Ai Makabi, Hiroshi Matsumoto and Kazuhide Yamamoto. Automatic Selection of Predicates for Common Sense Knowledge Expression. Proceedings of the Conference of the Pacific Association for Computational Linguistics (PACLING 2013), no page numbers (2013.9)
Strategize a Smooth Tenant-to-tenant Migration and Copilot Takeoff
Automatic Selection of Predicates for Common Sense Knowledge Expression
1. Automa'c
selec'on
of
predicates
for
common
sense
knowledge
expression
Ai
Makabi,
Kazuhide
Yamamoto,
Hiroshi
Matsumoto
Nagaoka
University
of
Technology
4. Related
Works
1/2
• Exis'ng
Upper
Ontologies
(SUMO,
Cyc,
etc.)
– Contain
many
general
concepts
– e.g.
Collec'on:
book
• A
Type
of:
Informa'on
bearing
object
the
form
of
paper
• Instance
of:
Kind
of
ar'fact
not
dis'nguished
by
brand
or
model
• Merits:
– Exploit
rigorously-‐defined
CSK
• Demerits:
– Knowledge
representa'on
cannot
be
matched
fully
with
actual
expressions
5. Related
Works
2/2
• Defineing
the
CSK
as
some
rela'ons
are
added
to
sentences/words
(ConceptNet)
– e.g.
犬(dog)
• CapableOf:
散歩(walk),
寝る(sleep)
• SymbolOf:
忠誠(loyalty),
• Merits:
– Defini'on
is
be_er
suited
to
a
natural
language
processing
task
• Demerits:
– For
the
Japanese
ConceptNet,
the
most
concepts
are
collected
manually
• Coverage
of
CSK
is
excep'onally
low
8. Specific
Property
of
CSK
• We
make
the
three
hypothesis:
1) The
predicate
a
is
the
CSK
of
the
noun
n
when
the
pair
of
a
and
n
are
frequently
co-‐occurred
in
sentences.
2) The
predicate
a
which
co-‐occurs
with
any
nouns
is
not
the
appropriate
CSK
3) Whether
the
predicate
a
is
a
correct
CSK
or
not,
it
depends
on
the
number
of
unique
nouns
which
co-‐occurred
with
a.
9. Specific
Property
of
CSK
• We
make
the
three
hypothesis:
1) The
predicate
a
is
the
CSK
of
the
noun
n
when
the
pair
of
a
and
n
are
frequently
co-‐occurred
in
sentences.
2) The
predicate
a
which
co-‐occurs
with
any
nouns
is
not
the
appropriate
CSK
3) Whether
the
predicate
a
is
a
correct
CSK
or
not,
it
depends
on
the
number
of
unique
nouns
which
co-‐occurred
with
a.
15. Specific
Property
of
CSK
• We
make
the
three
hypothesis:
1) The
predicate
a
is
the
CSK
of
the
noun
n
when
the
pair
of
a
and
n
are
frequently
co-‐occurred
in
sentences.
2) The
predicate
a
which
co-‐occurs
with
any
nouns
is
not
the
appropriate
CSK
3) Whether
the
predicate
a
is
a
correct
CSK
or
not,
it
depends
on
the
number
of
unique
nouns
which
co-‐occurred
with
a.
25. Specific
Property
of
CSK
• We
make
the
three
hypothesis:
1) The
predicate
a
is
the
CSK
of
the
noun
n
when
the
pair
of
a
and
n
are
frequently
co-‐occurred
in
sentences.
2) The
predicate
a
which
co-‐occurs
with
any
nouns
is
not
the
appropriate
CSK
3) Whether
the
predicate
a
is
a
correct
CSK
or
not,
it
depends
on
the
number
of
unique
nouns
which
co-‐occurred
with
a.
