
L
of
the
c
ha
tbot
a
ppli
c
a
ti
on.
T
ha
t
e
xa
mpl
e
s
hows
the
knowle
dge
of
the
c
ha
tbot
f
or
us
e
r
s
to
input
their
r
o
le.
Af
ter
the
us
e
r
input
s
their
r
ole,
the
s
ubmi
tt
e
d
da
ta
wi
ll
be
c
a
ught
by
the
wildca
r
d
s
ymbol
in
tag
pa
tt
e
r
n
a
nd
will
be
s
e
t
in
va
r
iable
temp.
Af
ter
tha
t,
in
o
r
de
r
to
ge
t
a
r
e
s
pons
e
f
r
om
the
c
ha
tbot
,
the
va
lue
of
the
va
r
iable
temp
wi
ll
be
c
he
c
ke
d
.
2.
3.
T
e
xt
p
r
e
-
p
r
oc
e
s
s
in
g
T
e
xt
pr
e
pr
oc
e
s
s
ing
pr
oc
e
s
s
is
done
be
f
or
e
p
a
r
s
ing
the
mes
s
a
ge
to
the
br
a
in
f
i
le.
F
ir
s
tl
y,
nor
maliza
ti
on
is
done
,
whic
h
mes
s
a
ge
will
be
c
onve
r
ted
to
lowe
r
c
a
s
e
.
Af
ter
nor
malizing
the
mes
s
a
ge
,
tokeniz
ing
will
be
done
.
T
oke
nizing
is
a
pr
oc
e
s
s
that
s
pli
t
the
m
e
s
s
a
ge
int
o
tokens
.
E
a
c
h
token
will
be
c
he
c
ke
d
if
ther
e
is
a
mi
s
typed
wor
d
.
P
r
oc
e
s
s
s
top
wor
d
r
e
moval
will
be
done
a
f
ter
c
he
c
king
the
ty
po
wor
d
a
f
ter
that
s
temmi
ng
will
be
done
wi
th
Na
z
ief
&
Adr
iani
s
temmi
ng
a
lgor
it
hm
.
S
temmi
ng
s
tar
ts
by
c
he
c
king
the
c
ombi
na
ti
on
pr
e
f
ix
-
s
uf
f
ix
pr
oc
e
s
s
.
I
f
a
c
om
mon
c
ombi
na
ti
on
de
tec
ted,
the
s
uf
f
ix
is
then
c
h
e
c
ke
d
a
nd
s
temmed,
a
nd
a
f
ter
that
,
the
pr
e
f
ix
will
be
c
he
c
ke
d
a
nd
s
temmed,
a
nd
f
inally
,
a
ne
w
wor
d
wil
l
be
r
e
c
e
ived.
I
f
a
unique
c
ombi
na
ti
on
is
de
tec
ted,
the
pr
e
f
ix
will
be
c
he
c
ke
d,
a
nd
s
tem
a
nd
a
f
ter
that
s
uf
f
ix
will
be
c
he
c
ke
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
2
,
Ap
r
il
2020:
8
90
-
89
8
892
a
nd
s
temmed,
then
a
ga
in,
a
ne
w
wo
r
d
will
be
r
e
c
e
ived.
F
igu
r
e
3
s
hows
the
a
c
ti
vit
y
dia
gr
a
m
f
or
the
pr
e
pr
oc
e
s
s
ing
pr
oc
e
s
s
.
F
igur
e
1.
Us
e
c
a
s
e
diagr
a
m
F
igur
e
2
.
E
xa
mpl
e
of
AI
M
L
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
Us
e
r
s
to
r
ies
c
oll
e
c
ti
on
v
ia
int
e
r
ac
ti
v
e
c
hatbot
to
s
u
ppor
t
r
e
quir
e
me
nt
s
gather
ing
(F
e
r
li
ana
Dw
it
am
)
893
F
igur
e
3
.
Ac
ti
vit
y
d
iagr
a
m
p
r
e
pr
oc
e
s
s
ing
p
r
oc
e
s
s
2.
4.
Us
e
r
s
t
or
y
T
he
us
e
r
s
tor
y
c
onc
e
pt
is
a
pa
r
t
o
f
the
a
gil
e
de
ve
lopm
e
nt
method
that
wa
s
f
ir
s
t
p
r
opos
e
d
in
e
xtr
e
me
pr
ogr
a
mm
ing
f
o
r
ga
ther
ing
us
e
r
r
e
quir
e
ments
th
a
t
a
r
e
ne
e
de
d
by
the
s
ys
tem
[
4]
.
I
t
then
be
c
omes
ke
y
in
the
s
of
twa
r
e
de
ve
lopm
e
nt
p
r
oc
e
s
s
[
23]
.
Us
e
r
s
tor
ies
f
oc
us
on
the
int
e
r
a
c
ti
on
be
twe
e
n
the
us
e
r
or
c
us
tom
e
r
[
3]
,
a
nd
they
will
de
s
c
r
ibe
f
unc
ti
ons
t
ha
t
a
r
e
ne
e
de
d
f
or
the
s
ys
tem
in
the
f
or
m
of
a
s
tor
y
[
4]
.
E
xa
mpl
e
f
or
m
of
the
us
e
r
s
tor
y
c
a
n
be
s
e
e
n
in
F
igu
r
e
4
[
24]
:
a
.
<
r
ole>
r
e
pr
e
s
e
nts
the
r
ole
o
f
the
pe
r
s
on
that
p
r
o
vide
the
s
tor
y.
b.
<
a
c
ti
vit
y>
r
e
pr
e
s
e
nts
the
r
e
quir
e
ment.
c.
<
bus
ines
s
va
lue>
r
e
pr
e
s
e
nts
the
r
e
a
s
on
or
va
lue
of
the
a
bove
r
e
quir
e
ment.
F
igur
e
4
.
Us
e
r
s
tor
y
3.
RE
S
UL
T
S
A
ND
AN
AL
YSI
S
T
he
r
e
s
ult
o
f
the
r
e
s
e
a
r
c
h
is
a
c
ha
tbot
a
ppli
c
a
ti
o
n
that
us
e
d
f
o
r
ga
ther
ing
s
of
twa
r
e
r
e
qui
r
e
ments
us
ing
us
e
r
s
tor
ies
a
nd
Na
z
ief
&
Adr
iani
s
temmi
ng
a
lgor
it
hm
.
