I
nte
rna
t
io
na
l J
o
urna
l o
f
I
nfo
rm
a
t
ics a
nd
Co
m
m
un
ica
t
io
n T
ec
hn
o
lo
g
y
(
I
J
-
I
CT
)
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
20
2
6
,
p
p
.
1
1
9
7
~
1
2
0
7
I
SS
N:
2252
-
8
7
7
6
,
DOI
:
1
0
.
1
1
5
9
1
/iji
ct
.
v
1
5
i
3
.
pp
1
1
9
7
-
1
2
0
7
1197
J
o
ur
na
l ho
m
ep
a
g
e
:
h
ttp
:
//ij
ict.
ia
esco
r
e.
co
m
An
AI
-
po
wered
k
no
wledg
e
g
ra
ph
-
ba
sed
questio
n
a
nswering
sy
stem
for
C
ha
ra
k
Sa
mhita
:
in
tegra
ting
sa
nskrit
NLP
a
nd
g
ra
ph
da
ta
scien
ce
Sh
a
ra
y
u
M
ira
s
da
r,
M
a
ng
es
h
B
edek
a
r
D
e
p
a
r
t
me
n
t
o
f
C
o
mp
u
t
e
r
E
n
g
i
n
e
e
r
i
n
g
a
n
d
Te
c
h
n
o
l
o
g
y
,
D
r
.
V
i
s
h
w
a
n
a
t
h
K
a
r
a
d
M
I
T
W
o
r
l
d
P
e
a
c
e
U
n
i
v
e
r
s
i
t
y
,
P
u
n
e
,
I
n
d
i
a
Art
icle
I
nfo
AB
S
T
RAC
T
A
r
ticle
his
to
r
y:
R
ec
eiv
ed
Au
g
1
2
,
2
0
2
5
R
ev
is
ed
Ap
r
1
6
,
2
0
2
6
Acc
ep
ted
Ma
y
1
7
,
2
0
2
6
Orig
in
a
ti
n
g
in
I
n
d
ia,
A
y
u
r
v
e
d
a
is
an
a
n
c
ien
t
m
e
d
ica
l
sy
ste
m
fo
c
u
se
d
on
h
o
li
stic
h
e
a
li
n
g
th
a
t
c
o
n
sid
e
rs
t
h
e
m
in
d
,
b
o
d
y
,
a
n
d
s
p
iri
t.
Th
is
st
u
d
y
u
ti
li
z
e
s
k
n
o
wle
d
g
e
g
ra
p
h
(KG
)
tec
h
n
o
lo
g
y
to
d
e
v
e
lo
p
a
KG
m
o
d
e
l
f
o
r
an
Ay
u
rv
e
d
a
q
u
e
stio
n
-
a
n
d
-
a
n
sw
e
r
sy
ste
m
.
Th
e
sy
ste
m
in
c
l
u
d
e
s
m
o
d
u
les
fo
r
k
n
o
wle
d
g
e
e
x
trac
ti
o
n
fr
o
m
च
र
क
स
ं
ह
िता
,
क
ा
य
ह
च
ह
क
ा
,
भ
ै
ष
र
ा
वल
ी
a
n
d
ग
ु
ण
स
ं
ि
,
c
o
n
stru
c
ti
o
n
of
KG
fro
m
t
h
is
e
x
trac
ted
k
n
o
wle
d
g
e
a
n
d
c
o
n
str
u
c
t
io
n
of
AI
su
p
p
o
rted
Q
u
e
stio
n
a
n
sw
e
r
sy
ste
m
.
In
th
e
m
e
th
o
d
o
l
o
g
y
,
d
o
m
a
in
-
s
p
e
c
ifi
c
KG
is
c
o
n
stru
c
te
d
in
Ne
o
4
j.
E
n
ti
ti
e
s
su
c
h
as
d
ise
a
se
s
(
Vy
a
d
h
i
ा
ध
ी
),
sy
m
p
to
m
s
(Lak
sh
a
n
a
ल
ण
),
d
o
s
h
a
s
(
द
ो
ष
),
h
e
rb
s,
a
n
d
trea
tme
n
ts
a
re
in
c
o
rp
o
ra
ted
.
Ad
v
a
n
c
e
d
S
a
n
sk
rit
n
a
tu
ra
l
la
n
g
u
a
g
e
p
ro
c
e
ss
in
g
(
NL
P
)
p
i
p
e
li
n
e
s
u
sin
g
By
T5
-
S
a
n
sk
rit
,
S
a
n
sk
rit
BERT
,
a
n
d
fi
n
e
-
tu
n
e
d
Bio
B
ERT
fa
c
il
it
a
te
n
a
m
e
d
e
n
ti
ty
re
c
o
g
n
it
io
n
(NER)
a
n
d
r
e
latio
n
e
x
trac
ti
o
n
.
G
ra
p
h
-
b
a
se
d
re
a
so
n
in
g
m
o
d
e
ls
su
c
h
as
g
ra
p
h
a
tt
e
n
ti
o
n
n
e
two
rk
s
(G
AT)
a
n
d
g
ra
p
h
re
a
so
n
i
n
g
e
n
h
a
n
c
e
d
lan
g
u
a
g
e
m
o
d
e
ls
(
G
REAS
EL
M
)
e
n
h
a
n
c
e
m
u
lt
i
-
h
o
p
re
a
so
n
in
g
a
c
ro
ss
Ay
u
rv
e
d
ic
c
o
n
c
e
p
ts.
Ev
a
l
u
a
ti
o
n
wa
s
c
o
n
d
u
c
ted
u
si
n
g
a
g
o
l
d
-
sta
n
d
a
rd
a
n
n
o
tate
d
d
a
tas
e
t
of
C
h
a
ra
k
S
a
m
h
it
a
v
e
rse
s
m
a
p
p
e
d
to
d
ise
a
se
–
sy
m
p
t
o
m
–
trea
tme
n
t
re
latio
n
sh
ip
s.
P
e
rfo
rm
a
n
c
e
m
e
tri
c
s
in
c
lu
d
e
d
p
re
c
isio
n
,
r
e
c
a
ll
,
F1
-
sc
o
re
,
mean
re
c
ip
ro
c
a
l
ra
n
k
(M
RR),
a
n
d
o
v
e
rlap
c
o
e
fficie
n
t
.
S
u
p
e
ri
o
r
a
c
c
u
ra
c
y
can
be
se
e
n
in
th
e
p
ro
p
o
se
d
m
o
d
e
l
as
c
o
m
p
a
re
d
to
b
a
se
li
n
e
BERT
-
QA
a
n
d
su
b
g
ra
p
h
QA
a
p
p
r
o
a
c
h
e
s.
Th
is
re
se
a
rc
h
h
a
s
in
teg
ra
ted
S
a
n
sk
rit
c
o
m
p
u
tati
o
n
a
l
l
in
g
u
isti
c
s
a
n
d
KG
sc
ien
c
e
.
Th
e
a
p
p
ro
a
c
h
m
e
n
t
io
n
e
d
in
t
h
is
p
a
p
e
r
h
a
s
m
e
n
ti
o
n
e
d
a
fra
m
e
wo
rk
th
a
t
is
sc
a
lab
le,
i
n
terp
re
tab
le
a
n
d
c
u
lt
u
ra
ll
y
si
g
n
ifi
c
a
n
t
.
W
it
h
th
e
f
o
c
u
s
on
Ay
u
r
v
e
d
a
,
th
e
m
e
th
o
d
o
lo
g
y
a
lso
m
e
n
ti
o
n
s
t
h
e
p
o
ten
ti
a
l
f
o
r
d
e
v
e
lo
p
i
n
g
c
ro
ss
-
c
u
lt
u
ra
l
m
e
d
ica
l
q
u
e
stio
n
s
–
a
n
sw
e
rin
g
sy
ste
m
s,
th
e
re
b
y
b
rid
g
i
n
g
a
n
c
ien
t
wisd
o
m
with
m
o
d
e
rn
tec
h
n
o
l
o
g
ica
l
a
p
p
ro
a
c
h
e
s.
K
ey
w
o
r
d
s
:
Ay
u
r
v
ed
a
C
h
ar
ak
S
am
h
ita
Kn
o
wled
g
e
g
r
ap
h
Natu
r
al
lan
g
u
ag
e
p
r
o
ce
s
s
in
g
Qu
esti
o
n
-
an
s
wer
s
y
s
tem
s
T
h
is
is
an
o
p
e
n
a
c
c
e
ss
a
rticle
u
n
d
e
r
th
e
CC
BY
-
SA
li
c
e
n
se
.
C
o
r
r
e
s
p
o
nd
ing
A
uth
o
r
:
Sh
ar
ay
u
Mir
asd
ar
Dep
ar
tm
en
t o
f
C
o
m
p
u
ter
E
n
g
i
n
ee
r
in
g
a
n
d
T
ec
h
n
o
lo
g
y
Dr
.
Vis
h
wan
ath
Kar
ad
MI
T
W
o
r
ld
Peac
e
Un
iv
er
s
ity
Pu
n
e,
I
n
d
ia
E
m
ail:
s
h
ar
ay
u
.
m
ir
asd
ar
@
d
es
p
u
.
ed
u
.
in
1.
I
NT
RO
D
UCT
I
O
N
Ay
u
r
v
ed
a
,
th
e
an
cien
t
I
n
d
ia
n
m
ed
ical
p
r
ac
tice
s
tan
d
s
tall
a
m
o
n
g
th
e
m
o
d
er
n
m
e
d
ical
p
r
a
ctice
s
ev
en
af
ter
s
ev
er
al
h
u
n
d
r
e
d
y
ea
r
s
.
It
is
a
h
o
lis
tic
an
d
p
r
e
v
en
tio
n
-
f
o
cu
s
ed
s
y
s
tem
of
n
atu
r
al
h
ea
l
in
g
th
at
o
r
i
g
in
ated
f
r
o
m
th
e
wis
d
o
m
of
a
n
cien
t
I
n
d
ian
m
aster
s
[
1
]
,
[
2
]
.
I
ts
p
er
s
o
n
alize
d
ap
p
r
o
ac
h
h
elp
s
ad
d
r
es
s
th
e
u
n
iq
u
e
n
ee
d
s
of
each
in
d
iv
id
u
al,
m
ak
in
g
it
a
v
alu
ab
le
p
r
ac
tice
f
o
r
m
an
y
s
ee
k
in
g
alter
n
ativ
e
or
co
m
p
lem
en
tar
y
h
ea
lin
g
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
20
2
6
:
1
1
9
7
-
1
2
0
7
1198
m
eth
o
d
s
.
T
h
e
Ay
u
r
v
ed
ic
tr
ea
tis
es
lik
e
C
h
a
r
a
k
S
a
mh
ita
,
K
a
yc
h
ikits
a
,
B
h
a
is
h
a
jya
R
a
tn
a
w
a
li,
Dra
vy
a
Gu
n
a
S
a
n
g
r
a
h
a
elab
o
r
ate
on
f
u
n
d
a
m
en
tal
asp
ec
ts
of
p
h
y
s
io
lo
g
y
,
an
ato
m
y
,
d
is
ea
s
e
etio
lo
g
y
,
an
d
p
ath
o
g
en
esis
,
alo
n
g
with
d
iag
n
o
s
tic
cr
iter
ia,
v
ar
io
u
s
m
eth
o
d
s
of
tr
ea
tm
e
n
t
(
ह
चह
क
ा
),
an
d
p
r
o
g
n
o
s
is
[
3
]
.
In
th
e
f
ield
of
m
ed
ical
s
cien
ce
,
th
e
n
ee
d
is
f
o
r
an
AI
s
y
s
tem
wh
ich
is
ca
p
ab
le
of
c
o
llectin
g
a
n
d
p
r
o
ce
s
s
in
g
d
ata
f
r
o
m
v
ar
io
u
s
s
o
u
r
ce
s
an
d
ass
is
tin
g
v
aid
y
a
(
ay
u
r
v
ed
ic
d
o
cto
r
s
)
)
in
th
eir
r
esp
o
n
s
ib
ilit
ies
[
3
]
.
A
k
n
o
w
led
g
e
g
r
ap
h
(
KG)
,
wh
ich
p
r
o
v
i
d
es
s
tr
u
ctu
r
ed
an
d
in
ter
p
r
etab
le
r
e
p
r
esen
tatio
n
s
of
d
o
m
ain
s
p
ec
if
ic
k
n
o
wle
d
g
e,
p
air
ed
with
a
q
u
esti
o
n
an
s
wer
in
g
(
QA
)
s
y
s
tem
,
s
er
v
es
as
an
ef
f
ec
tiv
e
s
o
lu
tio
n
to
m
ee
t
th
is
r
eq
u
ir
em
en
t
[
4
]
–
[
6
]
.
R
ec
en
t
ad
v
an
ce
s
in
g
r
ap
h
n
eu
r
al
n
et
wo
r
k
s
(
GNNs)
an
d
k
n
o
wled
g
e
-
awa
r
e
lan
g
u
ag
e
m
o
d
els
h
av
e
f
u
r
th
er
im
p
r
o
v
e
d
r
ea
s
o
n
in
g
ca
p
ab
ilit
ies
ac
r
o
s
s
b
io
m
ed
ical
an
d
clin
ical
co
n
tex
t
s
[
7
]
,
[
8
]
.
Ho
wev
er
,
ap
p
licatio
n
of
th
ese
m
eth
o
d
s
to
Ay
u
r
v
ed
ic
tex
ts
s
h
o
ws
s
o
m
e
u
n
iq
u
e
ch
allen
g
es
lik
e
th
e
lin
g
u
is
tic
co
m
p
le
x
ity
of
San
s
k
r
i
t
[
9
]
,
[
1
0
]
.
Do
m
ain
s
p
ec
if
icity
as
th
e
co
n
ce
p
ts
m
e
n
tio
n
ed
do
not
d
ir
ec
tly
m
ap
o
n
to
m
ed
ical
ca
teg
o
r
ies
[
1
1
]
,
[
1
2
]
,
an
d
QA
s
y
s
tem
s
ca
lab
ilit
y
wh
ich
n
ee
d
s
m
u
lti
-
hop
tr
a
v
er
s
al
[
1
3
]
,
[
1
4
]
.
T
h
is
s
tu
d
y
tack
les
lo
n
g
-
s
tan
d
in
g
ch
allen
g
es
in
th
e
co
m
p
u
t
atio
n
al
p
r
o
ce
s
s
in
g
of
an
cien
t
San
s
k
r
it
m
ed
ical
tex
ts
by
in
tr
o
d
u
ci
n
g
an
AI
-
d
r
iv
e
n
KG
-
ba
s
ed
QA
s
y
s
tem
f
o
r
th
e
C
h
ar
ak
Sam
h
ita
an
d
f
ew
more
Ay
u
r
v
ed
ic
tex
ts
.
T
h
e
f
r
a
m
ewo
r
k
co
n
tr
ib
u
tes
in
f
o
u
r
m
ajo
r
w
ay
s
.
First,
it
d
ev
elo
p
s
a
San
s
k
r
it
n
atu
r
al
lan
g
u
ag
e
p
r
o
ce
s
s
in
g
(
NL
P
)
p
ip
elin
e
th
at
in
teg
r
ates
B
y
T
5
-
Sa
n
s
k
r
it
(
f
o
r
to
k
e
n
izatio
n
)
an
d
San
s
k
r
itB
E
R
T
(
r
o
b
u
s
t
e
n
tity
r
ec
o
g
n
itio
n
s
u
ch
as
vy
a
d
h
i
,
la
ksh
a
n
a
,
an
d
a
u
s
h
a
d
h
i
).