29. ?%8+Table 0)$.)I
1+,+'-5)/0+12(#+*)
NUMBER OF DELETING PREDICATES FOR EACH NOUN (N=THE
.$0)+()-$-
UNIQUE NUMBER OF CO-OCCURRED PREDICATES)
Scope of the nouns Deletion
N!700 427
700N!1,100 267
1,100N!1,600 143
1,600N!2,500 73
others 33
R$-*21+0)##)#+)ee)
/0+12(#+*)0+)-$#)RW:9)
-1)1+,+#+).0$%),,)
-$-*)*)2-($00+(#,B)
/0+12(#+*))
However, the 33 predicates, which get deleted when
S:P3V-1+0*#-1X9)LD3VA+X9)KP3V*++9),$$3X9)GP3V8+($%+X9))
G63V-$#2-5X9)FP3V#3+9)1$/#9)/0+.+0X9)E;P3V(-X9)@P3V3-$4X9))
P3V($%+X9)9L73V#2-3X9)9963V%-BX9)6P3V8+9)-++19)*$$#X
can be used to nearly all nouns, so we consider
are not common sense knowledge, and delete from
as incorrectly predicates. Figure 6 shows a list of
A. Evaluation We compare following (1) Do predicates (2) Do predicates (3) Remove by normalized We compare 6%/,+)$.)1+,+'-5)/0+12(#+*
30. relate), B. Evaluation We take their assigned follows (Table The proposed noun as the On the other which frequently much higher “犬(dog)”, “一緒(be together)” appeared in :
やる(do), かける(build, hang, run, lack)
(predicates the weighted scores for predicates co-occurring with noun
using Figure Harman 6. Added
The normalized deleting CSK
predicates frequency. for
each
for A all noun
predicate noun
is correct
common sense knowledge for a noun when the predicate
score is high. The equation of Harman normalized frequency
is as follows (n: noun, a: predicate, na,n: appearance of predicate a with noun n).
use the selected predicates as common sense knowl-edge,
and add them to each noun. In particular, we calculate
weighted scores for predicates co-occurring with noun
Harman normalized frequency. A predicate is correct
common sense knowledge for a noun when the predicate
TF(a, n) =
log2(!
na,n + 1)
log2(
high. The equation of Harman normalized k nk,n)
frequency
follows (n: noun, a: predicate, na,n: appearance fre-quency
of predicate a with noun n).
• The
following
equa'on
computes
weighted
scores
for
predicates
co-‐occurring
with
noun
using
Harman
normalized
frequency
A
predicate
is
appreciate
as
correct
CSK
for
a
noun
when
TF(the
predicate
a, n) =
score
is
high.
log2(!
na,n + 1)
log2(
k nk,n)
(1)
Figure 6. The deleting predicates for all noun
use the selected predicates as common sense knowl-edge,
and add them to each noun. In particular, we calculate
weighted scores for predicates co-occurring with noun
Harman normalized frequency. A predicate is correct
common sense knowledge for a noun when the predicate
high. The equation of Harman normalized frequency
follows (n: noun, a: predicate, na,n: appearance fre-quency
of predicate a with noun n).
TF(a, n) =
log2(!
na,n + 1)
log2(
k nk,n)
for all noun
predicates as common sense knowl-edge,
noun. In particular, we calculate
predicates co-occurring with noun
frequency. A predicate is correct
a noun when the predicate
Harman normalized frequency
predicate, na,n: appearance fre-quency
noun n).
log2(na,n + 1)
!
(1)
noun
:
predicate
:
appearance
frequency
of
predicate
a
with
noun
n
31. Baselines
1) Do
not
delete
the
any
predicates,
just
use
the
weighted
predicates
by
Harman
normalized
frequency
(baseline
1)
2) Do
not
delete
the
any
predicates,
just
use
the
weighted
predicates
by
TF-‐IDF
score
(baseline
2)
3) Remove
the
427
dele'ng
predicates
in
N≤700,
and
use
the
weighted
predicates
by
Harman
normalized
frequency
(baseline
3)
35. Error
Analysis
1/3
• Although
a
predicate
co-‐occurs
with
a
noun
many
'mes,
there
are
unrelated
pairs
– Do
not
check
the
dependency
rela'on
between
them
Solu'on:
Use
only
the
predicates
which
depend
on
the
target
nouns
as
candidate
of
CSK
36. Error
Analysis
2/3
• Could
not
assign
nouns,
which
can
also
be
used
as
suffix
to
appropriate
predicates
–
美しい月です (This
is
the
beau'ful
moon)
– 月ごとに決済する
(We
make
a
charge
for
each
month)
Solu'on:
U'lize
the
rela'on
of
another
co-‐occurred
nouns
e.g.,
If
the
“月”
is
co-‐occurred
with
a
noun
“太陽
(sun)”,
it
may
mean
the
moon
37. Error
Analysis
3/3
• Include
nouns
which
are
used
for
defining
the
rela'on
of
nouns
– 原因
(cause)
– 理由
(reason)
Solu'on:
Discuss
how
we
limit
the
nouns
of
adding
target
38. Conclusion
• Described
the
selec'on
method
of
appropriate
predicate
as
CSK
for
construc'ng
the
CSKB.
– Method
for
sta's'cally
selec'ng
CSK
of
nouns
u'lizing
the
unique
number
of
co-‐occurred
predicates.
• Evaluated
sets
of
CSK
which
are
assigned
to
each
noun
compared
with
three
baselines
– Demonstrated
assumed
characteris'cs
of
CKS
in
our
study
– Gave
a
subjec've
evalua'on
• Plan
to
make
a
quan'ta've
evalua'on