I
n
thi
s
c
ha
tbot
AI
M
L
us
e
d
f
or
the
c
ha
tbot
knowle
dge
.
A
us
e
r
a
c
c
e
ptanc
e
tes
t
is
a
ppli
e
d
in
thi
s
r
e
s
e
a
r
c
h
to
e
va
luate
the
c
ha
tbot
's
pe
r
f
o
r
manc
e
,
p
e
r
c
e
ived
by
the
us
e
r
s
.
3.
1.
Chat
b
ot
im
p
lem
e
n
t
at
ion
F
igur
e
5
(
a
)
dis
play
the
s
plas
h
s
c
r
e
e
n
of
the
c
ha
tb
ot
a
ppli
c
a
ti
ons
.
S
plas
h
s
c
r
e
e
n
will
be
s
how
whe
n
us
e
r
ope
n
the
a
ppli
c
a
ti
ons
.
F
igur
e
5
(
b
)
s
hows
the
c
ha
t
r
oom
li
s
t
menu.
T
his
menu
s
how
li
s
t
c
ha
t
r
o
om
that
us
e
r
c
r
e
a
ted.
I
n
thi
s
menu
us
e
r
c
a
n
c
r
e
a
te
or
de
lete
c
ha
t
r
oo
m.
C
ha
t
r
oom
is
r
e
pr
e
s
e
nted
by
a
c
a
r
d.
E
a
c
h
c
a
r
d
of
dis
play
index
o
f
c
a
r
d
,
the
na
me
of
the
a
ppli
c
a
ti
o
n
a
nd
r
ole
of
us
e
r
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
2
,
Ap
r
il
2020:
8
90
-
89
8
894
F
igur
e
6
s
hows
the
us
e
r
s
tor
y
menu.
T
he
us
e
r
s
tor
y
menu
is
f
or
the
de
ve
loper
s
o
they
c
a
n
s
e
e
a
li
s
t
of
us
e
r
s
tor
ies
that
ha
ve
be
e
n
obt
a
ined
f
r
om
the
u
s
e
r
.
E
a
c
h
us
e
r
s
tor
y
r
e
pr
e
s
e
nted
by
a
c
a
r
d
that
e
ve
r
y
c
a
r
d
s
hows
the
s
tor
y,
r
e
a
s
on,
a
ppli
c
a
ti
on
na
me,
a
nd
th
e
da
te
that
the
us
e
r
s
tor
y
obtaine
d.
I
n
e
a
c
h
c
a
r
d
a
ls
o
ha
ve
a
de
lete
butt
on
s
o
the
de
ve
loper
c
a
n
de
lete
the
us
e
r
s
tor
y.
At
the
top
of
the
us
e
r
s
tor
y
,
the
r
e
is
a
f
ield
that
c
a
n
be
us
e
d
by
the
de
ve
loper
to
s
e
a
r
c
h
the
us
e
r
s
tor
y
by
a
ppli
c
a
ti
on
na
me,
s
tor
y,
a
nd
r
e
a
s
on.
T
his
li
s
t
o
f
the
us
e
r
s
tor
y
is
a
lr
e
a
dy
s
or
ted
by
de
s
c
e
nding
s
o
the
late
s
t
da
ta
will
a
lwa
y
s
be
dis
playe
d
in
the
f
i
r
s
t
r
ow,
F
igur
e
7
s
hows
the
c
ha
t
r
ooms
a
nd
a
n
e
xa
mpl
e
of
givi
ng
a
us
e
r
s
tor
y
.
I
n
thi
s
c
ha
t
r
oom,
us
e
r
s
c
a
n
c
ha
t
with
the
bot
a
nd
give
the
us
e
r
s
tor
y.
T
he
f
ir
s
t
-
ti
me
bot
wil
l
a
s
k
the
r
ole
of
the
us
e
r
.
Af
te
r
the
us
e
r
tells
thei
r
bot,
the
r
ole
a
nd
the
bot
wi
ll
then
s
tor
e
the
r
ole
in
f
or
mation
,
a
nd
the
us
e
r
c
a
n
s
tar
t
to
give
the
us
e
r
s
tor
y
.
E
v
e
r
y
ti
me
the
us
e
r
gives
the
us
e
r
the
r
e
a
s
on
the
bot
wi
ll
a
s
k
c
onf
ir
mation
whe
ther
or
not
the
us
e
r
ha
s
given
by
the
us
e
r
c
or
r
e
c
tl
y.
I
f
the
s
tor
y
a
nd
the
r
e
a
s
on
a
lr
e
a
dy
c
or
r
e
c
t,
the
us
e
r
s
tor
y
will
be
a
c
c
e
pted
a
nd
s
a
ve
d
.
3.
2.
Chat
b
ot
m
e
t
r
ics
S
e
ve
r
a
l
metr
ics
we
r
e
us
e
d
to
e
va
luate
the
bot’
s
pe
r
f
or
manc
e
qua
nti
tatively.
T
he
metr
ics
us
e
d
include
tot
a
l
e
laps
e
d
ti
me,
tot
a
l
number
of
us
e
r
tu
r
n,
tot
a
l
number
s
ys
tem
tur
n,
the
tot
a
l
number
o
f
t
ur
ns
pe
r
tas
k,
da
n
tot
a
l
e
laps
e
d
ti
me
pe
r
tur
n.
T
otal
e
laps
e
d
ti
me
c
a
lcula
tes
the
tot
a
l
e
laps
e
d
ti
me
f
r
om
whe
n
the
us
e
r
s
tar
ts
the
c
onve
r
s
a
ti
on
unti
l
whe
n
the
us
e
r
f
ini
s
he
s
s
ubmi
t
a
us
e
r
s
tor
y.
T
he
tot
a
l
number
of
u
s
e
r
tur
n
c
a
lcula
tes
the
number
of
us
e
r
c
ha
ts
f
r
om
whe
n
the
us
e
r
s
tar
ts
the
c
onve
r
s
a
ti
on
unti
l
whe
n
the
us
e
r
f
ini
s
he
s
s
ubmi
t
a
us
e
r
s
tor
y.
T
he
tot
a
l
number
of
s
ys
tem’
s
tur
n
c
a
lcula
tes
the
number
of
bot
r
e
s
pons
e
s
f
r
o
m
whe
n
the
us
e
r
s
tar
ts
the
c
onve
r
s
a
ti
on
unti
l
whe
n
the
us
e
r
f
ini
s
he
s
s
ubmi
t
a
us
e
r
s
tor
y.