Seco
n
d
,
it
d
ev
el
o
p
s
a
Neo
4
j
-
b
ased
Ay
u
r
v
ed
ic
KG
th
at
ex
p
licitl
y
m
o
d
els
d
is
ea
s
e
–
s
y
m
p
to
m
–
tr
ea
tm
en
t
r
elatio
n
s
h
ip
s
,
th
e
r
eb
y
ca
p
t
u
r
in
g
th
e
in
tr
icate
in
ter
d
ep
e
n
d
en
cies
d
e
s
cr
ib
ed
in
th
e
C
h
ar
a
k
Sam
h
it
a.
T
h
ir
d
,
th
e
s
y
s
tem
in
teg
r
ates
g
r
ap
h
r
ea
s
o
n
i
n
g
alg
o
r
ith
m
s
,
in
clu
d
in
g
g
r
ap
h
atten
tio
n
n
etwo
r
k
s
(
GAT
)
an
d
g
r
ap
h
r
ea
s
o
n
in
g
en
h
an
ce
d
lan
g
u
ag
e
m
o
d
el
(
GR
E
ASEL
M
)
,
to
f
ac
ilit
ate
m
u
lti
-
hop
r
ea
s
o
n
in
g
f
o
r
co
m
p
le
x
u
s
er
q
u
er
ies.
Fin
ally
,
th
e
f
r
a
m
ewo
r
k
u
n
d
er
g
o
es
ex
p
er
im
en
tal
ev
al
u
atio
n
ag
ai
n
s
t
s
tr
o
n
g
b
aselin
es,
d
em
o
n
s
tr
atin
g
s
ig
n
if
ican
t
im
p
r
o
v
e
m
en
ts
in
ac
cu
r
ac
y
,
in
ter
p
r
etab
ilit
y
,
an
d
d
o
m
ain
r
elev
an
ce
.
T
h
u
s
th
is
s
tu
d
y
ad
v
an
ce
s
d
ig
ital
Ay
u
r
v
ed
a
r
e
s
ea
r
ch
,
o
f
f
er
i
n
g
a
s
ca
lab
le,
ex
p
lain
ab
le,
an
d
cli
n
ically
r
elev
a
n
t
QA
s
y
s
tem
c
ap
ab
le
of
b
r
id
g
i
n
g
class
ical
Ay
u
r
v
ed
ic
wis
d
o
m
with
m
o
d
er
n
AI
-
d
r
i
v
en
h
e
alth
ca
r
e
in
f
o
r
m
atics
[
1
5
]
,
[
1
6
]
.
2.
L
I
T
E
R
AT
U
RE
R
E
VI
E
W
2
.
1
.
Ay
urv
eda
a
nd
t
he
Cha
ra
k
Sa
m
hita
C
h
ar
ak
Sam
h
ita
is
Ay
u
r
v
ed
a'
s
d
ef
in
itiv
e
tr
ea
tis
e
wh
ich
m
en
tio
n
s
th
e
co
n
ce
p
ts
of
3
d
o
s
h
as
(
Vaa
t,
Pit
ta,
Kap
h
a)
,
7
Dh
atu
(
b
o
d
y
tis
s
u
es)
an
d
3
m
ala
(
waste
p
r
o
d
u
cts)
alo
n
g
with
t
h
e
v
ar
io
u
s
tr
ea
tm
e
n
t
s
tr
ateg
ies
[
1
7
]
,
[
1
8
]
.
It
r
em
ai
n
s
r
elev
an
t
to
m
o
d
er
n
h
ea
lth
ca
r
e
b
y
o
f
f
er
in
g
a
h
o
lis
tic
ap
p
r
o
a
ch
wh
ich
v
e
r
y
well
alig
n
s
with
co
n
tem
p
o
r
ar
y
p
r
i
n
cip
les
[
1
9
]
.
B
u
t,
m
o
d
e
r
n
ac
c
ess
is
co
n
s
tr
ain
ed
by
u
n
s
tr
u
ct
u
r
ed
tex
t
f
o
r
m
s
an
d
s
ca
r
city
of
co
m
p
u
tatio
n
al
r
eso
u
r
ce
s
[
2
0
]
.
2
.
2
.
QA
o
v
er
KG
s
KG
-
b
ased
QA
h
as
em
er
g
ed
as
a
p
r
o
m
is
in
g
p
ar
ad
i
g
m
,
en
a
b
lin
g
s
tr
u
ctu
r
ed
q
u
esti
o
n
in
ter
p
r
e
tatio
n
an
d
r
ea
s
o
n
in
g
.
T
ec
h
n
iq
u
es
in
clu
d
e
s
u
b
g
r
ap
h
s
ea
r
ch
in
g
[
2
1
]
,
r
elatio
n
-
awa
r
e
m
ap
p
in
g
with
tr
an
s
f
o
r
m
er
s
,
an
d
m
u
lti
-
h
o
p
r
ea
s
o
n
in
g
u
s
in
g
GNNs
[
2
2
]
.
A
d
v
an
ce
d
m
o
d
els
s
u
ch
as
GR
E
ASEL
M
an
d
KG
r
an
k
[
2
3
]
d
em
o
n
s
tr
ate
im
p
r
o
v
ed
r
ea
s
o
n
in
g
by
co
m
b
in
i
n
g
lan
g
u
ag
e
m
o
d
els
with
s
tr
u
ctu
r
ed
g
r
ap
h
s
in
h
ea
lth
ca
r
e
[
2
4
]
,
[
2
5
]
.
2
.
3
.
G
ra
ph
da
t
a
s
cience
in
hea
lt
hca
re
Gr
ap
h
d
ata
s
cien
ce
(
GDS)
l
ev
er
ag
es
n
etwo
r
k
-
b
ased
alg
o
r
ith
m
s
to
ex
tr
ac
t
p
atter
n
s
,
cl
u
s
ter
s
,
an
d
p
ath
s
in
co
m
p
lex
s
y
s
tem
s
[
2
6
]
.
T
h
e
in
teg
r
atio
n
of
g
r
ap
h
s
im
ilar
ity
,
o
v
er
lap
co
ef
f
ici
en
ts
,
an
d
atten
tio
n
m
ec
h
an
is
m
s
h
as
s
h
o
wn
s
tr
o
n
g
p
o
ten
tial
f
o
r
co
m
p
lex
d
ec
is
io
n
-
m
ak
in
g
[
2
7
]
,
[
2
8
]
.
2
.
4
.
Sa
ns
k
rit
NL
P
a
nd
Co
mp
uta
t
io
na
l
Cha
lleng
es:
San
s
k
r
it,
ch
ar
ac
ter
ized
by
it
s
h
ig
h
ly
in
f
lecte
d
m
o
r
p
h
o
l
o
g
y
a
n
d
San
d
h
i
r
u
les,
p
o
s
es
ch
allen
g
es
f
o
r
t
o
k
en
izatio
n
an
d
s
e
m
an
tic
p
ar
s
in
g
[
2
9
]
.
R
ec
en
t
ad
v
a
n
ce
s
s
u
ch
as
B
y
T
5
-
San
s
k
r
it
an
d
d
o
m
ain
-
s
p
ec
if
ic
ad
ap
tatio
n
s
of
B
E
R
T
m
o
d
els
f
o
r
San
s
k
r
it
o
f
f
er
r
o
b
u
s
t
s
o
lu
tio
n
s
f
o
r
San
d
h
i
s
p
litt
in
g
,
NE
R
,
an
d
s
em
an
tic
r
o
le
la
b
elin
g
.
2.
5
.
AI
-
enha
nced
a
y
urv
edic
s
y
s
t
em
s
Sev
er
al
s
tu
d
ies
h
a
v
e
h
ig
h
lig
h
t
ed
th
e
s
y
n
er
g
y
b
etwe
en
AI
a
n
d
Ay
u
r
v
ed
a
.
Fo
r
ex
a
m
p
le,
g
r
a
p
h
-
b
ased
d
iag
n
o
s
tic
s
y
s
tem
s
h
av
e
b
ee
n
ex
p
lo
r
ed
f
o
r
Ay
u
r
v
ed
ic
d
i
ag
n
o
s
is
[
3
0
]
,
wh
ile
m
o
d
er
n
KG
ap
p
licatio
n
s
in
h
ea
lth
ca
r
e
d
em
o
n
s
tr
ate
th
e
in
t
er
p
r
etab
ilit
y
ad
v
an
tag
e
of
s
y
m
b
o
lic.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
A
n
A
I
-
p
o
w
ered
kn
o
w
led
g
e
g
r
a
p
h
-
b
a
s
ed
q
u
esti
o
n
a
n
s
w
er
in
g
s
ystem
fo
r
C
h
a
r
a
k
…
(
S
h
a
r
a
y
u
Mir
a
s
d
a
r
)
1199
3.
RE
S
E
ARCH
G
AP
AND
O
B
J
E
CT
I
VE
S
Alth
o
u
g
h
p
r
io
r
s
tu
d
ies
h
a
v
e
a
d
v
an
ce
d
KG
-
QA
[
3
1
]
,
th
eir
a
p
p
licatio
n
to
class
ical
San
s
k
r
it
m
ed
ical
tex
ts
r
em
ain
s
lar
g
ely
u
n
ex
p
lo
r
ed
.
C
u
r
r
en
t
r
esear
ch
on
Ay
u
r
v
ed
a
m
ain
ly
f
o
cu
s
es
on
co
n
ce
p
tu
al
an
aly
s
is
or
s
u
r
v
ey
-
b
ased
h
ea
lth
ca
r
e
ad
o
p
tio
n
s
tu
d
ies,
with
lim
ited
c
o
m
p
u
tatio
n
al
im
p
lem
e
n
tatio
n
s
.
Mo
r
eo
v
e
r
,
w
h
ile
g
r
ap
h
d
ata
s
cien
ce
h
as
d
em
o
n
s
tr
ated
s
ig
n
if
ican
t
u
tili
ty
in
h
ea
lth
ca
r
e
d
ec
is
io
n
-
m
ak
in
g
,
i
ts
in
teg
r
atio
n
with
San
s
k
r
it
NL
P
f
o
r
QA
o
v
e
r
a
n
cien
t
m
ed
ical
tr
ea
tis
es
is
n
o
v
el.
T
h
e
p
r
im
ar
y
o
b
jectiv
e
of
th
is
r
esear
ch
is
to
d
esig
n
an
d
im
p
lem
e
n
t
an
AI
-
p
o
wer
ed
f
r
a
m
ewo
r
k
th
at
in
te
g
r
ates
San
s
k
r
it
NL
P
with
KG
m
eth
o
d
o
lo
g
ies
to
s
u
p
p
o
r
t
ad
v
an
ce
d
QA
o
v
er
th
e
C
h
ar
ak
Sam
h
ita.
4.
M
E
T
H
O
DO
L
O
G
Y
T
h
e
m
eth
o
d
o
lo
g
y
f
o
llo
ws
a
f
o
u
r
-
s
tag
e
p
ip
elin
e
in
Fig
u
r
e
1
.
Fig
u
r
e
1.
Fo
u
r
-
s
tag
e
p
ip
eli
n
e
4
.1
.
Sa
ns
k
rit
NL
P
p
ipelin
e
A
h
y
b
r
i
d
p
ip
elin
e
was
d
ev
elo
p
ed
to
ex
tr
ac
t
s
tr
u
ctu
r
ed
Ay
u
r
v
ed
ic
k
n
o
wled
g
e
f
r
o
m
class
ical
San
s
k
r
it
tex
ts
,
in
teg
r
atin
g
ad
v
a
n
ce
d
tr
an
s
f
o
r
m
er
-
b
ased
m
o
d
els
with
d
o
m
ai
n
-
s
p
ec
if
ic
p
o
s
t
-
p
r
o
ce
s
s
in
g
.
Nam
ed
en
tity
r
ec
o
g
n
itio
n
(
NE
R
)
Mo
d
u
le:
Fin
e
-
tu
n
ed
B
y
T
5
-
San
s
k
r
it
an
d
San
s
k
r
itB
E
R
T
m
o
d
els
wer
e
e
m
p
lo
y
ed
to
id
en
tif
y
k
ey
Ay
u
r
v
ed
ic
en
titi
es,
in
clu
d
in
g
Vy
a
d
h
i
(
d
is
ea
s
es),
L
ak
s
h
an
a
(
s
y
m
p
to
m
s
)
,
Dr
a
v
y
a
(
h
e
r
b
s
/m
ed
icin
es),
an
d
co
n
s
titu
tio
n
al
f
ac
to
r
s
s
u
ch
as
Do
s
h
a,
Dh
atu
,
a
n
d
Ma
la.
R
el
atio
n
class
if
icatio
n
b
etwe
en
i
d
en
tifie
d
en
titi
es
is
done
with
th
e
h
el
p
of
th
e
B
io
B
E
R
T
m
o
d
el.
I
m
p
o
r
tan
t
r
elatio
n
s
h
ip
s
lik
e,
Vy
a
d
h
i
-
L
a
k
s
h
an
a,
Vy
ad
h
i
-
Dr
a
v
y
a,
an
d
Dr
av
y
a
–
Do
s
h
a
a
r
e
ca
p
t
u
r
ed
by
th
e
s
y
s
tem
.
Acc
u
r
ate
s
em
an
tic
lin
k
ag
e
b
etwe
en
e
n
titi
es
is
ca
p
tu
r
ed
by
f
in
e
-
tu
n
in
g
th
e
m
o
d
el
on
a
n
n
o
tated
San
s
k
r
it
-
Ay
u
r
v
e
d
i
co
r
p
o
r
a.
In
t
h
e
s
tag
e
of
P
o
s
t
-
p
r
o
ce
s
s
in
g
an
d
Valid
atio
n
,
th
e
ex
tr
ac
ted
e
n
titi
es
an
d
r
elatio
n
s
wer
e
alig
n
ed
ag
a
in
s
t
Ay
u
r
v
ed
ic
o
n
to
lo
g
ies.
T
h
is
m
u
lti
-
s
tag
e
p
ip
elin
e
e
n
s
u
r
es
r
o
b
u
s
t
e
x
tr
a
ctio
n
of
s
tr
u
ctu
r
ed
k
n
o
wled
g
e,
wh
ich
will
be
th
e
f
o
u
n
d
ati
o
n
f
o
r
d
o
w
n
s
tr
ea
m
task
s
s
u
ch
as
KG
co
n
s
tr
u
ctio
n
an
d
QA
s
y
s
tem
s
in
th
e
Ay
u
r
v
ed
ic
d
o
m
ain
.
4
.2
.
Ay
urv
edic
k
no
wledg
e
g
ra
ph
dev
elo
pm
ent
Usi
n
g
th
e
Neo
4
j
g
r
a
p
h
d
atab
a
s
e,
s
tr
u
ctu
r
ed
r
elatio
n
s
h
ip
s
b
et
wee
n
k
ey
d
o
m
ain
en
titi
es
wer
e
ca
p
tu
r
e
d
an
d
Ay
u
r
v
ed
ic
KG
was
co
n
s
tr
u
cted
.
KG
n
o
d
es
wer
e
V
ya
d
h
i
,
La
ksh
a
n
a
,
Do
s
h
a
s
,
Dra
vy
a
,
an
d
Tr
ea
tmen
ts
,
wh
ile
ed
g
es
ca
p
tu
r
ed
s
em
an
tic
r
elatio
n
s
lik
e
“ca
u
s
es”
,
“tr
ea
ted
_
b
y”
,
“a
s
s
o
cia
ted
_
w
ith
”
,
an
d
“b
a
la
n
ce
s
”
.
E
ac
h
node
a
n
d
ed
g
e
is
an
n
o
ta
ted
with
p
r
o
p
er
ties
s
u
ch
as
San
s
k
r
it
ter
m
s
,
E
n
g
lis
h
tr
an
s
liter
atio
n
s
,
an
d
tex
tu
al
r
ef
er
en
ce
s
f
r
o
m
class
ical
s
o
u
r
ce
s
,
s
u
p
p
o
r
tin
g
m
u
ltil
in
g
u
al
an
d
p
r
o
v
e
n
an
ce
-
awa
r
e
q
u
er
ie
s
.