T
otal
e
laps
e
d
ti
me
pe
r
tur
n
c
a
lcula
tes
the
e
laps
e
d
ti
me
f
or
a
us
e
r
to
f
ini
s
h
his
/
he
r
tur
n;
thi
s
is
c
a
lcula
ted
by
divi
ding
the
tot
a
l
e
laps
e
d
ti
me
by
the
tot
a
l
number
of
tur
ns
.
T
he
metr
ics
e
va
luate
the
pr
ovis
ion
of
30
us
e
r
s
tor
ies
given
by
thr
e
e
us
e
r
s
in
whic
h
the
ti
me
is
mea
s
ur
e
d
manua
ll
y
us
ing
a
ti
mer
.
T
a
ble
1
s
hows
the
s
a
mpl
e
metr
ic
of
a
us
e
r
whe
n
tes
ti
ng
the
c
ha
tbot
a
ppli
c
a
ti
on,
pe
r
f
o
r
mi
ng
a
tot
a
l
o
f
8
t
a
s
ks
.
E
a
c
h
tas
k
in
T
a
ble
1
r
e
pr
e
s
e
nts
a
n
a
c
ti
vit
y
in
whic
h
a
us
e
r
is
c
omm
unica
ti
ng
us
ing
the
c
ha
tbot
to
p
r
ov
ide
us
e
r
s
tor
ies
,
with
the
f
ol
lowing
number
s
a
r
e
c
a
lcula
ted
f
or
the
metr
ics
.
(
a
)
(
b)
F
igur
e
5
.
(
a
)
S
plas
hs
c
r
e
e
n
(
b)
C
ha
t
r
oom
li
s
t
menu
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
Us
e
r
s
to
r
ies
c
oll
e
c
ti
on
v
ia
int
e
r
ac
ti
v
e
c
hatbot
to
s
u
ppor
t
r
e
quir
e
me
nt
s
gather
ing
(F
e
r
li
ana
Dw
it
am
)
895
F
igur
e
6
.
L
is
t
of
us
e
r
s
tor
ies
F
igur
e
7
.
C
ha
t
r
oom
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
2
,
Ap
r
il
2020:
8
90
-
89
8
896
T
a
ble
1.
C
ha
tbot
metr
ics
T
ot
a
l
e
la
ps
e
d t
im
e
(
s
e
c
onds
)
T
ot
a
l
numbe
r
of
us
e
r
t
ur
n
T
ot
a
l
numbe
r
of
s
ys
te
m t
ur
n
T
ot
a
l
tu
r
ns
pe
r
ta
s
k
T
ot
a
l
e
la
ps
e
d t
im
e
pe
r
tu
r
n (
s
e
c
onds
)
T
a
s
k 1
50
.
70
6
6
12
4
.
23
T
a
s
k 2
37
.
18
7
7
14
2
.
66
T
a
s
k 3
36
.
14
6
6
12
3
.
01
T
a
s
k 4
32
.
00
6
6
12
2
.
67
T
a
s
k 5
39
.
24
6
6
12
3
.
27
T
a
s
k 6
44
.
57
6
6
12
3
.
71
T
a
s
k 7
44
.
44
7
7
14
3
.
17
T
a
s
k 8
48
.
68
6
6
12
4
.
06
T
ot
a
l
332
.
95
50
50
100
26
.
77
A
ve
r
a
ge
41
.
62
6
6
12
3
.
35
3.
3.
P
r
e
li
m
in
ar
y
u
s
e
r
ac
c
e
p
t
an
c
e
t
e
s
t
A
pr
e
li
mi
na
r
y
us
e
r
a
c
c
e
ptanc
e
tes
t
wa
s
c
onduc
te
d
to
e
va
luate
the
pe
r
c
e
ived
us
e
f
ulnes
s
(
P
U)
,
us
e
r
a
tt
it
ude
towa
r
ds
(
A)
,
be
ha
vior
a
l
int
e
nti
on
to
u
s
e
(
B
I
)
,
a
nd
pe
r
c
e
ived
e
a
s
e
of
us
e
(
P
E
OU
)
,
f
oll
owing
the
tec
hnology
a
c
c
e
ptanc
e
model
pr
opos
e
d
by
D
a
vis
[
25]
.
T
h
e
que
s
ti
onna
ir
e
us
e
s
a
f
ive
-
point
L
ik
e
r
t
s
c
a
le
us
e
d
in
or
de
r
to
de
ter
mi
ne
the
leve
l
of
us
e
r
a
c
c
e
ptanc
e
,
whic
h
is
c
ompr
is
e
d
of
s
tr
ongly
a
g
r
e
e
(
5)
,
a
gr
e
e
(
4)
,
ne
utr
a
l
(
3)
,
dis
a
gr
e
e
(
2)
,
a
nd
s
tr
ongly
dis
a
gr
e
e
(
1)
.
T
hir
ty
-
thr
e
e
r
e
s
ponde
nts
pa
r
ti
c
ipate
d
in
thi
s
que
s
ti
onna
ir
e
.
I
n
or
de
r
to
tes
t
the
c
a
pa
bil
it
ies
of
the
c
ha
tbot
,
we
us
e
a
univer
s
it
y
inf
or
mation
s
ys
tem,
a
s
the
c
a
s
e
s
tudy
f
or
whic
h
the
us
e
r
s
will
p
r
ovide
f
e
e
dba
c
k.
Us
e
r
a
c
c
e
ptanc
e
tes
t
is
done
to
33
us
e
r
s
us
ing
a
we
b
-
ba
s
e
d
univer
s
it
y
inf
or
mation
s
ys
tem,
c
a
ll
e
d
M
yUM
N,
a
s
a
tes
t
c
a
s
e
f
or
whic
h
the
us
e
r
s
will
p
r
ovide
us
e
r
s
tor
ies
on
.
T
he
r
e
s
ponde
nt
is
M
yUM
N’
s
us
e
r
s
whic
h
a
r
e
s
tudent,
lec
tur
e
r
o
r
unive
r
s
it
y
s
taf
f
s
.
T
he
r
e
s
ponde
nt’
s
a
ge
is
a
bove
20
ye
a
r
s
old
,
both
male
a
nd
f
e
male
r
e
s
ponde
nt
s
.
F
ir
s
tl
y,
r
e
s
pond
e
nts
we
r
e
a
s
k
e
d
to
downloa
d
the
APK
f
il
e
f
or
the
c
ha
tbot
a
ppli
c
a
ti
on.