GDS
alg
o
r
ith
m
s
(
Pag
eRan
k
,
Gr
ap
h
Similar
ity
,
Ov
er
la
C
o
ef
f
icien
t)
wer
e
ap
p
lied
to
en
h
an
ce
g
r
ap
h
-
b
a
s
ed
r
ea
s
o
n
in
g
an
d
k
n
o
wled
g
e
d
is
co
v
er
y
.
4
.3
.
QA
s
y
s
t
e
m
a
rc
hite
ct
ure
T
h
e
ay
u
r
v
ed
ic
KG
is
tr
av
er
s
e
d
by
a
p
r
o
p
o
s
ed
QA
s
y
s
tem
to
r
esp
o
n
d
to
th
e
q
u
er
ies
wr
itt
en
in
eith
e
r
San
s
k
r
it
or
E
n
g
lis
h
.
T
h
e
p
i
p
elin
e
d
esig
n
is
as
f
o
llo
ws
:
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
20
2
6
:
1
1
9
7
-
1
2
0
7
1200
−
Qu
esti
o
n
p
ar
s
in
g
:
On
ce
th
e
u
s
er
ask
s
q
u
esti
o
n
s
,
th
ey
ar
e
p
r
o
ce
s
s
ed
th
r
o
u
g
h
th
e
NL
P
m
o
d
u
le,
wh
ich
p
er
f
o
r
m
s
to
k
en
izatio
n
an
d
d
e
p
en
d
en
c
y
p
a
r
s
in
g
,
to
ex
tr
ac
t
m
ea
n
in
g
f
u
l
ter
m
s
an
d
s
y
n
tact
ic
s
tr
u
ctu
r
es.
T
h
is
p
r
o
ce
s
s
m
ak
es
s
u
r
e
th
at
b
o
th
San
s
k
r
it
an
d
E
n
g
lis
h
in
p
u
t
s
ar
e
n
o
r
m
alize
d
f
o
r
f
u
r
th
e
r
p
r
o
ce
s
s
in
g
.
−
E
n
tity
lin
k
in
g
:
T
h
e
ter
m
s
,
ex
tr
ac
ted
in
th
e
ab
o
v
e
p
h
ase,
a
r
e
m
ap
p
ed
to
KG
n
o
d
es
u
s
in
g
em
b
ed
d
i
n
g
-
b
ased
s
im
ilar
ity
m
ea
s
u
r
es.
T
h
is
s
tep
r
ec
o
n
ciles
lin
g
u
is
tic
v
ar
iatio
n
s
by
m
ap
p
in
g
u
s
er
v
o
ca
b
u
lar
y
to
s
tan
d
ar
d
ized
Ay
u
r
v
ed
ic
e
n
titi
es
(
d
is
ea
s
e,
s
y
m
p
to
m
,
h
er
b
,
tr
e
atm
en
t,
d
o
s
h
a)
.
−
Gr
ap
h
tr
av
er
s
al
:
Qu
er
y
r
eso
lu
t
io
n
with
in
th
e
KG
is
ac
h
iev
ed
th
r
o
u
g
h
lay
er
ed
tr
a
v
er
s
al
s
tr
ateg
ies.
Simp
le
q
u
er
ies
ar
e
p
r
o
ce
s
s
ed
u
s
in
g
s
u
b
g
r
a
p
h
s
ea
r
ch
to
e
x
tr
ac
t
ex
p
licit
r
elatio
n
s
s
u
ch
as
Vy
ad
h
i
–
tr
ea
ted
_
b
y
–
Her
b
s
.
Mo
r
e
co
m
p
le
x
in
f
o
r
m
atio
n
n
ee
d
s
,
wh
ich
r
e
q
u
ir
e
in
f
er
en
ce
,
ar
e
a
d
d
r
ess
ed
v
ia
m
u
lti
-
hop
r
ea
s
o
n
in
g
tech
n
i
q
u
e
s.
GAT
an
d
GR
E
ASEL
M
ar
e
em
p
lo
y
ed
to
ca
p
tu
r
e
co
n
tex
tu
al
d
e
p
en
d
en
cies
an
d
p
r
o
p
a
g
ate
s
em
an
tic
s
ig
n
als
ac
r
o
s
s
m
u
ltip
le
n
o
d
es
an
d
ed
g
es.
T
h
is
en
ab
les
th
e
d
is
co
v
er
y
of
in
d
ir
ec
t
ass
o
ciatio
n
s
o
th
er
wis
e
h
id
d
e
n
in
th
e
g
r
ap
h
s
tr
u
ctu
r
e
.
By
in
teg
r
atin
g
d
ir
ec
t
r
etr
iev
al
with
ad
v
an
ce
d
in
f
er
en
ce
,
t
h
e
s
y
s
tem
s
u
p
p
o
r
ts
r
o
b
u
s
t
an
d
s
em
an
tically
en
r
ic
h
ed
ex
p
lo
r
atio
n
of
Ay
u
r
v
e
d
ic
k
n
o
wled
g
e.
−
An
s
wer
r
an
k
in
g
:
KG
R
an
k
m
o
d
el,
c
o
m
b
in
e
d
with
s
im
ilar
ity
m
ea
s
u
r
es,
ar
e
u
s
ed
to
r
a
n
k
th
e
an
s
wer
s
wh
ich
en
s
u
r
es
th
e
m
o
s
t
r
elev
a
n
t
an
d
s
em
an
tically
c
o
r
r
ec
t
r
es
p
o
n
s
es
b
ased
on
th
e
p
r
i
o
r
ities
.
T
h
e
s
y
s
tem
is
d
esig
n
ed
to
p
r
o
v
id
e
in
ter
p
r
eta
b
le,
k
n
o
wled
g
e
-
g
r
o
u
n
d
e
d
a
n
s
wer
s
to
u
s
er
q
u
er
ies.
Fo
r
ex
am
p
le,
wh
e
n
th
e
in
p
u
t
is
“W
h
ich
h
erb
is
u
s
ed
in
th
e
tr
ea
tmen
t
of
R
a
a
jya
ksh
ma
(
र
ा
जया
)
a
cc
o
r
d
in
g
to
C
h
a
r
a
k
S
a
mh
ita
?
”
,
th
e
q
u
e
r
y
u
n
d
er
g
o
es
m
u
ltip
le
s
tag
es
of
p
r
o
ce
s
s
in
g
.
First,
th
e
NL
P
m
o
d
u
le
will
do
en
tity
r
ec
o
g
n
itio
n
,
id
en
tif
y
in
g
R
a
a
jy
a
ksh
ma
as
a
Vy
ad
h
i
(
d
is
ea
s
e)
an
d
“h
er
b
”
as
a
Dr
av
y
a
(
m
e
d
i
cin
e)
.
T
h
ese
en
titi
es
ar
e
th
en
m
ap
p
ed
to
th
e
KG.
As
th
is
is
th
e
r
elatio
n
s
h
ip
b
etwe
en
Vy
ad
h
i
an
d
Her
b
,
th
e
s
y
s
tem
s
ea
r
ch
es
alo
n
g
th
e
Vy
ad
h
i
–
D
r
av
y
a
ed
g
es
to
id
en
tify
ap
p
r
o
p
r
iate
h
e
r
b
s
f
o
r
th
is
d
is
ea
s
e.
In
th
is
ca
s
e,
th
e
tr
av
er
s
al
r
etr
iev
es
न
ा
गप
ु
,
क
प
पू
र
,
ल
व
an
d
o
th
er
s
wh
ich
ar
e
u
s
ed
in
th
e
tr
ea
tm
en
t
of
R
a
a
jy
a
ksh
ma
.
Fin
ally
,
th
e
an
s
wer
is
en
r
ich
ed
with
s
u
p
p
o
r
tin
g
v
er
s
e
r
ef
er
en
ce
s
f
r
o
m
th
e
C
h
a
r
a
k
S
a
mh
ita
,
e
n
s
u
r
in
g
tr
a
n
s
p
ar
en
cy
an
d
cu
ltu
r
al
g
r
o
u
n
d
in
g
.
T
h
is
in
ter
p
r
etab
le
p
ip
elin
e
g
u
ar
an
tees
n
o
t
o
n
l
y
ac
cu
r
ate
r
esp
o
n
s
es
but
also
ex
p
lain
ab
ilit
y
an
d
au
th
en
ticity
in
th
e
c
o
n
tex
t
of
Ay
u
r
v
ed
ic
k
n
o
wled
g
e.
5.
E
XP
E
R
I
M
E
N
T
A
L
DE
SI
G
N
AND
E
VA
L
UA
T
I
O
N
P
RO
T
O
CO
L
S
E
v
alu
atio
n
m
etr
ics
:
Sy
s
tem
p
er
f
o
r
m
an
ce
was
r
ig
o
r
o
u
s
ly
ass
es
s
ed
by
co
m
b
in
in
g
q
u
an
t
itativ
e
an
d
g
r
ap
h
-
b
ased
m
etr
ics.
Pre
cisi
o
n
,
R
ec
all,
an
d
F1
-
s
co
r
e
wer
e
u
s
ed
as
co
r
e
m
ea
s
u
r
es
to
ev
al
u
ate
th
e
co
r
r
ec
t
n
ess
of
r
etr
iev
e
d
a
n
s
wer
s
with
r
esp
ec
t
to
g
o
ld
-
s
tan
d
a
r
d
a
n
n
o
tat
io
n
s
.
Mean
r
e
cip
r
o
ca
l
r
an
k
(
MRR
)
was
u
s
ed
to
m
ea
s
u
r
e
how
h
ig
h
ly
th
e
co
r
r
ec
t
an
s
wer
was
g
en
er
ated
in
th
e
r
an
k
ed
r
esu
lts
lis
t.
T
h
e
o
v
er
lap
c
o
ef
f
icien
t
q
u
an
tifie
d
th
e
d
eg
r
ee
of
en
ti
ty
o
v
er
lap
b
etwe
en
s
y
s
tem
o
u
tp
u
ts
an
d
r
ef
e
r
en
ce
KG
an
n
o
tatio
n
s
,
en
s
u
r
in
g
s
em
an
tic
co
n
s
is
ten
cy
.
T
h
e
g
r
ap
h
s
im
ilar
ity
in
d
ex
was
also
ca
lcu
lated
u
s
in
g
s
tr
u
ctu
r
al
e
m
b
ed
d
in
g
d
is
tan
ce
s
wh
ich
ca
p
tu
r
ed
cl
o
s
en
ess
of
ca
p
tu
r
ed
s
u
b
g
r
ap
h
s
.
T
o
g
eth
er
,
th
ese
m
etr
ics
p
r
o
v
id
e
a
b
alan
ce
d
ev
alu
atio
n
f
r
am
ewo
r
k
wh
ic
h
co
v
er
s
b
o
th
an
s
wer
-
lev
el
co
r
r
ec
tn
ess
an
d
g
r
ap
h
-
lev
el
f
id
elity
,
an
d
also
p
r
o
v
id
e
th
e
ac
cu
r
ac
y
an
d
in
ter
p
r
etab
ilit
y
of
th
e
p
r
o
p
o
s
ed
QA
s
y
s
tem
.
5
.
1
.
E
v
a
lua
t
i
o
n
p
ro
t
o
co
ls
A
m
u
lti
-
s
tag
e
ev
alu
atio
n
p
r
o
t
o
co
l
was
d
esig
n
ed
to
c
o
m
p
r
e
h
en
s
iv
ely
v
alid
ate
th
e
p
r
o
p
o
s
ed
s
y
s
tem
ac
r
o
s
s
en
tity
r
ec
o
g
n
itio
n
,
r
elat
io
n
ex
tr
ac
tio
n
,
an
d
QA
task
s
.
−
E
n
tity
r
ec
o
g
n
itio
n
task
:
NE
R
m
o
d
els
wer
e
e
v
alu
ated
a
g
ain
s
t
a
m
an
u
ally
a
n
n
o
tated
g
o
ld
-
s
tan
d
ar
d
d
ataset,
with
Pre
cisi
o
n
,
R
ec
all
,
an
d
F1
-
s
co
r
es
co
m
p
u
te
d
to
m
ea
s
u
r
e
en
tity
-
lev
el
ac
c
u
r
ac
y
.
−
R
elatio
n
ex
tr
ac
tio
n
task
:
R
elat
io
n
s
s
u
ch
as
V
ya
d
h
i
–
La
ksh
a
n
a
,
V
ya
d
h
i
–
Dra
vy
a
,
an
d
Dra
vy
a
–
Do
s
h
a
wer
e
ev
alu
ated
u
s
in
g
d
o
m
ai
n
-
s
p
ec
if
ic
lab
els.
Mo
d
el
p
er
f
o
r
m
an
ce
was
co
m
p
a
r
ed
u
s
in
g
F1
-
s
co
r
es,
h
ig
h
lig
h
tin
g
th
e
ef
f
ec
tiv
en
ess
of
B
io
B
E
R
T
-
b
ased
r
elatio
n
class
if
icatio
n
in
San
s
k
r
it
-
Ay
u
r
v
ed
ic
co
n
tex
ts
.
−
QA
task
:
E
n
d
-
to
-
en
d
s
y
s
tem
ev
alu
at
io
n
was
ca
r
r
ied
out
u
s
in
g
q
u
er
ies
p
o
s
ed
in
b
o
th
San
s
k
r
it
an
d
E
n
g
lis
h
.
Pre
d
icted
an
s
wer
s
w
er
e
v
alid
ated
ag
ain
s
t
g
o
ld
-
s
tan
d
ar
d
an
n
o
tatio
n
s
as
well
as
ex
p
er
t
-
r
ev
iewe
d
Ay
u
r
v
ed
ic
r
ef
er
e
n
ce
s
,
en
s
u
r
in
g
clin
ical
an
d
c
u
ltu
r
al
f
id
elity
.
To
estab
lis
h
th
e
r
o
b
u
s
tn
ess
of
r
esu
lts
,
p
air
ed
t
-
test
s
wer
e
p
er
f
o
r
m
e
d
ac
r
o
s
s
b
aselin
es
an
d
t
h
e
p
r
o
p
o
s
ed
s
y
s
tem
.
Statis
tical
s
ig
n
if
ican
ce
was
co
n
f
ir
m
e
d
at
p
<
0
.
0
5
,
e
n
s
u
r
in
g
th
at
im
p
r
o
v
em
en
ts
wer
e
not
due
to
r
a
n
d
o
m
v
ar
iatio
n
5
.
2
.
Da
t
a
s
et
d
escript
io
n
T
h
e
d
ataset
u
s
ed
f
o
r
tr
ain
in
g
an
d
ex
tr
ac
tio
n
co
n
s
is
ted
of
C
h
ar
ak
a
Saṃ
h
ita
v
e
r
s
es,
m
an
u
a
lly
cu
r
ated
an
d
s
eg
m
en
ted
in
t
o
Ay
u
r
v
ed
ic
en
titi
es
s
u
ch
as
V
yā
d
h
i
(
d
is
ea
s
es),
La
kṣa
ṇ
a
(
s
y
m
p
to
m
s
)
,
Oṣa
d
h
i
(
h
er
b
s
)
,
Gu
ṇ
a
(
q
u
alities
)
,
R
a
s
a
,
K
a
r
ma
,
an
d
Dra
vy
a
–
C
o
mp
o
u
n
d
r
ela
tio
n
s
h
ip
s
.
A
to
tal
of
1
2
,
4
8
0
tag
g
ed
s
en
ten
ce
s
(
ap
p
r
o
x
.
1
.
8
5
lak
h
to
k
e
n
s
)
wer
e
p
r
ep
ar
e
d
in
B
-
I
-
O
f
o
r
m
at,
co
m
b
in
in
g
b
o
th
m
an
u
ally
a
n
n
o
tated
v
er
s
es
an
d
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
A
n
A
I
-
p
o
w
ered
kn
o
w
led
g
e
g
r
a
p
h
-
b
a
s
ed
q
u
esti
o
n
a
n
s
w
er
in
g
s
ystem
fo
r
C
h
a
r
a
k
…
(
S
h
a
r
a
y
u
Mir
a
s
d
a
r
)
1201
ad
d
itio
n
al
s
em
i
-
au
t
o
m
atica
lly
an
n
o
tated
te
x
t
v
er
i
f
ied
by
ex
p
er
ts
.
T
h
e
f
in
al
KG
was
co
n
s
tr
u
cte
d
h
av
in
g
1
7
,
1
5
9
n
o
d
es
an
d
4
6
1
2
4
ed
g
es.
T
h
is
d
ataset
s
e
r
v
ed
as
th
e
in
p
u
t
f
o
r
b
o
th
NE
R
tr
ain
in
g
an
d
d
o
wn
s
tr
ea
m
lin
k
e
x
tr
ac
tio
n
.
6.
RE
SU
L
T
S
AND
AN
AL
Y
SI
S
T
h
e
ex
p
e
r
im
en
tal
ev
al
u
atio
n
d
em
o
n
s
tr
ates
th
e
ef
f
ec
tiv
en
es
s
of
th
e
p
r
o
p
o
s
ed
San
s
k
r
itB
E
R
T
+
KG
+
GDS
p
ip
elin
e
in
Ay
u
r
v
ed
ic
Q
A.
E
n
tity
r
ec
o
g
n
itio
n
p
er
f
o
r
m
an
ce
s
h
o
w
in
T
a
b
le
1.
T
ab
le
1.
E
n
tity
r
ec
o
g
n
itio
n
p
e
r
f
o
r
m
a
n
ce
M
o
d
e
l
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
s
c
o
r
e
En
t
i
t
i
e
s
C
o
v
e
r
e
d
B
ER
T
0
.
7
8
0
.
7
4
0
.
7
6
V
y
a
d
h
i
,
La
k
sh
a
n
a
B
i
o
B
E
R
T
0
.
8
1
0
.
7
9
0
.
8
0
V
y
a
d
h
i
,
La
k
sh
a
n
a
,
D
r
a
v
y
a
S
a
n
s
k
r
i
t
B
ER
T
0
.
8
6
0
.
8
2
0
.
8
4
V
y
a
d
h
i
,
La
k
sh
a
n
a
,
D
o
s
h
a
B
y
T
5
-
S
a
n
s
k
r
i
t
0
.
8
8
0
.
8
6
0
.
8
7
V
y
a
d
h
i
,
La
k
sh
a
n
a
,
D
r
a
v
y
a
,
D
o
sh
a
P
r
o
p
o
se
d
m
o
d
e
l
(
B
y
T
5
B
i
o
B
ER
T)
0
.
9
1
0
.
8
9
0
.
9
0
A
l
l
A
y
u
r
v
e
d
i
c
E
n
t
i
t
i
e
s
E
n
tity
r
ec
o
g
n
itio
n
f
o
r
m
s
th
e
b
ac
k
b
o
n
e
of
QA
s
y
s
tem
s
.
T
h
e
b
aselin
e
B
E
R
T
m
o
d
el
s
h
o
wed
lim
itatio
n
s
in
h
an
d
lin
g
San
s
k
r
it
m
o
r
p
h
o
lo
g
y
.
B
io
B
E
R
T
im
p
r
o
v
e
d
r
esu
lts
by
d
o
m
ain
a
d
a
p
tatio
n
but
lack
ed
San
s
k
r
it
s
en
s
itiv
ity
.
San
s
k
r
itB
E
R
T
an
d
B
y
T
5
-
San
s
k
r
it
s
ig
n
i
f
ican
tly
im
p
r
o
v
ed
r
ec
o
g
n
itio
n
of
in
f
lecte
d
f
o
r
m
s
,
o
u
tp
er
f
o
r
m
in
g
ea
r
lier
m
o
d
els.
T
h
e
h
y
b
r
id
ap
p
r
o
ac
h
co
m
b
i
n
in
g
B
y
T
5
-
San
s
k
r
it
with
B
io
B
E
R
T
ac
h
iev
ed
th
e
h
ig
h
est
F1
-
s
co
r
e
(
0
.
9
0
)
,
d
em
o
n
s
tr
atin
g
r
o
b
u
s
tn
ess
in
ex
tr
ac
tin
g
en
titi
es
lik
e
Do
s
h
a
an
d
Dra
vy
a
,
wh
ich
wer
e
not
well
ca
p
tu
r
ed
by
g
en
er
al
m
o
d
els.
T
h
ese
r
esu
lts
co
n
f
i
r
m
th
at
San
s
k
r
it
-
s
p
ec
ialized
m
o
d
els
o
u
tp
er
f
o
r
m
g
en
er
ic
b
io
m
e
d
ical
o
n
es,
s
u
p
p
o
r
tin
g
p
r
io
r
claim
s
in
co
m
p
u
tatio
n
al
San
s
k
r
it
lin
g
u
is
tics
.
Fig
u
r
e
2
illu
s
tr
ates
m
o
d
el
p
e
r
f
o
r
m
an
ce
f
o
r
en
tit
y
r
ec
o
g
n
itio
n
.
T
h
e
p
r
o
p
o
s
ed
h
y
b
r
id
m
o
d
el
s
u
r
p
ass
es
all
b
aselin
es
with
an
F1
-
s
co
r
e
of
0
.
9
0
,
h
ig
h
lig
h
tin
g
th
e
im
p
o
r
tan
ce
of
d
o
m
ain
-
s
p
e
cif
ic
ad
ap
tatio
n
.
Fig
u
r
e
2
.
C
o
m
p
a
r
is
o
n
of
NE
R
m
o
d
el
p
e
r
f
o
r
m
an
ce
R
elatio
n
ex
tr
ac
tio
n
,
as
s
h
o
w
n
in
T
a
b
le
2,
was
test
ed
on
d
is
ea
s
e
–
s
y
m
p
to
m
an
d
d
is
ea
s
e
–
tr
ea
tm
en
t
m
ap
p
in
g
s
.
T
r
an
s
f
o
r
m
er
-
o
n
ly
QA
s
tr
u
g
g
led
d
u
e
to
a
b
s
en
c
e
of
s
tr
u
ctu
r
al
co
n
s
tr
ain
ts
.
B
io
B
E
R
T
im
p
r
o
v
ed
r
esu
lts
but
co
u
ld
not
ca
p
tu
r
e
Ay
u
r
v
ed
ic
o
n
to
lo
g
ies.
I
n
teg
r
a
tio
n
of
KG
with
r
elatio
n
m
ap
p
in
g
[
3
2
]
en
h
an
ce
d
p
er
f
o
r
m
an
ce
by
e
n
f
o
r
cin
g
g
r
ap
h
-
b
ased
co
n
s
is
ten
cy
.
T
h
e
p
r
o
p
o
s
ed
Hy
b
r
id
KG
-
QA
s
y
s
tem
ac
h
iev
ed
th
e
h
ig
h
est
F1
(
0
.
8
7
)
,
d
em
o
n
s
tr
atin
g
th
at
KG
s
co
m
b
in
e
d
with
lan
g
u
ag
e
m
o
d
els
s
ig
n
if
ican
tly
im
p
r
o
v
e
r
elatio
n
ac
cu
r
ac
y
,
alig
n
in
g
with
r
esu
lts
in
h
ea
lth
ca
r
e
KG
r
esear
c
h
[
3
3
]
.
T
ab
le
2.
R
elatio
n
ex
tr
ac
tio
n
r
e
s
u
lts
M
o
d
e
l
P
r
e
c
i
s
i
o
n
R
e
c
a
l
l
F1
-
sc
o
r
e
R
e
l
a
t
i
o
n
s
H
a
n
d
l
e
d
Tr
a
n
sf
o
r
mer
-
o
n
l
y
QA
0
.
7
2
0
.
7
0
0
.
7
1
V
y
a
d
h
i
–
La
k
sh
a
n
a
B
i
o
B
E
R
T
(
b
a
sel
i
n
e
)
0
.
7
6
0
.
7
3
0
.
7
4
V
y
a
d
h
i
–
La
k
sh
a
n
a
,
V
y
a
d
h
i
–
D
r
a
v
y
a
KG
+
R
e
l
a
t
i
o
n
M
a
p
p
i
n
g
0
.
8
1
0
.
7
7
0
.
7
9
V
y
a
d
h
i
–
La
k
sh
a
n
a
,
D
r
a
v
y
a
–
D
o
sh
a
P
r
o
p
o
se
d
H
y
b
r
i
d
KG
-
QA
0
.
8
8
0
.
8
5
0
.
8
7
A
l
l
R
e
l
a
t
i
o
n
s
Fig
u
r
e
3
co
m
p
ar
es
r
elatio
n
e
x
tr
ac
tio
n
p
e
r
f
o
r
m
an
ce
ac
r
o
s
s
m
o
d
els.
T
h
e
p
r
o
p
o
s
ed
h
y
b
r
i
d
KG
-
QA
s
y
s
tem
ac
h
iev
ed
s
u
p
er
io
r
ac
cu
r
ac
y
,
p
ar
ticu
lar
l
y
in
m
u
lti
-
r
elatio
n
al
q
u
er
ies
(
e
.
g
.
,
Vy
ad
h
i
-
Dr
av
y
a
-
Do
s
h
a
ch
ain
s
)
.
B
io
B
E
R
T
s
h
o
wed
m
o
d
er
ate
g
ain
s
,
b
u
t
with
o
u
t
KG
co
n
s
tr
ain
ts
,
it
p
r
o
d
u
ce
d
s
em
an
tic
d
r
if
t.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
20
2
6
:
1
1
9
7
-
1
2
0
7
1202
KG
in
teg
r
atio
n
p
r
e
v
en
ts
s
p
u
r
i
o
u
s
r
elatio
n
s
by
en
f
o
r
cin
g
Ay
u
r
v
ed
ic
s
tr
u
ctu
r
al
r
u
les.
T
h
is
p
r
o
v
es
th
at
h
y
b
r
id
s
y
m
b
o
lic
–
n
eu
r
al
s
y
s
tem
s
o
u
t
p
er
f
o
r
m
p
u
r
ely
n
eu
r
al
QA
s
y
s
tem
s
f
o
r
d
o
m
ain
-
s
p
ec
if
ic
t
ex
t,
co
n
s
is
ten
t
with
m
ed
ical
QA
f
in
d
in
g
s
[
3
4
]
.
Fig
u
r
e
3
.
R
elatio
n
ex
tr
ac
ti
o
n
a
cc
u
r
ac
y
ac
r
o
s
s
m
o
d
els
T
ab
le
3
s
h
o
ws
QA
ac
cu
r
ac
y
in
b
o
th
San
s
k
r
it
an
d
E
n
g
lis
h
q
u
er
ies.
Su
b
g
r
ap
h
QA
ac
h
iev
ed
lo
w
ac
cu
r
ac
y
due
to
lim
ited
r
e
aso
n
in
g
.
T
r
an
s
f
o
r
m
er
QA
i
m
p
r
o
v
e
d
s
lig
h
tly
b
u
t
lack
e
d
in
ter
p
r
etab
ilit
y
.
GR
E
ASEL
M
p
r
o
v
id
ed
s
tr
o
n
g
er
r
ea
s
o
n
in
g
,
alig
n
in
g
with
pr
io
r
wo
r
k
in
m
u
lti
-
hop
QA.
T
h
e
p
r
o
p
o
s
ed
h
y
b
r
id
QA
s
y
s
tem
s
ig
n
if
ican
tly
o
u
tp
er
f
o
r
m
e
d
all
b
aselin
es,
r
ea
ch
in
g
87%
(
San
s
k
r
it)
an
d
90%
(
E
n
g
lis
h
)
ac
cu
r
ac
y
,
with
h
ig
h
er
MRR
an
d
o
v
er
la
p
s
co
r
es.
T
h
is
h
ig
h
lig
h
ts
th
e
b
en
ef
it
of
KG
in
teg
r
atio
n
w
ith
n
eu
r
al
m
o
d
els,
co
n
f
ir
m
in
g
o
b
s
er
v
atio
n
s
f
r
o
m
h
ea
lth
ca
r
e
QA.
T
ab
le
3.
QA
task
ac
cu
r
ac
y
(
Sa
n
s
k
r
it
an
d
E
n
g
lis
h
q
u
e
r
ies)
M
o
d
e
l
A
c
c
u
r
a
c
y
(
S
a
n
sk
r
i
t
)
A
c
c
u
r
a
c
y
(
E
n
g
l
i
s
h
)
M
R
R
O
v
e
r
l
a
p
C
o
e
f
f
i
c
i
e
n
t
S
u
b
g
r
a
p
h
QA
0
.
6
8
0
.
7
1
0
.
5
2
0
.
6
1
Tr
a
n
sf
o
r
mer
QA
0
.
7
2
0
.
7
6
0
.
5
6
0
.
6
6
G
R
EA
S
EL
M
0
.
8
0
0
.
8
3
0
.
6
8
0
.
7
4
P
r
o
p
o
se
d
H
y
b
r
i
d
QA
0
.
8
7
0
.
9
0
0
.
7
5
0
.
8
1
Fig
u
r
e
4
h
ig
h
li
g
h
ts
ac
cu
r
ac
y
d
if
f
er
en
ce
s
ac
r
o
s
s
q
u
er
y
lan
g
u
ag
es.
W
h
ile
b
aselin
e
m
o
d
els
p
er
f
o
r
m
ed
b
etter
on
E
n
g
lis
h
q
u
er
ies,
th
ey
s
tr
u
g
g
led
with
San
s
k
r
it
d
u
e
to
m
o
r
p
h
o
lo
g
ical
co
m
p
le
x
ity
.
T
h
e
p
r
o
p
o
s
ed
h
y
b
r
id
s
y
s
tem
p
e
r
f
o
r
m
ed
co
n
s
is
ten
tly
ac
r
o
s
s
b
o
th
lan
g
u
a
g
es,
v
alid
atin
g
its
ab
ilit
y
to
h
an
d
le
co
m
p
lex
San
s
k
r
i
t
m
o
r
p
h
o
lo
g
y
u
s
in
g
B
y
T
5
-
San
s
k
r
it
[
3
5
]
.
T
h
is
b
ilin
g
u
al
r
o
b
u
s
tn
ess
is
cr
u
cial
f
o
r
p
r
ac
tical
u
s
e
am
o
n
g
A
y
u
r
v
ed
ic
s
ch
o
lar
s
an
d
m
o
d
er
n
clin
ician
s
.
T
h
ese
r
esu
lts
alig
n
with
e
ar
lier
claim
s
th
at
d
o
m
ain
-
s
p
ec
if
ic
NL
P
p
ip
elin
es
ar
e
ess
en
tial
f
o
r
class
ical
tex
t
p
r
o
ce
s
s
in
g
.
Fig
u
r
e
4
.
QA
ac
cu
r
ac
y
co
m
p
a
r
is
o
n
(
San
s
k
r
it
vs
E
n
g
lis
h
q
u
e
r
ies)
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
A
n
A
I
-
p
o
w
ered
kn
o
w
led
g
e
g
r
a
p
h
-
b
a
s
ed
q
u
esti
o
n
a
n
s
w
er
in
g
s
ystem
fo
r
C
h
a
r
a
k
…
(
S
h
a
r
a
y
u
Mir
a
s
d
a
r
)
1203
Fig
u
r
e
5
s
h
o
ws
th
e
e
f
f
icien
cy
-
ac
cu
r
ac
y
t
r
ad
eo
f
f
.