T
he
n,
us
e
r
s
e
xpe
r
im
e
nted
with
the
a
pp,
ba
s
e
d
on
t
he
a
ppli
c
a
ti
on
de
s
c
r
ipt
ion
e
xplaining
a
bout
us
e
r
s
t
or
y
a
nd
wha
t
the
c
ha
tbot
a
im
s
to
pe
r
f
o
r
m.
Af
ter
the
us
e
r
s
tr
ied
a
nd
e
xpe
r
im
e
nted
with
giv
ing
f
e
e
dba
c
ks
r
e
ga
r
ding
the
M
yUM
N
we
b
a
ppli
c
a
ti
on
a
nd
c
omm
unica
ti
n
g
with
the
c
ha
tbot
,
que
s
ti
onna
ir
e
s
we
r
e
dis
tr
ibut
e
d,
a
nd
int
e
r
view
s
we
r
e
c
onduc
ted
to
ge
t
f
e
e
dba
c
k
f
r
om
the
us
e
r
s
,
mainly
f
oc
us
ing
on
the
f
ou
r
a
s
pe
c
ts
of
the
tec
hnology
a
c
c
e
ptanc
e
model
mentioned
pr
e
vious
ly.
T
he
que
s
ti
onna
ir
e
s
’
a
ns
we
r
s
he
e
t
is
f
or
med
ba
s
e
d
on
the
int
e
r
va
ls
us
ing
the
L
iker
t
S
c
a
le,
with
int
e
r
p
r
e
tation,
a
s
s
hown
in
T
a
ble
2
be
low.
B
a
s
e
d
on
the
que
s
ti
onna
ir
e
s
f
il
led
by
the
us
e
r
s
a
f
ter
us
ing
the
c
ha
tbot
a
ppli
c
a
ti
on
,
we
c
a
lcula
ted
the
a
ve
r
a
ge
s
c
or
e
f
r
om
the
que
s
ti
onna
ir
e
f
o
r
e
a
c
h
f
a
c
tor
in
the
tec
hnology
a
c
c
e
ptanc
e
model.
R
e
s
ul
ts
a
r
e
a
s
f
oll
ow,
83
.
79%
f
or
us
e
r
a
tt
it
ude
towa
r
ds
the
c
ha
tbot
a
ppli
c
a
ti
on,
85
.
45%
f
or
pe
r
c
e
ived
us
e
f
ulnes
s
,
84.
55%
f
or
pe
r
c
e
ived
e
a
s
e
of
us
e
,
a
nd
83.
03
%
f
or
us
e
r
s
’
be
ha
vio
r
a
l
int
e
nti
on
to
us
e
.
T
he
g
r
a
phic
a
l
r
e
s
ult
of
us
e
r
a
c
c
e
ptanc
e
tes
t
c
a
n
be
s
e
e
n
in
F
igur
e
8.
I
nter
iew
s
a
ls
o
s
hows
that
the
us
e
r
s
’
a
s
s
take
ho
lder
s
of
the
M
yUM
N
we
b
a
ppli
c
a
ti
on.
T
he
ove
r
a
ll
r
e
s
ult
s
s
e
e
ms
not
to
dif
f
e
r
f
r
om
one
a
s
pe
c
t
to
a
nother
,
ho
we
ve
r
if
we
look
c
los
e
ly,
pe
r
c
e
ived
us
e
f
ulnes
s
a
c
hieve
d
th
e
highes
t
s
c
or
e
,
thi
s
a
s
pe
c
t
mea
s
ur
e
s
the
u
s
e
r
s
’
a
c
c
e
ptanc
e
r
e
ga
r
ding
how
we
ll
the
c
ha
tbot
is
pe
r
c
e
ived
to
be
us
e
f
ul
to
s
uppor
t
the
r
e
quir
e
ments
ga
ther
ing
a
c
ti
vit
y
(
f
r
om
the
de
ve
loper
s
po
int
of
view
)
a
nd
the
f
e
e
dba
c
k
pr
ovis
ion
a
c
ti
vit
y
(
f
r
om
the
us
e
r
s
’
pe
r
s
pe
c
ti
ve
)
by
pr
ovidi
ng
a
pos
s
ibi
li
ty
o
f
a
na
tur
a
l
c
omm
unica
ti
on
to
s
tr
uc
tur
e
the
f
e
e
dba
c
ks
f
oll
owing
the
us
e
r
s
tor
y
f
o
r
mat,
e
li
c
it
ing
the
f
e
e
dba
c
k
pr
ovider
s
’
r
ole
,
their
ne
e
ds
,
a
nd
the
bus
ines
s
va
lue
r
e
ga
r
ding
the
ne
e
ds
/r
e
quir
e
ments
pr
ovided.
F
igur
e
8.
Us
e
r
a
c
c
e
ptanc
e
r
e
s
ult
8
3
,
7
9
%
8
5
,
4
5
%
8
4
,
5
5
%
8
3
,
0
3
%
8
1
,
0
0
%
8
2
,
0
0
%
8
3
,
0
0
%
8
4
,
0
0
%
8
5
,
0
0
%
8
6
,
0
0
%
A
PU
PE
O
U
BI
Us
e
r
A
c
c
e
pt
a
n
c
e
T
e
s
t
Evaluation Warning : The document was created with Spire.PDF for Python.
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
Us
e
r
s
to
r
ies
c
oll
e
c
ti
on
v
ia
int
e
r
ac
ti
v
e
c
hatbot
to
s
u
ppor
t
r
e
quir
e
me
nt
s
gather
ing
(F
e
r
li
ana
Dw
it
am
)
897
T
he
be
ha
vior
a
l
int
e
nti
on
to
us
e
s
c
or
e
d
the
lo
we
s
t
of
the
f
our
a
s
pe
c
ts
,
howe
ve
r
,
looki
ng
a
t
the
int
e
r
pr
e
tation
s
c
or
e
s
hown
in
T
a
ble
2,
it
s
ti
ll
mana
ge
d
to
be
a
ble
to
be
int
e
r
pr
e
ted
a
s
S
tr
ongl
y
Agr
e
e
.
I
nter
view
s
c
onduc
ted
a
f
ter
the
que
s
ti
onna
ir
e
s
-
f
il
li
ng
a
c
ti
vit
y
s
hows
the
us
e
r
s
’
mot
ivat
ion
a
nd
int
e
nti
on
to
us
e
the
c
ha
tbot
a
ga
in
in
the
f
utu
r
e
.