Su
b
g
r
ap
h
s
ea
r
ch
in
cu
r
r
e
d
h
ig
h
d
ela
y
s
,
u
n
s
u
itab
le
f
o
r
r
ea
l
-
wo
r
ld
QA.
GAT
an
d
GR
E
ASEL
M
b
alan
ce
d
r
ea
s
o
n
in
g
d
ep
t
h
an
d
ef
f
icien
cy
.
T
h
e
p
r
o
p
o
s
ed
h
y
b
r
id
GDS
ac
h
iev
ed
th
e
lo
west
r
esp
o
n
s
e
tim
e
with
h
ig
h
est
ac
cu
r
a
cy
,
th
an
k
s
to
o
p
tim
ized
g
r
ap
h
s
im
ilar
ity
m
eth
o
d
s
.
T
h
ese
r
esu
lts
alig
n
with
f
in
d
in
g
s
th
at
g
r
ap
h
alg
o
r
ith
m
s
can
a
cc
eler
ate
QA
in
b
io
m
ed
ical
c
o
n
tex
ts
.
Fig
u
r
e
5
.
R
esp
o
n
s
e
tim
e
vs
p
a
th
ac
cu
r
ac
y
ac
r
o
s
s
alg
o
r
ith
m
s
T
ab
le
4
co
m
p
a
r
es
o
u
r
s
y
s
tem
with
s
tate
-
of
-
th
e
-
ar
t
h
ea
lth
ca
r
e
QA
s
y
s
tem
s
.
W
h
ile
E
HR
-
b
ased
KGs
an
d
b
io
m
ed
ical
QA
d
em
o
n
s
tr
ate
s
tr
o
n
g
p
e
r
f
o
r
m
an
ce
,
t
h
e
p
r
o
p
o
s
ed
C
h
ar
ak
KG
-
QA
o
u
tp
er
f
o
r
m
s
th
em
in
F1
-
s
co
r
e
(
0
.
9
0
)
an
d
in
ter
p
r
et
ab
ilit
y
.
T
h
is
v
alid
ates
th
e
s
y
s
tem
’
s
co
n
tr
ib
u
tio
n
to
cu
ltu
r
ally
g
r
o
u
n
d
e
d
y
et
co
m
p
u
tatio
n
ally
m
o
d
e
r
n
QA
s
y
s
tem
s
.
T
ab
le
4
.
C
o
m
p
a
r
ativ
e
r
esu
lts
with
h
ea
lth
ca
r
e
QA
s
y
s
tem
s
S
y
st
e
m
D
o
ma
i
n
F1
-
sc
o
r
e
M
R
R
I
n
t
e
r
p
r
e
t
a
b
i
l
i
t
y
KG
-
QA
(
M
e
d
i
c
a
l
EH
R
)
C
l
i
n
i
c
a
l
0
.
8
3
0
.
7
1
M
e
d
i
u
m
D
R
.
K
N
O
W
S
(
LL
M
+
K
G
)
B
i
o
me
d
i
c
a
l
0
.
8
5
0
.
7
3
H
i
g
h
A
d
a
p
t
i
v
e
KG
-
QA
M
e
d
i
c
a
l
AI
0
.
8
6
0
.
7
4
H
i
g
h
P
r
o
p
o
se
d
C
h
a
r
a
k
KG
-
QA
A
y
u
r
v
e
d
a
0
.
9
0
0
.
7
5
V
e
r
y
H
i
g
h
7.
QA
P
E
RF
O
RM
AN
CE
E
VA
L
UA
T
I
O
N
T
h
e
AI
-
p
o
wer
ed
QA
s
y
s
tem
was
test
ed
u
s
in
g
300
b
en
c
h
m
ar
k
q
u
esti
o
n
s
f
r
o
m
C
h
a
r
a
k
S
a
mh
ita
an
n
o
tated
co
r
p
u
s
.
Qu
esti
o
n
s
wer
e
ca
teg
o
r
ize
d
in
t
o
f
ac
t
o
id
,
ca
u
s
al,
an
d
r
ec
o
m
m
e
n
d
atio
n
q
u
er
ies.
E
v
al
u
atio
n
was
p
er
f
o
r
m
ed
u
s
in
g
B
io
B
E
R
T
(
f
in
e
-
tu
n
ed
on
San
s
k
r
it
co
r
p
u
s
)
+
Neo
4
j
tr
av
e
r
s
al
r
ea
s
o
n
in
g
.
T
ab
le
5
e
v
alu
ates
th
e
QA
s
y
s
tem
ac
r
o
s
s
th
r
ee
q
u
esti
o
n
c
ateg
o
r
ies.
Facto
id
q
u
er
ies
ac
h
iev
ed
th
e
h
ig
h
est
ac
cu
r
ac
y
(
9
1
.
3
%)
s
in
ce
an
s
wer
s
ar
e
o
f
ten
d
ir
ec
t
m
ap
p
in
g
s
f
r
o
m
V
ya
d
h
i
–
La
ks
h
a
n
a
r
elatio
n
s
h
ip
s
.
C
au
s
al
q
u
er
ies
s
h
o
wed
s
lig
h
t
ly
r
ed
u
ce
d
ac
cu
r
ac
y
(
8
7
.
6
%),
p
r
im
ar
ily
due
to
m
u
lti
-
hop
r
ea
s
o
n
in
g
er
r
o
r
s
in
Do
s
h
a
in
f
er
en
ce
.
R
ec
o
m
m
en
d
atio
n
q
u
er
ies,
wh
ich
r
eq
u
i
r
e
co
m
b
in
i
n
g
m
u
ltip
le
h
o
p
s
(
d
i
s
ea
s
e
→
Do
s
h
a
→
h
erb
),
ac
h
ie
v
ed
t
h
e
lo
west
ac
cu
r
ac
y
(
8
4
.
1
%).
Ho
wev
e
r
,
th
is
p
er
f
o
r
m
an
ce
is
s
till
h
ig
h
er
t
h
an
b
aselin
e
k
ey
wo
r
d
-
b
ased
QA
s
y
s
tem
s
,
wh
ich
ty
p
ically
ac
h
iev
e
b
el
o
w
75%
on
co
m
p
le
x
r
ea
s
o
n
in
g
task
s
[
3
6
]
.
Av
er
ag
e
s
y
s
tem
r
esp
o
n
s
e
tim
e
was
134
m
s
,
p
r
o
v
i
n
g
th
e
f
r
am
ewo
r
k
’
s
s
ca
lab
ilit
y
f
o
r
r
ea
l
-
tim
e
ap
p
licatio
n
s
.
T
h
e
o
v
e
r
all
F1
-
s
co
r
e
(
0
.
8
6
)
co
n
f
ir
m
s
s
tr
o
n
g
b
alan
c
e
b
etwe
en
p
r
ec
is
io
n
an
d
r
ec
all.
C
o
m
p
ar
e
d
to
b
io
m
ed
ical
QA
s
y
s
tem
s
lik
e
Pu
b
Me
d
KGQA,
o
u
r
s
y
s
t
em
d
em
o
n
s
tr
ates
s
u
p
er
io
r
r
ea
s
o
n
in
g
p
e
r
f
o
r
m
an
ce
due
to
A
y
u
r
v
e
d
ic
o
n
t
o
lo
g
y
-
d
r
iv
en
in
te
g
r
atio
n
.
T
ab
le
5
.
QA
ac
cu
r
ac
y
ac
r
o
s
s
q
u
esti
o
n
ty
p
es
Q
u
e
st
i
o
n
t
y
p
e
Ex
a
m
p
l
e
s
A
c
c
u
r
a
c
y
(
%)
F1
-
sc
o
r
e
A
v
g
.
R
e
s
p
o
n
se
Ti
me
(
ms)
F
a
c
t
o
i
d
“
W
h
a
t
a
r
e
t
h
e
s
y
m
p
t
o
ms
of
र
ा
जय
ा
(
t
u
b
e
r
c
u
l
o
s
i
s)
?
”
9
1
.
3
0
.
8
9
1
2
0
C
a
u
sa
l
“
W
h
i
c
h
D
o
sh
a
i
mb
a
l
a
n
c
e
c
a
u
ses
ा
स
(
a
st
h
ma)?
”
8
7
.
6
0
.
8
6
1
3
5
R
e
c
o
mm
e
n
d
a
t
i
o
n
“
W
h
i
c
h
h
e
r
b
t
r
e
a
t
s
र
w
i
t
h
P
i
t
t
a
i
mb
a
l
a
n
c
e
?
”
8
4
.
1
0
.
8
3
1
4
8
O
v
e
r
a
l
l
-
8
7
.
7
0
.
8
6
1
3
4
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
20
2
6
:
1
1
9
7
-
1
2
0
7
1204
T
h
e
p
r
ac
tical
u
tili
ty
of
th
e
s
y
s
tem
was
ev
alu
ated
o
v
e
r
th
e
g
en
er
ated
Ay
u
r
v
ed
ic
KG
a
n
d
th
e
QA
s
y
s
tem
.
A
s
tr
u
ctu
r
ed
v
alid
atio
n
was
p
er
f
o
r
m
e
d
with
th
e
ex
p
er
t
an
d
ce
r
tifie
d
Vaid
y
as
h
av
in
g
8
-
20
y
ea
r
s
of
clin
ical
ex
p
er
ien
ce
.
1
0
0
p
lu
s
r
ep
r
esen
tativ
e
q
u
er
ies
we
r
e
cu
r
ated
wh
ich
in
clu
d
e
d
Vy
ād
h
i,
L
a
k
ṣ
aṇ
a,
Do
ṣ
a
im
b
alan
ce
,
Dr
av
y
a,
R
as
a,
Gu
ṇ
a,
an
d
Yo
g
a
f
o
r
m
u
lati
o
n
s
.
T
h
e
s
y
s
tem
s
’
r
esp
o
n
s
e
w
as
v
alid
ated
by
th
e
ex
p
er
ts
b
ased
on
p
ar
am
eter
s
s
u
ch
as
clin
ical
r
elev
an
ce
,
co
m
p
leten
ess
,
alig
n
m
en
t
w
ith
class
ical
tex
t
s
,
an
d
p
r
ac
tical
tr
ea
tm
e
n
t
v
ali
d
ity
.
T
h
e
a
v
er
ag
e
e
x
p
er
t
s
a
tis
f
ac
tio
n
s
co
r
e
was
4
.
3
8
/5
,
in
d
icatin
g
s
tr
o
n
g
ag
r
ee
m
en
t
on
th
e
co
r
r
ec
tn
ess
an
d
u
s
ef
u
ln
ess
of
th
e
an
s
wer
s
.
Feed
b
ac
k
co
llected
f
r
o
m
Vai
d
y
as
also
h
elp
ed
to
r
ef
in
e
am
b
ig
u
o
u
s
m
ap
p
i
n
g
s
(
e.
g
.
,
s
y
m
p
to
m
clu
s
ter
s
)
an
d
r
eso
lv
e
s
y
n
o
n
y
m
v
ar
iati
o
n
s
.
T
h
is
v
alid
atio
n
co
n
f
ir
m
s
th
at
th
e
s
y
s
tem
is
not
o
n
ly
c
o
m
p
u
tatio
n
ally
ac
cu
r
ate
but
also
clin
ically
m
ea
n
in
g
f
u
l
f
o
r
r
ea
l
-
wo
r
ld
Ay
u
r
v
ed
ic
d
ec
is
io
n
s
u
p
p
o
r
t.
8.
CO
M
P
ARA
T
I
V
E
S
T
UDY
A
ND
I
M
P
RO
V
E
M
E
N
T
S
To
v
alid
ate
ef
f
ec
tiv
en
ess
,
we
co
m
p
ar
ed
o
u
r
AI
-
p
o
we
r
ed
KG
-
QA
s
y
s
tem
w
ith
a
b
aselin
e
k
ey
wo
r
d
-
m
atch
in
g
San
s
k
r
it
QA
s
y
s
tem
ad
ap
ted
f
r
o
m
p
r
io
r
wo
r
k
s
.
E
v
alu
atio
n
u
s
ed
th
e
s
a
m
e
300
test
q
u
er
ies.
T
ab
le
6
d
em
o
n
s
tr
ates
th
e
s
ig
n
if
ican
t
im
p
r
o
v
em
e
n
ts
ac
h
iev
ed
by
our
s
y
s
tem
co
m
p
ar
ed
to
a
b
aselin
e
k
ey
wo
r
d
-
b
ased
San
s
k
r
it
QA
s
y
s
tem
.
Acc
u
r
ac
y
im
p
r
o
v
e
d
f
r
o
m
7
2
.
4
%
to
8
7
.
7
%,
m
ain
l
y
due
to
th
e
s
em
an
tic
u
n
d
er
s
tan
d
i
n
g
of
San
s
k
r
it
tex
t
v
ia
B
io
B
E
R
T
f
in
e
-
tu
n
in
g
an
d
o
n
to
lo
g
y
-
d
r
iv
en
g
r
a
p
h
r
ea
s
o
n
in
g
.
T
h
e
F1
-
s
co
r
e
in
cr
ea
s
ed
f
r
o
m
0
.
7
1
to
0
.
8
6
,
i
n
d
icatin
g
a
b
alan
ce
d
g
ai
n
in
b
o
th
p
r
ec
is
io
n
a
n
d
r
ec
all.
Sy
s
tem
ef
f
icien
cy
also
im
p
r
o
v
e
d
,
with
av
er
a
g
e
r
esp
o
n
s
e
tim
e
r
e
d
u
ce
d
f
r
o
m
2
1
0
ms
to
1
3
4
m
s
,
o
win
g
to
o
p
tim
i
ze
d
Neo
4
j
tr
av
e
r
s
al
q
u
er
ies.
T
h
e
s
u
cc
ess
r
ate
f
o
r
m
u
lti
-
h
o
p
q
u
er
ies
r
o
s
e
s
u
b
s
tan
tially
(
+3
4
.
7
%),
d
em
o
n
s
tr
atin
g
th
e
s
y
s
tem
’
s
s
u
p
er
io
r
ity
in
h
an
d
lin
g
c
o
m
p
l
ex
r
ea
s
o
n
i
n
g
in
v
o
l
v
in
g
d
is
ea
s
es,
Do
s
h
as,
an
d
h
er
b
s
.
T
h
is
was
f
u
r
th
er
v
e
r
if
ied
with
th
e
h
elp
of
u
s
er
s
atis
f
ac
tio
n
s
u
r
v
ey
s
wh
ich
s
h
o
wed
im
p
r
o
v
ed
r
atin
g
s
f
r
o
m
3
.
2
/5
to
4
.
6
/5
.
T
h
ese
im
p
r
o
v
em
en
ts
h
ig
h
lig
h
t
th
e
s
y
n
er
g
y
of
San
s
k
r
it
NL
P
an
d
KG
in
te
g
r
atio
n
,
o
f
f
er
in
g
a
r
o
b
u
s
t
alter
n
ativ
e
to
co
n
v
e
n
tio
n
al
r
u
le
-
b
ased
QA
s
y
s
tem
s
.
T
ab
le
6
.
C
o
m
p
a
r
ativ
e
QA
p
er
f
o
r
m
an
ce
f
o
r
p
r
o
p
o
s
ed
V
s.
b
as
elin
e
s
y
s
tem
M
e
t
r
i
c
B
a
se
l
i
n
e
QA
(
K
e
y
w
o
r
d
-
b
a
s
e
d
)
O
u
r
S
y
s
t
e
m
(
K
G
+
B
i
o
B
ER
T)
I
mp
r
o
v
e
m
e
n
t
(
%)
A
c
c
u
r
a
c
y
(
%)
7
2
.
4
8
7
.
7
+
2
1
.
1
F1
-
sc
o
r
e
0
.
7
1
0
.
8
6
+
2
1
.
1
A
v
g
.
R
e
s
p
o
n
se
Ti
me
(
ms)
2
1
0
1
3
4
-
3
6
.
2
C
o
m
p
l
e
x
Q
u
e
r
y
S
u
c
c
e
s
s
R
a
t
e
(
%)
6
2
.
0
8
3
.
5
+
3
4
.
7
U
ser
S
a
t
i
sf
a
c
t
i
o
n
(
S
u
r
v
e
y
,
/
5
)
3
.
2
4
.
6
+
4
3
.
7
T
h
e
p
r
o
p
o
s
ed
AI
-
p
o
wer
ed
KG
-
QA
s
y
s
tem
d
em
o
n
s
tr
ates
how
San
s
k
r
it
NL
P
an
d
GDS
can
be
s
y
n
er
g
is
tically
co
m
b
in
ed
to
a
d
d
r
ess
th
e
ch
allen
g
es
of
r
etr
i
ev
in
g
Ay
u
r
v
ed
ic
k
n
o
wled
g
e
f
r
o
m
an
cie
n
t
tex
ts
s
u
ch
as
th
e
C
h
a
r
a
k
S
a
mh
it
a
.
C
o
m
p
ar
e
d
to
c
o
n
v
e
n
tio
n
al
k
ey
wo
r
d
-
b
ased
QA,
our
ap
p
r
o
ac
h
ex
h
ib
its
s
u
b
s
tan
tial
im
p
r
o
v
em
en
ts
in
ac
cu
r
ac
y
,
s
em
an
tic
u
n
d
e
r
s
tan
d
in
g
,
an
d
q
u
er
y
r
esp
o
n
s
e
ef
f
i
cien
cy
.
T
h
e
r
esu
lts
co
n
f
ir
m
th
at
d
o
m
ain
-
s
p
ec
if
ic
B
io
B
E
R
T
f
in
e
-
tu
n
in
g
ca
p
tu
r
e
s
co
n
tex
tu
al
m
ea
n
in
g
s
of
San
s
k
r
it
m
ed
ical
ter
m
s
an
d
clar
if
ies
s
y
n
o
n
y
m
s
.
In
th
e
KG
,
f
o
r
e
n
tity
n
o
d
es
(
Do
s
h
as
,
Vy
ad
h
i,
L
ak
s
h
an
,
Her
b
s
)
wh
ich
ar
e
n
o
t
d
ir
ec
tly
co
n
n
ec
ted
with
r
elatio
n
s
h
ip
e
d
g
e,
m
u
ti
-
h
o
p
p
i
n
g
is
r
eq
u
ir
ed
d
u
r
in
g
tr
a
v
er
s
al.
Neo
4
j
KG
allo
ws
us
to
ca
p
tu
r
e
th
is
ef
f
ec
tiv
ely
.
I
m
p
o
r
tan
tly
,
th
e
s
y
s
tem
en
h
an
ce
s
p
r
ac
tical
u
tili
ty
f
o
r
m
o
d
er
n
p
r
ac
titi
o
n
er
s
by
c
o
n
v
e
r
tin
g
class
ical
San
s
k
r
it
v
er
s
es
in
to
s
tr
u
ctu
r
ed
an
d
q
u
er
y
ab
le
k
n
o
wled
g
e.
Few
ch
allen
g
es
s
till
n
ee
d
to
be
ad
d
r
ess
ed
lik
e
s
ca
lin
g
th
e
s
y
s
tem
with
l
ar
g
er
San
s
k
r
it
co
r
p
o
r
a,
r
ar
e
e
n
titi
es,
an
d
d
ialec
tal
v
ar
iatio
n
s
.
T
h
e
QA
s
y
s
tem
will
be
more
r
o
b
u
s
t
af
ter
a
d
d
r
ess
in
g
th
ese
ch
allen
g
es.
9.
CO
NCLU
SI
O
N
T
h
is
s
tu
d
y
d
is
cu
s
s
es
th
e
AI
-
p
o
wer
ed
KG
-
b
ased
QA
s
y
s
tem
f
o
r
Ay
u
r
v
ed
ic
tr
ea
tis
e
‘
C
h
ar
a
k
Sam
h
ita’
wh
er
e
San
s
k
r
it
NL
P
a
n
d
Gr
a
p
h
d
ata
s
cien
ce
tech
n
iq
u
es
ar
e
in
teg
r
ated
.
In
th
e
p
r
o
p
o
s
ed
s
y
s
tem
,
u
n
s
tr
u
ctu
r
e
d
San
s
k
r
it
v
er
s
es
ar
e
tr
an
s
f
o
r
m
ed
in
to
a
s
tr
u
ctu
r
ed
Neo
4
j
KG
wh
ich
h
elp
s
to
c
o
n
n
ec
t
Ay
u
r
v
ed
ic
k
n
o
wled
g
e
an
d
m
o
d
er
n
co
m
p
u
tatio
n
al
m
e
th
o
d
s
.
By
f
in
e
-
tu
n
in
g
B
io
B
E
R
T
on
A
y
u
r
v
ed
a
-
s
p
ec
if
ic
d
atasets
,
th
e
s
y
s
tem
was
ab
le
to
ac
h
iev
e
en
h
a
n
ce
d
NE
R
f
o
r
c
o
r
e
Ay
u
r
v
ed
ic
c
o
n
ce
p
ts
s
u
ch
as
V
y
a
d
h
i
(
d
i
s
ea
s
es),
La
ksh
a
n
a
(
s
y
m
p
to
m
s
)
,
Dra
vy
a
(
h
er
b
s
)
,
R
a
s
a
,
Gu
n
a
,
an
d
K
a
r
ma
.
Me
asu
r
es
of
p
r
ec
is
io
n
,
r
ec
all,
a
n
d
F1
-
s
co
r
e
s
h
o
wed
s
ig
n
if
ican
t
im
p
r
o
v
em
en
t
as
co
m
p
ar
ed
to
t
r
ad
itio
n
al
r
u
le
-
b
as
ed
an
d
k
ey
wo
r
d
-
d
r
i
v
en
ap
p
r
o
a
ch
es.
T
h
e
in
clu
s
io
n
of
g
r
ap
h
em
b
ed
d
in
g
s
an
d
tr
a
v
er
s
al
alg
o
r
ith
m
s
en
ab
led
ef
f
ec
tiv
e
h
an
d
lin
g
of
m
u
lti
-
h
o
p
r
ea
s
o
n
in
g
q
u
e
r
ies,
p
r
o
v
id
i
n
g
not
o
n
ly
d
ir
ec
t
an
s
wer
s
but
also
e
x
p
lan
ato
r
y
co
n
tex
t.
Use
of
th
e
p
r
o
p
o
s
ed
s
y
s
tem
can
be
s
ee
n
at
m
u
ltip
le
lev
els
s
u
ch
as
p
e
r
s
o
n
alize
d
tr
ea
tm
en
t
f
o
r
th
e
p
ati
en
ts
,
clin
ical
d
ec
is
io
n
s
u
p
p
o
r
t,
ed
u
ca
tio
n
al
u
s
e
Evaluation Warning : The document was created with Spire.PDF for Python.
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
I
SS
N:
2252
-
8
7
7
6
A
n
A
I
-
p
o
w
ered
kn
o
w
led
g
e
g
r
a
p
h
-
b
a
s
ed
q
u
esti
o
n
a
n
s
w
er
in
g
s
ystem
fo
r
C
h
a
r
a
k
…
(
S
h
a
r
a
y
u
Mir
a
s
d
a
r
)
1205
ca
s
es
f
o
r
Ay
u
r
v
ed
a
r
esear
c
h
er
s
an
d
p
r
ac
titi
o
n
er
s
.
T
h
e
p
r
o
p
o
s
ed
ap
p
r
o
ac
h
h
ig
h
lig
h
t
s
th
e
p
o
ten
tial
of
in
teg
r
atin
g
d
ee
p
lear
n
in
g
wit
h
g
r
ap
h
-
b
ased
r
ea
s
o
n
in
g
f
o
r
ad
v
an
cin
g
d
ig
ital
h
u
m
an
ities
an
d
co
m
p
u
tatio
n
al
Ay
u
r
v
ed
a
.
Ov
er
all,
th
is
r
esear
ch
co
n
tr
ib
u
tes
to
war
d
s
cr
ea
tin
g
n
ex
t
g
en
e
r
atio
n
h
ea
lth
ca
r
e
k
n
o
wled
g
e
s
y
s
tem
s
,
wh
ile
p
r
eser
v
in
g
t
h
e
s
em
an
tic
r
ich
n
ess
of
San
s
k
r
it
test
an
d
al
s
o
en
ab
lin
g
I
n
tellig
en
t
q
u
er
y
in
g
.
10.
F
UT
UR
E
WO
RK
Alth
o
u
g
h
th
e
p
r
o
p
o
s
ed
s
y
s
tem
h
as
d
em
o
n
s
tr
ated
p
r
o
m
is
in
g
r
esu
lts
,
s
ev
er
al
av
en
u
es
f
o
r
im
p
r
o
v
em
e
n
t
r
em
ain
.
Firstl
y
,
ex
p
an
d
i
n
g
th
e
tr
ain
in
g
d
ataset
with
lar
g
er
San
s
k
r
it
c
o
r
p
o
r
a,
in
clu
d
in
g
S
u
s
h
r
u
ta
S
a
mh
ita
an
d
A
s
h
ta
n
g
a
Hri
d
a
y
a
,
will
en
h
an
ce
th
e
s
y
s
tem
’
s
co
v
er
ag
e
an
d
r
o
b
u
s
tn
ess
.
Ad
d
itio
n
ally
,
ap
p
ly
i
n
g
ac
tiv
e
lear
n
in
g
an
d
s
em
i
-
s
u
p
e
r
v
is
ed
an
n
o
tati
on
m
eth
o
d
s
co
u
ld
r
e
d
u
ce
th
e
d
ep
en
d
en
cy
on
m
an
u
ally
lab
elle
d
Ay
u
r
v
ed
ic
d
atasets
.
Fro
m
a
tech
n
ical
p
er
s
p
ec
tiv
e,
f
u
tu
r
e
wo
r
k
will
f
o
cu
s
on
m
u
lti
-
lin
g
u
al
in
teg
r
atio
n
,
wh
er
e
San
s
k
r
it
co
n
ce
p
ts
ar
e
alig
n
e
d
with
m
o
d
er
n
b
io
m
ed
ical
o
n
to
lo
g
ies
(
e.
g
.
,
UM
L
S,
Me
SH)
to
s
u
p
p
o
r
t
cr
o
s
s
-
d
o
m
ain
r
ea
s
o
n
in
g
.
I
n
c
o
r
p
o
r
atin
g
tr
an
s
f
o
r
m
er
-
b
ased
GNNs
co
u
ld
f
u
r
th
er
im
p
r
o
v
e
lin
k
p
r
ed
ictio
n
,
d
r
u
g
r
ep
u
r
p
o
s
in
g
,
an
d
p
e
r
s
o
n
alize
d
tr
ea
tm
e
n
t
r
e
co
m
m
en
d
atio
n
s
.
An
o
th
e
r
p
r
o
m
is
in
g
d
ir
ec
tio
n
in
v
o
l
v
es
lev
er
ag
in
g
e
x
p
lain
ab
le
AI
(
XAI
)
tech
n
iq
u
es
to
m
ak
e
th
e
s
y
s
tem
’
s
r
ec
o
m
m
en
d
ati
o
n
s
m
o
r
e
in
ter
p
r
etab
le
to
p
r
a
ctitio
n
er
s
,
en
s
u
r
in
g
tr
u
s
t
an
d
a
d
o
p
tio
n
in
clin
ic
al
s
ettin
g
s
.
Fu
r
th
er
m
o
r
e
,
th
e
QA
f
r
am
ewo
r
k
can
be
e
x
ten
d
ed
to
s
u
p
p
o
r
t
v
o
ice
-
b
ased
q
u
er
ies
in
San
s
k
r
it
an
d
r
e
g
io
n
al
I
n
d
ia
n
lan
g
u
ag
es,
en
h
an
cin
g
ac
ce
s
s
ib
ilit
y
f
o
r
tr
ad
itio
n
al
h
ea
ler
s
an
d
s
tu
d
e
n
ts
.
C
o
llab
o
r
ativ
e
p
latf
o
r
m
s
c
o
u
ld
also
allo
w
Ay
u
r
v
ed
a
s
ch
o
lar
s
to
c
o
n
tr
ib
u
te
an
n
o
tatio
n
s
an
d
co
r
r
ec
tio
n
s
,
th
e
r
eb
y
im
p
r
o
v
in
g
s
y
s
tem
ac
cu
r
ac
y
o
v
er
tim
e.
In
s
u
m
m
ar
y
,
wh
ile
th
is
s
tu
d
y
r
ep
r
es
en
ts
a
f
o
u
n
d
atio
n
al
s
tep
to
war
d
AI
-
d
r
i
v
en
Ay
u
r
v
ed
ic
k
n
o
wled
g
e
s
y
s
tem
s
,
f
u
tu
r
e
ef
f
o
r
ts
will
em
p
h
asize
s
ca
lab
ilit
y
,
in
ter
o
p
er
ab
ilit
y
,
an
d
r
ea
l
-
wo
r
ld
d
e
p
l
o
y
m
en
t
to
m
ax
im
ize
im
p
ac
t
in
b
o
th
r
esear
ch
an
d
clin
ical
p
r
ac
tice.
ACK
NO
WL
E
DG
M
E
N
T
B
o
th
th
e
au
th
o
r
s
h
av
e
co
n
tr
ib
u
ted
eq
u
ally
to
th
e
wr
itin
g
,
r
e
v
iew
an
d
r
ev
is
io
n
o
f
th
e
m
an
u
s
cr
ip
t
an
d
h
av
e
r
ea
d
an
d
a
p
p
r
o
v
ed
th
e
f
i
n
al
v
er
s
io
n
.
F
UNDING
I
NF
O
R
M
A
T
I
O
N
N
o
f
u
n
d
in
g
was r
ec
eiv
e
d
f
o
r
c
o
n
d
u
ctin
g
th
is
s
tu
d
y
.
AUTHO
R
CO
NT
RI
B
UT
I
O
NS
ST
A
T
E
M
E
N
T
T
h
is
jo
u
r
n
al
u
s
es
th
e
C
o
n
t
r
ib
u
to
r
R
o
les
T
a
x
o
n
o
m
y
(
C
R
ed
iT)
to
r
ec
o
g
n
ize
in
d
iv
i
d
u
al
au
th
o
r
co
n
tr
ib
u
tio
n
s
,
r
ed
u
ce
au
th
o
r
s
h
ip
d
is
p
u
tes,
an
d
f
ac
ilit
ate
co
llab
o
r
atio
n
.
Na
m
e
of
Aut
ho
r
C
M
So
Va
Fo
I
R
D
O
E
Vi
Su
P
Fu
Sh
ar
ay
u
Mir
asd
ar
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
✓
Ma
n
g
esh
B
ed
ek
ar
✓
✓
✓
✓
✓
✓
✓
C
:
C
o
n
c
e
p
t
u
a
l
i
z
a
t
i
o
n
M
:
M
e
t
h
o
d
o
l
o
g
y
So
:
So
f
t
w
a
r
e
Va
:
Va
l
i
d
a
t
i
o
n
Fo
:
Fo
r
mal
a
n
a
l
y
s
i
s
I
:
I
n
v
e
s
t
i
g
a
t
i
o
n
R
:
R
e
so
u
r
c
e
s
D
:
D
a
t
a
C
u
r
a
t
i
o
n
O
:
W
r
i
t
i
n
g
-
O
r
i
g
i
n
a
l
D
r
a
f
t
E
:
W
r
i
t
i
n
g
-
R
e
v
i
e
w
&
E
d
i
t
i
n
g
Vi
:
Vi
su
a
l
i
z
a
t
i
o
n
Su
:
Su
p
e
r
v
i
s
i
o
n
P
:
P
r
o
j
e
c
t
a
d
mi
n
i
st
r
a
t
i
o
n
Fu
:
Fu
n
d
i
n
g
a
c
q
u
i
si
t
i
o
n
CO
NF
L
I
C
T
OF
I
N
T
E
R
E
S
T
T
h
e
au
th
o
r
s
d
ec
lar
e
no
co
n
f
lict
of
in
ter
est.