T
a
ble
2.
Us
e
r
a
c
c
e
ptanc
e
tes
t
s
c
or
e
int
e
r
pr
e
tation
I
nt
e
r
pr
e
ta
ti
on
P
e
r
c
e
nt
a
ge
S
c
or
e
V
a
lu
e
S
tr
ongl
y D
is
a
gr
e
e
0%
-
19
.
99
%
1
D
is
a
gr
e
e
2
0
%
-
39
.
99
%
2
Ne
u
tr
a
l
4
0
%
-
59
.
99
%
3
A
gr
e
e
6
0
%
-
79
.
99
%
4
S
tr
ongl
y A
gr
e
e
8
0
%
-
100%
5
4.
CONC
L
USI
ON
T
he
c
onduc
ted
r
e
s
e
a
r
c
h
ha
s
de
ve
loped
a
c
ha
tbot
a
ppli
c
a
ti
on
ba
s
e
d
on
the
Andr
oid
mobi
le
platf
or
m
,
to
s
uppor
t
the
r
e
quir
e
ments
ga
ther
ing
a
c
ti
vit
y
by
pr
ovidi
ng
wa
ys
to
br
idge
the
c
omm
unica
ti
on
ga
p
be
twe
e
n
the
de
ve
loper
s
of
a
s
of
twa
r
e
pr
oduc
t
a
nd
their
s
take
ho
lder
s
.
T
he
r
e
qui
r
e
ments
that
a
r
e
ga
ther
e
d
a
r
e
then
na
tur
a
ll
y
e
nf
or
c
e
d
by
the
c
ha
tbot
to
f
oll
ow
the
us
e
r
s
tor
y
f
or
mat
c
ontaining
the
us
e
r
s
’
r
ole,
r
e
quir
e
ments
/nee
ds
,
a
nd
it
s
bus
ines
s
va
lue.
T
he
c
ha
tbot
a
ls
o
im
pleme
nted
the
Na
z
ief
&
Adr
iani
s
t
e
mm
ing
a
lgor
it
h
m
f
or
the
s
pe
c
if
ic
ne
e
d
of
s
temmi
ng
wor
ds
wr
it
ten
in
B
a
ha
s
a
I
ndone
s
ia.
F
ur
ther
mor
e
,
it
uti
li
z
e
s
the
Ar
ti
f
icia
l
I
ntelli
ge
nc
e
M
a
r
kup
L
a
ngua
ge
f
or
the
bot’
s
knowle
dge
ba
s
e
in
whic
h
the
bot
r
e
s
pons
e
is
de
c
ided,
given
a
us
e
r
c
ha
t
ha
s
be
e
n
s
ubmi
tt
e
d.
A
pr
e
li
mi
n
a
r
y
us
e
r
a
c
c
e
ptanc
e
tes
t
ba
s
e
d
on
the
tec
hnology
a
c
c
e
ptanc
e
model
s
how
s
pr
omi
s
ing
r
e
s
ult
s
.
T
he
t
e
s
t
f
oc
us
e
d
on
e
va
luating
f
our
a
s
pe
c
ts
of
us
e
r
a
c
c
e
ptanc
e
whic
h
a
r
e
the
us
e
r
a
tt
it
ude
towa
r
d
the
c
ha
tbot
a
p
pli
c
a
ti
on,
pe
r
c
e
ived
us
e
f
ulnes
s
,
pe
r
c
e
ived
e
a
s
e
of
us
e
,
a
nd
the
us
e
r
s
’
be
ha
vior
a
l
int
e
nti
on
to
us
e
.
T
he
c
ha
tbot
r
e
c
e
ived
s
c
or
e
s
f
or
the
f
ou
r
a
s
pe
c
ts
a
s
f
oll
ow,
83.
79%
f
o
r
us
e
r
a
tt
it
ude
towa
r
ds
the
c
ha
tbot
a
ppli
c
a
ti
on,
85.
4
5%
f
o
r
pe
r
c
e
ived
us
e
f
ulnes
s
,
84.
55%
f
or
pe
r
c
e
ived
e
a
s
e
of
us
e
,
a
nd
83.
03%
f
or
us
e
r
s
’
be
ha
vior
a
l
int
e
nti
on
t
o
us
e
.
F
utur
e
wo
r
ks
include
e
mpl
oying
a
mor
e
a
dva
nc
e
d
na
tur
a
l
langua
ge
pr
oc
e
s
s
ing
method
f
or
f
e
e
db
a
c
ks
wr
it
ten
in
B
a
ha
s
a
I
ndone
s
ia,
de
ve
lopi
ng
the
we
b
da
s
hboa
r
d
f
or
the
s
upe
r
us
e
r
(
de
ve
loper
s
)
to
mana
ge
the
AI
M
L
br
a
in
f
il
e
a
nd
mana
ge
s
ub
mi
tt
e
d
f
e
e
dba
c
k
by
f
il
ter
ing
a
nd
c
las
s
if
ying
them
int
o
mea
ningf
ul
inf
or
mation
f
or
the
s
of
twa
r
e
e
nginee
r
s
.
F
u
r
ther
mo
r
e
,
a
mo
r
e
thor
ough
us
e
r
a
c
c
e
ptanc
e
tes
t
c
ould
a
ls
o
be
c
onduc
ted
to
f
ur
ther
e
va
luate
the
c
ha
tbot
’
s
pe
r
f
o
r
manc
e
a
nd
i
ts
us
e
r
s
’
pe
r
s
pe
c
ti
ve
tow
a
r
d
the
c
ha
tbot
.
AC
KNOWL
E
DGE
M
E
NT
T
his
r
e
s
e
a
r
c
h
wa
s
s
uppor
ted
by
the
M
obil
e
De
v
e
lopm
e
nt
L
a
bor
a
tor
y
in
Unive
r
s
it
a
s
M
ult
im
e
dia
Nus
a
ntar
a
.
W
e
a
ls
o
thank
our
c
oll
e
a
gue
s
f
r
om
t
he
F
a
c
ult
y
o
f
E
nginee
r
ing
a
nd
I
nf
o
r
matics
who
pr
ovided
ins
ight
a
nd
e
xpe
r
ti
s
e
that
g
r
e
a
tl
y
a
s
s
is
ted
the
r
e
s
e
a
r
c
h,
a
lt
hough
they
may
not
a
g
r
e
e
with
a
ll
o
f
the
int
e
r
pr
e
tations
/conc
lus
ions
of
thi
s
pa
pe
r.
RE
F
E
RE
NC
E
S
[1
]
Pres
s
ma
n
RS,
“
So
ft
w
are
E
n
g
i
n
eer
i
n
g
A
Pract
i
t
i
o
n
er’s
A
p
p
ro
ac
h
7
th
E
d
i
t
i
o
n
,
”
S
o
f
t
w
a
r
e
E
n
g
i
n
e
er
i
n
g
A
P
r
a
ct
i
t
i
o
n
er
’
s
A
p
p
r
o
a
c
h
7
th
Ed
-
R
o
g
e
r
S
.
P
r
e
s
s
m
a
n
.
2
0
1
0
.
[2
]
So
mmerv
i
l
l
e
I
,
“So
ft
w
are
E
n
g
i
n
eer
i
n
g
,
”
S
o
f
t
w
a
r
e
E
n
g
i
n
eer
i
n
g
.
2
0
1
0
.
[3
]
L
ams
w
eerd
e
A
V
an
,
“
Req
u
i
reme
n
t
s
E
n
g
i
n
eeri
n
g
:
Fro
m
Sy
s
t
em
G
o
al
s
t
o
U
ML
Mo
d
e
l
s
t
o
So
f
t
w
are
Sp
ec
i
fi
ca
t
i
o
n
s
,”
Ch
a
n
g
e
,
2
0
1
0
.
[4
]
O
’h
E
o
c
h
a
C,
Co
n
b
o
y
K
,
“
T
h
e
ro
l
e
o
f
t
h
e
u
s
er
s
t
o
r
y
ag
i
l
e
p
ract
i
ce
i
n
i
n
n
o
v
a
t
i
o
n
,”
Lect
u
r
e
N
o
t
e
s
i
n
B
u
s
i
n
es
s
In
f
o
r
m
a
t
i
o
n
P
r
o
ce
s
s
i
n
g
,
p
p
.
2
0
-
3
0
,
2
0
1
0
.
[5
]
A
b
d
u
l
-
K
a
d
er
SA
,
J
o
h
n
D
,
”
Su
r
v
ey
o
n
Ch
a
t
b
o
t
D
e
s
i
g
n
T
ech
n
i
q
u
es
i
n
Sp
eec
h
Co
n
v
ers
a
t
i
o
n
S
y
s
t
ems
,
”
In
t
J
A
d
v
Co
m
p
u
t
S
ci
A
p
p
l
.
2
0
1
5
.
[6
]
H
u
a
n
g
J
,
Z
h
o
u
M,
Y
a
n
g
D
,
“
E
x
t
ract
i
n
g
ch
at
b
o
t
k
n
o
w
l
ed
g
e
fro
m
o
n
l
i
n
e
d
i
s
c
u
s
s
i
o
n
f
o
ru
m
s
,
”
IJCA
I
In
t
er
n
a
t
i
o
n
a
l
Jo
i
n
t
Co
n
f
e
r
en
ce
o
n
A
r
t
i
f
i
ci
a
l
I
n
t
e
l
l
i
g
e
n
ce
,
p
p
.
6
-
1
2
,
2
0
0
7
.
[7
]
Sh
aw
ar
B,
A
t
w
el
l
E
,
“
U
s
i
n
g
d
i
al
o
g
u
e
co
r
p
o
ra
t
o
t
ra
i
n
a
ch
at
b
o
t
,
”
P
r
o
c
Co
r
p
u
s
Li
n
g
u
i
s
t
,
p
p
.
6
8
1
-
6
9
0
,
2
0
0
3
.
[8
]
O
racl
e,
“
Can
V
i
r
t
u
a
l
E
x
p
eri
e
n
ces
Rep
l
ace
Real
i
t
y
?
T
h
e
fu
t
u
re
ro
l
e
fo
r
h
u
man
s
i
n
d
e
l
i
v
eri
n
g
cu
s
t
o
mer
ex
p
er
i
en
ce,
”
O
r
a
cl
e
,
p
p
.
1
-
1
9
,
2
0
1
6
.
[9
]
N
g
u
y
e
n
M
-
H
,
“
Fro
m
Fo
rt
u
n
e
5
0
0
s
t
o
s
mal
l
b
u
s
i
n
es
s
es
,
real
b
u
s
i
n
e
s
s
e
s
are
al
read
y
u
s
i
n
g
ch
a
t
b
o
t
s
t
o
i
mp
r
o
v
e
t
h
ei
r
s
erv
i
ce,
”
2
0
1
7
.
[1
0
]
Beav
er
L
,
“
Ch
at
b
o
t
s
are
g
a
i
n
i
n
g
t
ract
i
o
n
In
t
er
n
et
,
”
2
0
1
7
.
[
O
n
l
i
n
e
].
A
v
a
i
l
a
b
l
e
:
h
t
t
p
s
:
/
/
w
w
w
.
b
u
s
i
n
e
s
s
i
n
s
i
d
er.
c
o
m/
ch
at
b
o
t
s
-
are
-
g
ai
n
i
n
g
-
t
ract
i
o
n
-
2
0
1
7
-
5
/
?IR=
T
.
A
cces
s
e
d
:
2
O
ct
o
b
er
2
0
1
8
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
S
S
N
:
1693
-
6930
T
E
L
KO
M
NI
KA
T
e
lec
omm
un
C
omput
E
l
C
ontr
o
l
,
Vol.
18
,
No
.
2
,
Ap
r
il
2020:
8
90
-
89
8
898
[1
1
]
V
an
E
e
u
w
e
n
M,
“
Mo
b
i
l
e
co
n
v
ers
a
t
i
o
n
a
l
co
mmerce
:
mes
s
e
n
g
er
ch
a
t
b
o
t
s
as
t
h
e
n
ex
t
i
n
t
erface
b
e
t
w
ee
n
b
u
s
i
n
e
s
s
e
s
an
d
co
n
s
u
mers
,
”
U
n
i
v
Twe
n
t
e
.
2
0
1
7
.
[1
2
]
D
at
a
b
o
k
s
.
co
.
i
d
,
“
Pen
g
g
u
n
a
Po
n
s
e
l
In
d
o
n
es
i
a
Men
ca
p
a
i
1
4
2
%
d
ar
i
Po
p
u
l
as
i
[In
t
ern
e
t
],
”
2
0
1
7
.