DATA
AV
AI
L
AB
I
L
I
T
Y
T
h
e
p
ar
tial
d
ata
o
f
th
is
s
tu
d
y
is
av
ailab
le
in
p
u
b
lic
r
ep
o
s
ito
r
y
at
“
h
ttp
s
://i
ee
e
-
d
atap
o
r
t.o
r
g
/d
o
cu
m
en
ts
/ay
u
r
k
o
s
h
-
d
ataset
-
m
ac
h
in
e
-
r
ea
d
ab
le
-
ay
u
r
v
e
d
ic
-
k
n
o
wled
g
e
-
r
eso
u
r
c
e
-
k
n
o
wled
g
e
-
g
r
ap
h
-
an
d
”,
w
h
ile
th
e
r
em
ain
i
n
g
d
ata
ar
e
a
v
ailab
le
o
n
r
eq
u
e
s
t
f
r
o
m
th
e
co
r
r
esp
o
n
d
in
g
a
u
th
o
r
,
as
th
ey
c
o
n
tai
n
in
f
o
r
m
atio
n
th
a
t
co
u
ld
co
m
p
r
o
m
is
e
th
e
p
r
iv
ac
y
o
f
th
e
r
ese
ar
ch
p
ar
ticip
a
n
ts
an
d
ar
e
th
er
e
f
o
r
p
u
b
licly
n
o
t
av
ailab
le
d
u
e
to
eth
ical
co
n
f
id
en
tial r
estrictio
n
s
.
Evaluation Warning : The document was created with Spire.PDF for Python.
I
SS
N
:
2
2
5
2
-
8
7
7
6
I
n
t J I
n
f
&
C
o
m
m
u
n
T
ec
h
n
o
l
,
Vo
l.
1
5
,
No
.
3
,
Sep
tem
b
er
20
2
6
:
1
1
9
7
-
1
2
0
7
1206
RE
F
E
R
E
NC
E
S
[
1
]
S
.
K
.
V
e
r
ma,
M
.
P
a
n
d
e
y
,
A
.
S
h
a
r
ma
,
a
n
d
D
.
S
i
n
g
h
,
“
Ex
p
l
o
r
i
n
g
A
y
u
r
v
e
d
a
:
p
r
i
n
c
i
p
l
e
s
a
n
d
t
h
e
i
r
a
p
p
l
i
c
a
t
i
o
n
i
n
m
o
d
e
r
n
m
e
d
i
c
i
n
e
,
”
Bu
l
l
e
t
i
n
o
f
t
h
e
N
a
t
i
o
n
a
l
Re
s
e
a
r
c
h
C
e
n
t
re
,
v
o
l
.
4
8
,
n
o
.
1
,
2
0
2
4
,
d
o
i
:
1
0
.
1
1
8
6
/
s4
2
2
6
9
-
0
2
4
-
0
1
2
3
1
-
0.
[
2
]
C
.
T.
C
h
e
,
V
.
G
e
o
r
g
e
,
T.
P
.
I
j
i
n
u
,
P
.
P
u
s
h
p
a
n
g
a
d
a
n
,
a
n
d
K
.
A
n
d
r
a
e
-
M
a
r
o
b
e
l
a
,
“
Tr
a
d
i
t
i
o
n
a
l
m
e
d
i
c
i
n
e
,
”
Ph
a
rm
a
c
o
g
n
o
s
y
:
Fu
n
d
a
m
e
n
t
a
l
s,
A
p
p
l
i
c
a
t
i
o
n
s
a
n
d
S
t
r
a
t
e
g
y
,
p
p
.
1
5
–
3
0
,
2
0
1
7
,
d
o
i
:
1
0
.
1
0
1
6
/
B
9
7
8
-
0
-
12
-
8
0
2
1
0
4
-
0
.
0
0
0
0
2
-
0.
[
3
]
F
.
M
i
a
o
,
X
.
Zh
o
u
,
S
.
X
i
a
o
,
a
n
d
S
.
Z
h
a
n
g
,
“
A
g
r
a
p
h
si
m
i
l
a
r
i
t
y
a
l
g
o
r
i
t
h
m
b
a
s
e
d
o
n
g
r
a
p
h
p
a
r
t
i
t
i
o
n
i
n
g
a
n
d
a
t
t
e
n
t
i
o
n
me
c
h
a
n
i
sm,
”
El
e
c
t
r
o
n
i
c
s
(
S
w
i
t
zer
l
a
n
d
)
,
v
o
l
.
1
3
,
n
o
.
1
9
,
2
0
2
4
,
d
o
i
:
1
0
.
3
3
9
0
/
e
l
e
c
t
r
o
n
i
c
s
1
3
1
9
3
7
9
4
.
[
4
]
P
.
Jo
h
r
i
,
S
.
K
.
K
h
a
t
r
i
,
A
.
T
.
A
l
-
Ta
a
n
i
,
M
.
S
a
b
h
a
r
w
a
l
,
S
.
S
u
v
a
n
o
v
,
a
n
d
A
.
K
u
mar,
“
N
a
t
u
r
a
l
l
a
n
g
u
a
g
e
p
r
o
c
e
ssi
n
g
:
h
i
s
t
o
r
y
,
e
v
o
l
u
t
i
o
n
,
a
p
p
l
i
c
a
t
i
o
n
,
a
n
d
f
u
t
u
r
e
w
o
r
k
,
”
L
e
c
t
u
re
N
o
t
e
s
i
n
N
e
t
w
o
r
k
s
a
n
d
S
y
st
e
m
s
,
v
o
l
.
1
6
7
,
p
p
.
3
6
5
-
3
7
5
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
9
8
1
-
15
-
9
7
1
2
-
1
_
3
1
.
[
5
]
B
.
A
b
u
-
S
a
l
i
h
,
M
.
A
L
-
Q
u
r
i
s
h
i
,
M
.
A
l
w
e
sh
a
h
,
M
.
A
L
-
S
m
a
d
i
,
R
.
A
l
f
a
y
e
z
,
a
n
d
H
.
S
a
a
d
e
h
,
“
H
e
a
l
t
h
c
a
r
e
k
n
o
w
l
e
d
g
e
g
r
a
p
h
c
o
n
st
r
u
c
t
i
o
n
:
A
s
y
st
e
ma
t
i
c
r
e
v
i
e
w
o
f
t
h
e
st
a
t
e
-
of
-
t
h
e
-
a
r
t
,
o
p
e
n
i
ssu
e
s,
a
n
d
o
p
p
o
r
t
u
n
i
t
i
e
s,
”
J
o
u
r
n
a
l
o
f
Bi
g
D
a
t
a
,
v
o
l
.
1
0
,
n
o
.
1
,
2
0
2
3
,
d
o
i
:
1
0
.
1
1
8
6
/
s4
0
5
3
7
-
0
2
3
-
0
0
7
7
4
-
9.
[
6
]
M
.
N
z
o
m
o
a
n
d
D
.
M
o
o
d
l
e
y
,
“
I
n
t
e
g
r
a
t
i
n
g
k
n
o
w
l
e
d
g
e
g
r
a
p
h
s
a
n
d
B
a
y
e
s
i
a
n
n
e
t
w
o
r
k
s
:
A
h
y
b
r
i
d
a
p
p
r
o
a
c
h
f
o
r
e
x
p
l
a
i
n
a
b
l
e
d
i
se
a
s
e
r
i
sk
p
r
e
d
i
c
t
i
o
n
,
”
Pr
o
c
e
e
d
i
n
g
s
-
2
0
2
5
I
EEE
4
9
t
h
A
n
n
u
a
l
C
o
m
p
u
t
e
rs,
S
o
f
t
w
a
r
e
,
a
n
d
A
p
p
l
i
c
a
t
i
o
n
s
C
o
n
f
e
r
e
n
c
e
,
C
O
MP
S
A
C
2
0
2
5
,
p
p
.
8
3
4
–
8
4
4
,
2
0
2
5
,
d
o
i
:
1
0
.
1
1
0
9
/
C
O
M
P
S
A
C
6
5
5
0
7
.
2
0
2
5
.
0
0
1
1
1
.
[
7
]
R
.
D
a
s
a
n
d
M
.
S
o
y
l
u
,
“
A
k
e
y
r
e
v
i
e
w
o
n
g
r
a
p
h
d
a
t
a
sci
e
n
c
e
:
T
h
e
p
o
w
e
r
o
f
g
r
a
p
h
s
i
n
sc
i
e
n
t
i
f
i
c
st
u
d
i
e
s
,
”
C
h
e
m
o
m
e
t
r
i
c
s
a
n
d
I
n
t
e
l
l
i
g
e
n
t
L
a
b
o
r
a
t
o
r
y
S
y
st
e
m
s
,
v
o
l
.
2
4
0
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
h
e
m
o
l
a
b
.
2
0
2
3
.
1
0
4
8
9
6
.
[
8
]
S
.
S
o
n
g
,
“
A
n
o
n
l
i
n
e
q
u
e
st
i
o
n
a
n
sw
e
r
i
n
g
sy
s
t
e
m
b
a
se
d
o
n
s
u
b
-
g
r
a
p
h
se
a
r
c
h
i
n
g
,
”
p
p
.
1
–
4.
[
9
]
Y
.
C
u
i
,
H
.
Y
u
,
X
.
G
u
o
,
H
.
C
a
o
,
a
n
d
L.
W
a
n
g
,
“
R
A
K
C
R
:
R
e
v
i
e
w
s
s
e
n
t
i
me
n
t
-
a
w
a
r
e
b
a
se
d
k
n
o
w
l
e
d
g
e
g
r
a
p
h
c
o
n
v
o
l
u
t
i
o
n
a
l
n
e
t
w
o
r
k
s fo
r
p
e
r
so
n
a
l
i
z
e
d
r
e
c
o
mm
e
n
d
a
t
i
o
n
,
”
Ex
p
e
r
t
S
y
s
t
e
m
s
w
i
t
h
A
p
p
l
i
c
a
t
i
o
n
s
,
v
o
l
.
2
4
8
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
e
s
w
a
.
2
0
2
4
.
1
2
3
4
0
3
.
[
1
0
]
M
.
R
.
R
e
z
a
e
i
,
R
.
S
.
F
a
r
d
,
J.
L
.
P
a
r
k
e
r
,
R
.
G
.
K
r
i
sh
n
a
n
,
a
n
d
M
.
La
n
k
a
r
a
n
y
,
“
A
d
a
p
t
i
v
e
k
n
o
w
l
e
d
g
e
g
r
a
p
h
s
e
n
h
a
n
c
e
med
i
c
a
l
q
u
e
st
i
o
n
a
n
sw
e
r
i
n
g
:
B
r
i
d
g
i
n
g
t
h
e
g
a
p
b
e
t
w
e
e
n
LLM
S
a
n
d
e
v
o
l
v
i
n
g
me
d
i
c
a
l
k
n
o
w
l
e
d
g
e
,
”
a
rx
i
v
,
2
0
2
5
.
[
1
1
]
A
.
B
e
r
n
a
s
c
o
n
i
,
A
.
Za
n
g
a
,
P
.
J.
F
.
Lu
c
a
s,
M
.
S
c
u
t
a
r
i
,
A
.
Tr
a
m
a
,
a
n
d
F
.
S
t
e
l
l
a
,
“
A
c
a
u
s
a
l
n
e
t
w
o
r
k
m
o
d
e
l
t
o
e
st
i
m
a
t
e
t
h
e
c
a
r
d
i
o
t
o
x
i
c
e
f
f
e
c
t
o
f
o
n
c
o
l
o
g
i
c
a
l
t
r
e
a
t
m
e
n
t
s
i
n
y
o
u
n
g
b
r
e
a
st
c
a
n
c
e
r
s
u
r
v
i
v
o
r
s,
”
Pr
o
g
ress
i
n
Ar
t
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
0
7
/
s
1
3
7
4
8
-
024
-
0
0
3
4
8
-
7.
[
1
2
]
R
.
Y
a
n
g
e
t
a
l
.
,
“
K
G
-
R
a
n
k
:
E
n
h
a
n
c
i
n
g
l
a
r
g
e
l
a
n
g
u
a
g
e
mo
d
e
l
s
f
o
r
m
e
d
i
c
a
l
Q
A
w
i
t
h
k
n
o
w
l
e
d
g
e
g
r
a
p
h
s
a
n
d
r
a
n
k
i
n
g
t
e
c
h
n
i
q
u
e
s
,
”
Bi
o
N
L
P
2
0
2
4
-
2
3
r
d
Me
e
t
i
n
g
o
f
t
h
e
A
C
L
S
p
e
c
i
a
l
I
n
t
e
r
e
st
G
ro
u
p
o
n
B
i
o
m
e
d
i
c
a
l
N
a
t
u
r
a
l
L
a
n
g
u
a
g
e
P
ro
c
e
ss
i
n
g
,
P
ro
c
e
e
d
i
n
g
s
o
f
t
h
e
Wo
r
k
sh
o
p
a
n
d
S
h
a
r
e
d
T
a
s
k
s
,
p
p
.
1
5
5
–
1
6
6
,
2
0
2
4
,
d
o
i
:
1
0
.
1
8
6
5
3
/
v
1
/
2
0
2
4
.
b
i
o
n
l
p
-
1
.
1
3
.
[
1
3
]
A
.
Li
,
Q
.
W
e
i
,
C
.
H
a
n
,
a
n
d
X
.
X
i
n
g
,
“
R
e
s
e
a
r
c
h
o
n
t
h
e
c
o
n
s
t
r
u
c
t
i
o
n
o
f
sm
a
r
t
c
a
r
e
q
u
e
s
t
i
o
n
a
n
sw
e
r
i
n
g
s
y
s
t
e
m
b
a
s
e
d
o
n
k
n
o
w
l
e
d
g
e
g
r
a
p
h
,
”
Pro
c
e
d
i
a
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
,
v
o
l
.
2
1
4
,
n
o
.
C
,
p
p
.
1
5
9
5
–
1
6
0
2
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
p
r
o
c
s.
2
0
2
2
.
1
1
.
3
4
8
.
[
1
4
]
D
.
M
a
,
X
.
H
e
,
M
.
W
a
n
g
,
Q
.
F
a
n
g
,
H
.
Zh
u
,
a
n
d
P
.
H
u
,
“
R
ET
R
A
C
TED
:
K
n
o
w
l
e
d
g
e
g
r
a
p
h
q
u
e
st
i
o
n
a
n
sw
e
r
i
n
g
b
a
sed
o
n
r
e
l
a
t
i
o
n
a
l
a
t
t
e
n
t
i
o
n
e
n
h
a
n
c
e
m
e
n
t
a
n
d
f
e
a
t
u
r
e
c
r
o
ss
-
c
o
d
i
n
g
,
”
J
o
u
r
n
a
l
o
f
I
n
t
e
l
l
i
g
e
n
t
&
Fu
zz
y
S
y
st
e
m
s
,
v
o
l
.
4
7
,
n
o
.
1
_
s
u
p
p
l
,
p
p
.