[
O
n
l
i
n
e
].
A
v
ai
l
a
b
l
e
:
h
t
t
p
s
:
/
/
d
at
a
b
o
k
s
.
k
a
t
ad
a
t
a.
co
.
i
d
/
d
at
a
p
u
b
l
i
s
h
/
2
0
1
7
/
0
8
/
2
9
/
p
en
g
g
u
n
a
-
p
o
n
s
el
-
i
n
d
o
n
es
i
a
-
men
ca
p
ai
-
1
4
2
-
d
ar
i
-
p
o
p
u
l
a
s
i
.
A
cces
s
ed
:
2
O
c
t
o
b
er
2
0
1
8
.
[1
3
]
Mari
et
t
o
B,
A
g
u
i
ar
M
d
a
s
G
,
A
g
u
i
as
R
V
,
“
A
rt
i
fi
c
i
al
In
t
el
l
i
g
en
ce
Mar
k
u
p
L
an
g
u
ag
e
:
A
Bri
ef
T
u
t
o
ri
a
l
,
”
I
n
t
J
Co
m
p
u
t
S
ci
E
n
g
S
u
r
v
.
2
0
1
3
.
[1
4
]
W
al
l
ace
R,
“
T
h
e
el
eme
n
t
s
o
f
A
IML
s
t
y
l
e,
”
A
l
i
ce
A
I
F
o
u
n
d
.
2
0
0
3
.
[1
5
]
Mard
i
a
n
a
T
,
A
d
j
i
T
B,
H
i
d
ay
a
h
I,
“
St
emmi
n
g
i
n
f
l
u
e
n
c
e
o
n
s
i
mi
l
ari
t
y
d
et
ec
t
i
o
n
o
f
ab
s
t
rac
t
w
ri
t
t
en
i
n
In
d
o
n
es
i
a,
”
TE
LKO
M
NIK
A
Tel
ec
o
m
m
u
n
i
c
a
t
i
o
n
Co
m
p
u
t
i
n
g
E
l
ec
t
r
o
n
i
c
s
a
n
d
Co
n
t
r
o
l
,
v
o
l
.
1
4
,
n
o
.
1
,
p
p
.
2
1
9
-
2
2
7
,
2
0
1
6
.
[1
6
]
A
g
u
s
t
a
L
,
“
Co
mp
ari
s
o
n
o
f
P
o
rt
er'
s
St
emmi
n
g
A
l
g
o
r
i
t
h
m
w
i
t
h
N
az
i
ef
&
A
d
r
i
an
i
's
A
l
g
o
r
i
t
h
m
fo
r
St
emm
i
n
g
In
d
o
n
e
s
i
a
n
T
e
x
t
D
o
c
u
men
t
s
(
i
n
Bah
as
a
:
Perb
a
n
d
i
n
g
an
A
l
g
o
ri
t
ma
St
emm
i
n
g
Po
r
t
er
D
e
n
g
a
n
A
l
g
o
ri
t
ma
N
az
i
e
f
&
A
d
r
i
an
i
U
n
t
u
k
St
em
m
i
n
g
D
o
k
u
me
n
T
e
k
s
Bah
a
s
a
In
d
o
n
e
s
i
a
)
,
”
Ko
n
f
Na
s
S
i
s
t
d
a
n
In
f
o
r
m
,
2
0
0
9
.
[1
7
]
H
i
d
ay
a
t
u
l
l
a
h
A
F,
Rat
n
as
ar
i
CI,
W
i
s
n
u
g
r
o
h
o
S,
“
A
n
a
l
y
s
i
s
o
f
St
emm
i
n
g
In
fl
u
en
ce
o
n
I
n
d
o
n
es
i
an
T
w
ee
t
Cl
as
s
i
fi
cat
i
o
n
,
”
TE
LKO
M
NIK
A
Tel
eco
m
m
u
n
i
ca
t
i
o
n
Co
m
p
u
t
i
n
g
E
l
ec
t
r
o
n
i
cs
a
n
d
Co
n
t
r
o
l
,
v
o
l
.
1
4
,
n
o
.
2
,
p
p
.
6
6
5
-
6
7
3
,
2
0
1
6
.
[1
8
]
W
at
e
q
u
l
i
s
Sy
ai
fu
d
i
n
Y
,
Sap
u
t
ra
PY
,
Pu
s
p
i
t
as
ar
i
D
,
“
T
h
e
i
mp
l
emen
t
at
i
o
n
o
f
w
eb
s
er
v
i
ce
b
as
e
d
t
ex
t
p
rep
r
o
ces
s
i
n
g
t
o
meas
u
re
In
d
o
n
es
i
an
s
t
u
d
e
n
t
t
h
es
i
s
s
i
m
i
l
ar
i
t
y
l
ev
e
l
,
”
M
A
TE
C
W
eb
o
f
Co
n
f
e
r
en
ce
s
,
2
0
1
8
.
[1
9
]
A
u
l
i
a
A
,
K
h
ai
ran
i
D
,
H
ak
i
em
N
,
“
D
ev
e
l
o
p
men
t
o
f
a
re
t
ri
e
v
al
s
y
s
t
em
fo
r
A
l
H
a
d
i
t
h
i
n
Ba
h
as
a
(ca
s
e
s
t
u
d
y
:
H
a
d
i
t
h
Bu
k
h
ari
)
,”
2
0
1
7
5
th
In
t
e
r
n
a
t
i
o
n
a
l
Co
n
f
er
en
ce
o
n
Cyb
e
r
a
n
d
IT
S
er
v
i
ce
M
a
n
a
g
em
e
n
t
,
CIT
SM
2
0
1
7
.
[2
0
]
Pu
rb
a
s
ari
,
I.
Y
.
,
A
n
g
g
rae
n
y
,
F.
T
.
,
Mah
ara
n
i
,
M.
F.
,
“
Rep
l
i
es
Id
en
t
i
f
i
cat
i
o
n
i
n
Q
u
e
s
t
i
o
n
A
n
s
w
eri
n
g
S
y
s
t
em
u
s
i
n
g
V
ect
o
r
Sp
ace
Mo
d
el
,
”
A
t
l
a
n
t
i
s
H
i
g
h
l
i
g
h
t
s
i
n
E
n
g
i
n
ee
r
i
n
g
(A
H
E
)
,
v
o
l
.
1
,
p
p
.
4
5
9
-
4
6
4
,
2
0
1
8
.