4
9
9
–
5
1
3
,
D
e
c
.
2
0
2
4
,
d
o
i
:
1
0
.
3
2
3
3
/
JI
F
S
-
2
3
3
6
5
0
.
[
1
5
]
S
.
R
a
st
o
g
i
e
t
a
l
.
,
“
A
s
u
r
v
e
y
o
f
p
a
t
i
e
n
t
s
v
i
si
t
i
n
g
a
n
A
y
u
r
v
e
d
i
c
t
e
a
c
h
i
n
g
h
o
sp
i
t
a
l
f
o
r
f
a
c
t
o
r
s
i
n
f
l
u
e
n
c
i
n
g
t
h
e
d
e
c
i
s
i
o
n
t
o
c
h
o
o
s
e
a
y
u
r
v
e
d
a
a
s
a
h
e
a
l
t
h
c
a
r
e
p
r
o
v
i
d
e
r
,
”
J
o
u
rn
a
l
o
f
Ay
u
rv
e
d
a
a
n
d
I
n
t
e
g
r
a
t
i
v
e
Me
d
i
c
i
n
e
,
v
o
l
.
1
4
,
n
o
.
1
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
a
i
m.
2
0
2
1
.
1
0
0
5
3
9
.
[
1
6
]
M
.
H
a
u
b
e
n
,
M
.
R
a
f
i
,
I
.
A
b
d
e
l
a
z
i
z
,
a
n
d
O
.
H
a
ssa
n
z
a
d
e
h
,
“
K
n
o
w
l
e
d
g
e
g
r
a
p
h
s
i
n
P
h
a
r
ma
c
o
v
i
g
i
l
a
n
c
e
:
A
s
c
o
p
i
n
g
r
e
v
i
e
w
,
”
C
l
i
n
i
c
a
l
T
h
e
ra
p
e
u
t
i
c
s
,
v
o
l
.
4
6
,
n
o
.
7
,
p
p
.
5
4
4
–
5
5
4
,
J
u
l
.
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
c
l
i
n
t
h
e
r
a
.
2
0
2
4
.
0
6
.
0
0
3
.
[
1
7
]
C
.
M
.
S
u
n
e
e
r
a
,
J.
P
r
a
k
a
sh
,
a
n
d
P
.
K
.
S
i
n
g
h
,
“
Q
u
e
s
t
i
o
n
a
n
sw
e
r
i
n
g
o
v
e
r
k
n
o
w
l
e
d
g
e
g
r
a
p
h
s
u
si
n
g
B
E
R
T
b
a
se
d
r
e
l
a
t
i
o
n
ma
p
p
i
n
g
,
”
Ex
p
e
rt
S
y
st
e
m
s
,
v
o
l
.
4
0
,
n
o
.
1
0
,
D
e
c
.
2
0
2
3
,
d
o
i
:
1
0
.
1
1
1
1
/
e
x
s
y
.
1
3
4
5
6
.
[
1
8
]
S
.
R
.
P
a
r
a
d
k
a
r
,
“
R
o
l
e
o
f
m
o
d
e
r
n
d
i
a
g
n
o
s
t
i
c
m
e
t
h
o
d
s
i
n
a
y
u
r
v
e
d
i
c
d
i
a
g
n
o
s
i
s
:
C
o
n
c
e
p
t
s a
n
d
p
r
o
s
p
e
c
t
s
,
”
Ay
u
sh
d
h
a
r
a
,
v
o
l
.
4
,
n
o
.
4
.
[
1
9
]
T.
D
a
v
e
n
p
o
r
t
a
n
d
R
.
K
a
l
a
k
o
t
a
,
“
Th
e
p
o
t
e
n
t
i
a
l
f
o
r
a
r
t
i
f
i
c
i
a
l
i
n
t
e
l
l
i
g
e
n
c
e
i
n
h
e
a
l
t
h
c
a
r
e
,
”
Fu
t
u
r
e
H
e
a
l
t
h
c
a
re
J
o
u
rn
a
l
,
v
o
l
.
6
,
n
o
.
2
,
pp
.
9
4
–
9
8
,
2
0
1
9
,
d
o
i
:
1
0
.
7
8
6
1
/
f
u
t
u
r
e
h
o
sp
.
6
-
2
-
9
4
.
[
2
0
]
P
.
R
a
j
e
n
d
r
a
n
,
“
D
o
s
h
a
D
h
a
t
u
M
a
l
a
M
o
o
l
a
m
H
i
S
h
a
r
e
e
r
a
m,
”
I
n
t
e
rn
a
t
i
o
n
a
l
J
o
u
r
n
a
l
o
f
S
c
i
e
n
c
e
a
n
d
Re
s
e
a
r
c
h
(
I
J
S
R)
,
v
o
l
.
1
2
,
n
o
.
6
,
p
p
.
3
9
8
–
4
0
0
,
2
0
2
3
,
d
o
i
:
1
0
.
2
1
2
7
5
/
sr
2
3
6
0
2
1
1
1
7
4
5
.
[
2
1
]
J.
Z
h
a
n
g
,
Z.
P
e
i
,
W
.
X
i
o
n
g
,
a
n
d
Z
.
L
u
o
,
“
A
n
sw
e
r
e
x
t
r
a
c
t
i
o
n
w
i
t
h
g
r
a
p
h
a
t
t
e
n
t
i
o
n
n
e
t
w
o
r
k
f
o
r
k
n
o
w
l
e
d
g
e
g
r
a
p
h
q
u
e
s
t
i
o
n
a
n
sw
e
r
i
n
g
,
”
2
0
2
0
I
EEE
6
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
r
e
n
c
e
o
n
C
o
m
p
u
t
e
r
a
n
d
C
o
m
m
u
n
i
c
a
t
i
o
n
s,
I
C
C
C
2
0
2
0
,
p
p
.
1
6
4
5
–
1
6
5
0
,
2
0
2
0
,
d
o
i
:
1
0
.
1
1
0
9
/
I
C
C
C
5
1
5
7
5
.
2
0
2
0
.
9
3
4
5
0
0
0
.
[
2
2
]
S
.
A
g
h
a
e
i
,
K
.
A
n
g
e
l
e
,
a
n
d
A
.
F
e
n
se
l
,
“
B
u
i
l
d
i
n
g
k
n
o
w
l
e
d
g
e
s
u
b
g
r
a
p
h
s
i
n
q
u
e
st
i
o
n
a
n
sw
e
r
i
n
g
o
v
e
r
k
n
o
w
l
e
d
g
e
g
r
a
p
h
s,”
L
e
c
t
u
re
N
o
t
e
s
i
n
C
o
m
p
u
t
e
r
S
c
i
e
n
c
e
(
i
n
c
l
u
d
i
n
g
s
u
b
ser
i
e
s
L
e
c
t
u
r
e
N
o
t
e
s
i
n
Art
i
f
i
c
i
a
l
I
n
t
e
l
l
i
g
e
n
c
e
a
n
d
L
e
c
t
u
re
N
o
t
e
s
i
n
Bi
o
i
n
f
o
rm
a
t
i
c
s)
,
v
o
l
.
1
3
3
6
2
LN
C
S
,
p
p
.
2
3
7
–
2
5
1
,
2
0
2
2
,
d
o
i
:
1
0
.
1
0
0
7
/
9
7
8
-
3
-
0
3
1
-
0
9
9
1
7
-
5
_
1
6
.
[
2
3
]
Y
.
S
h
a
n
g
e
t
a
l
.
,
“
El
e
c
t
r
o
n
i
c
h
e
a
l
t
h
r
e
c
o
r
d
–
o
r
i
e
n
t
e
d
k
n
o
w
l
e
d
g
e
g
r
a
p
h
s
y
s
t
e
m
f
o
r
c
o
l
l
a
b
o
r
a
t
i
v
e
c
l
i
n
i
c
a
l
d
e
c
i
si
o
n
s
u
p
p
o
r
t
u
s
i
n
g
mu
l
t
i
c
e
n
t
e
r
f
r
a
g
me
n
t
e
d
me
d
i
c
a
l
d
a
t
a
:
d
e
si
g
n
a
n
d
a
p
p
l
i
c
a
t
i
o
n
s
t
u
d
y
,
”
J
o
u
r
n
a
l
o
f
M
e
d
i
c
a
l
I
n
t
e
r
n
e
t
R
e
se
a
rc
h
,
v
o
l
.
2
6
,
p
.
e
5
4
2
6
3
,
Ju
l
.
2
0
2
4
,
d
o
i
:
1
0
.
2
1
9
6
/
5
4
2
6
3
.
[
2
4
]
M
.
E
d
a
v
a
l
a
t
h
a
n
d
B
.
P
.
B
h
a
r
a
t
h
a
n
,
“
M
e
t
h
o
d
o
l
o
g
y
f
o
r
d
e
v
e
l
o
p
i
n
g
a
n
d
e
v
a
l
u
a
t
i
n
g
d
i
a
g
n
o
st
i
c
t
o
o
l
s
i
n
A
y
u
r
v
e
d
a
–
A
r
e
v
i
e
w
,
”
J
o
u
rn
a
l
o
f
Ay
u
rv
e
d
a
a
n
d
I
n
t
e
g
ra
t
i
v
e
Me
d
i
c
i
n
e
,
v
o
l
.
1
2
,
n
o
.
2
,
p
p
.
3
8
9
–
3
9
7
,
2
0
2
1
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
a
i
m
.
2
0
2
1
.
0
1
.
0
0
9
.
[
2
5
]
Y
.
F
a
n
g
e
t
a
l
.
,
“
K
n
o
w
l
e
d
g
e
g
r
a
p
h
-
e
n
h
a
n
c
e
d
m
o
l
e
c
u
l
a
r
c
o
n
t
r
a
st
i
v
e
l
e
a
r
n
i
n
g
w
i
t
h
f
u
n
c
t
i
o
n
a
l
p
r
o
mp
t
,
”
N
a
t
u
r
e
M
a
c
h
i
n
e
I
n
t
e
l
l
i
g
e
n
c
e
,
v
o
l
.
5
,
n
o
.
5
,
p
p
.
5
4
2
–
5
5
3
,
2
0
2
3
,
d
o
i
:
1
0
.
1
0
3
8
/
s
4
2
2
5
6
-
0
2
3
-
0
0
6
5
4
-
0.
[
2
6
]
W
i
k
i
p
e
d
i
a
C
o
n
t
r
i
b
u
t
o
r
s,
“
O
v
e
r
l
a
p
c
o
e
f
f
i
c
i
e
n
t
,
”
Wi
k
i
p
e
d
i
a
,
2
0
2
5
.
h
t
t
p
s:
/
/
e
n
.
w
i
k
i
p
e
d
i
a
.
o
r
g
/
w
i
k
i
/
O
v
e
r
l
a
p
_
c
o
e
f
f
i
c
i
e
n
t
(
a
c
c
e
s
se
d
J
u
n
.
0
9
,
2
0
2
4
)
.
[
2
7
]
Z.
W
a
n
g
,
“
A
c
o
mp
u
t
a
t
i
o
n
a
l
a
p
p
r
o
a
c
h
t
o
e
p
i
l
e
p
s
y
t
r
e
a
t
m
e
n
t
:
A
n
A
I
-
o
p
t
i
mi
z
e
d
g
l
o
b
a
l
n
a
t
u
r
a
l
p
r
o
d
u
c
t
p
r
e
scr
i
p
t
i
o
n
s
y
st
e
m,
”
2
0
2
5
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
:
/
/
a
r
x
i
v
.
o
r
g
/
a
b
s/
2
5
0
5
.
0
9
6
4
3
.
[
2
8
]
C
.
S
a
n
d
e
e
p
,
V
.
G
u
p
t
a
,
V
.
D
e
sh
mu
k
h
,
a
n
d
S
.
S
a
r
d
e
sh
m
u
k
h
,
“
I
n
t
e
g
r
a
t
i
n
g
A
y
u
r
v
e
d
i
c
p
h
i
l
o
so
p
h
y
w
i
t
h
m
o
d
e
r
n
t
e
c
h
n
o
l
o
g
i
e
s
f
o
r
d
r
u
g
r
e
sea
r
c
h
a
n
d
d
e
v
e
l
o
p
m
e
n
t
:
A
c
r
i
t
i
c
a
l
n
e
e
d
o
f
m
e
c
h
a
n
i
s
t
i
c
i
n
si
g
h
t
s
f
o
r
w
i
d
e
r
a
c
c
e
p
t
a
b
i
l
i
t
y
,
”
J
o
u
r
n
a
l
o
f
A
y
u
r
v
e
d
a
a
n
d
I
n
t
e
g
ra
t
i
v
e
Me
d
i
c
i
n
e
,
v
o
l
.
1
5
,
n
o
.
5
,
2
0
2
4
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
a
i
m.
2
0
2
4
.
1
0
0
9
5
4
.
[
2
9
]
H
.
Y
i
n
g
,
Z.
Z
h
a
o
,
Y
.
Z
h
a
o
,
S
.
Ze
n
g
,
a
n
d
S
.
Y
u
,
“
C
o
R
T
Ex
:
C
o
n
t
r
a
st
i
v
e
l
e
a
r
n
i
n
g
f
o
r
r
e
p
r
e
s
e
n
t
i
n
g
t
e
r
ms
v
i
a
e
x
p
l
a
n
a
t
i
o
n
s
w
i
t
h
a
p
p
l
i
c
a
t
i
o
n
s
o
n
c
o
n
s
t
r
u
c
t
i
n
g
b
i
o
m
e
d
i
c
a
l
k
n
o
w
l
e
d
g
e
g
r
a
p
h
s,
”
J
o
u
r
n
a
l
o
f
t
h
e
Am
e
ri
c
a
n
Me
d
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
Ass
o
c
i
a
t
i
o
n
,
v
o
l
.
3
0
,
n
o
.
9
,
p
p
.
1
5
4
3
–
1
5
5
1
,
D
e
c
.
2
0
2
3
,
[
O
n
l
i
n
e
]
.
A
v
a
i
l
a
b
l
e
:
h
t
t
p
:
/
/
a
r
x
i
v
.
o
r
g
/
a
b
s/
2
3
1
2
.
0
8
0
3
6
.
[
3
0
]
H
.
C
u
i
e
t
a
l
.
,
“
A
r
e
v
i
e
w
o
n
k
n
o
w
l
e
d
g
e
g
r
a
p
h
s
f
o
r
h
e
a
l
t
h
c
a
r
e
:
R
e
s
o
u
r
c
e
s,
a
p
p
l
i
c
a
t
i
o
n
s,
a
n
d
p
r
o
m
i
ses
,
”
J
o
u
r
n
a
l
o
f
Bi
o
m
e
d
i
c
a
l
I
n
f
o
rm
a
t
i
c
s
,
v
o
l
.
1
6
9
,
p
.
1
0
4
8
6
1
,
S
e
p
.
2
0
2
5
,
d
o
i
:
1
0
.
1
0
1
6
/
j
.
j
b
i
.
2
0
2
5
.
1
0
4
8
6
1
.
[
3
1
]
X
.
Zh
a
n
g
e
t
a
l
.
,
“
G
r
e
a
s
e
l
m
:
G
r
a
p
h
r
e
a
so
n
i
n
g
e
n
h
a
n
c
e
d
l
a
n
g
u
a
g
e
m
o
d
e
l
s
f
o
r
q
u
e
st
i
o
n
a
n
sw
e
r
i
n
g
,
”
I
C
L
R
2
0
2
2
-
1
0
t
h
I
n
t
e
r
n
a
t
i
o
n
a
l
C
o
n
f
e
re
n
c
e
o
n
L
e
a
rn
i
n
g
Re
p
res
e
n
t
a
t
i
o
n
s
,
2
0
2
2
.
Evaluation Warning : The document was created with Spire.PDF for Python.