[2
1
]
T
o
x
t
l
i
C,
Mo
n
ro
y
-
H
er
n
án
d
ez
A
,
Cran
s
h
aw
J
,
“
U
n
d
ers
t
a
n
d
i
n
g
c
h
at
b
o
t
-
med
i
at
e
d
t
as
k
man
ag
eme
n
t
,
”
Co
n
f
er
en
ce
o
n
H
u
m
a
n
F
a
c
t
o
r
s
i
n
Co
m
p
u
t
i
n
g
S
y
s
t
e
m
s
–
P
r
o
ceed
i
n
g
s
,
2
0
1
8
.
[2
2
]
T
ed
j
o
p
ran
o
t
o
,
M.
L
.
,
W
i
j
ay
a,
A
.
,
San
t
o
s
o
,
L
.
H
.
,
Su
h
art
o
n
o
,
D
,
“
Co
rrec
t
i
n
g
T
y
p
o
g
rap
h
i
ca
l
E
rr
o
r
a
n
d
U
n
d
ers
t
an
d
i
n
g
U
s
er
I
n
t
e
n
t
i
o
n
i
n
Ch
a
t
b
o
t
b
y
C
o
mb
i
n
i
n
g
N
-
G
ram
an
d
Mach
i
n
e
L
earn
i
n
g
U
s
i
n
g
Sc
h
ema
Ma
t
c
h
i
n
g
T
ech
n
i
q
u
e,
”
In
t
er
n
a
t
i
o
n
a
l
Jo
u
r
n
a
l
o
f
M
a
c
h
i
n
e
Lea
r
n
i
n
g
a
n
d
Co
m
p
u
t
i
n
g
,
v
o
l
.
9
,
n
o
.
4
,
p
p
.
4
7
1
-
4
7
6
,
2
0
1
9
.
[2
3
]
Rees
M
.
J
.
,
“
A
feas
i
b
l
e
u
s
er
s
t
o
ry
t
o
o
l
f
o
r
a
g
i
l
e
s
o
f
t
w
are
d
ev
e
l
o
p
men
t
?,
”
P
r
o
cee
d
i
n
g
s
-
A
s
i
a
-
P
a
ci
f
i
c
S
o
f
t
w
a
r
e
E
n
g
i
n
eer
i
n
g
Co
n
f
er
e
n
ce,
A
P
S
E
C
,
2
0
0
2
.
[2
4
]
Z
eaarao
u
i
A
.
,
Bo
u
g
ro
u
n
Z
.
,
Bel
k
as
m
i
M
.
G
.
,
Bo
u
ch
en
t
o
u
f
T
.
,
“
U
s
er
s
t
o
ri
e
s
t
em
p
l
a
t
e
fo
r
o
b
j
ect
-
o
ri
e
n
t
e
d
ap
p
l
i
ca
t
i
o
n
s
,
”
In
:
2
0
1
3
3
r
d
In
t
er
n
a
t
i
o
n
a
l
C
o
n
f
er
e
n
ce
o
n
In
n
o
va
t
i
ve
Co
m
p
u
t
i
n
g
Tech
n
o
l
o
g
y
,
IN
T
E
CH
2
0
1
3
.
[2
5
]
Da
v
i
s
F
.
D
.
,
Bag
o
zzi
R
.
P
.
,
W
ars
h
aw
P
.
R
.
,
“
U
s
er
A
ccep
t
a
n
ce
o
f
Co
mp
u
t
er
T
ech
n
o
l
o
g
y
:
A
Co
mp
a
ri
s
o
n
o
f
T
w
o
T
h
e
o
ret
i
cal
Mo
d
el
s
,
”
M
a
n
a
g
e
S
c
i
,
1
9
8
9
.
B
I
OG
RA
P
H
I
E
S
OF
AU
T
HO
RS
F
erl
i
a
n
a
D
w
i
ta
m
a
i
s
g
rad
u
at
ed
fro
m
U
n
i
v
ers
i
t
a
s
Mu
l
t
i
me
d
i
a
N
u
s
a
n
t
ara
i
n
N
o
v
em
b
er
201
9
w
i
t
h
a
b
ach
e
l
o
r’
s
o
f
Co
mp
u
t
e
r
S
ci
en
ce.
Sh
e
h
as
b
ack
g
ro
u
n
d
i
n
i
n
f
o
rmat
i
o
n
t
ech
n
o
l
o
g
y
an
d
h
o
l
d
s
k
ee
n
i
n
t
ere
s
t
s
i
n
t
h
e
area
o
f
re
q
u
i
r
emen
t
s
en
g
i
n
eer
i
n
g
an
d
ch
at
b
o
t
.
A
ndre
R
us
l
i
r
ecei
v
ed
h
i
s
Mas
t
er’
s
o
f
Sci
e
n
ce
d
e
g
ree
i
n
In
fo
rma
t
i
o
n
E
n
v
i
r
o
n
me
n
t
fro
m
T
o
k
y
o
D
e
n
k
i
U
n
i
v
ers
i
t
y
,
J
ap
a
n
,
i
n
2
0
1
7
,
fo
cu
s
i
n
g
i
n
t
h
e
So
f
t
w
are
Re
q
u
i
remen
t
s
E
n
g
i
n
eeri
n
g
f
i
el
d
.
H
e
i
s
cu
rre
n
t
l
y
a
l
ect
u
rer
a
n
d
re
s
earch
er
i
n
U
n
i
v
ers
i
t
a
s
M
u
l
t
i
me
d
i
a
N
u
s
an
t
ara
an
d
al
s
o
s
erv
i
n
g
a
s
t
h
e
h
ea
d
co
o
rd
i
n
at
o
r
o
f
t
h
e
Mo
b
i
l
e
D
ev
e
l
o
p
men
t
L
ab
o
ra
t
o
ry
.
H
i
s
re
s
earch
i
n
t
ere
s
t
s
i
n
c
l
u
d
e
req
u
i
remen
t
s
en
g
i
n
eeri
n
g
i
n
s
o
f
t
w
are
ap
p
l
i
cat
i
o
n
d
ev
e
l
o
p
men
t
,
n
at
u
ral
l
a
n
g
u
ag
e
p
ro
ce
s
s
i
n
g
,
an
d
h
u
ma
n
co
mp
u
t
er
i
n
t
erac
t
i
o
n
.
Evaluation Warning : The document was created with Spire.PDF for Python.