Investigating adjective denotation and collocation Ann Copestake Computer Laboratory, University of Cambridge Outline introduction: compositional semantics, GL and semantic space models. denotation and collocation  distribution of `magnitude’

Download Report

Transcript Investigating adjective denotation and collocation Ann Copestake Computer Laboratory, University of Cambridge Outline introduction: compositional semantics, GL and semantic space models. denotation and collocation  distribution of `magnitude’

Investigating adjective
denotation and collocation
Ann Copestake
Computer Laboratory,
University of Cambridge
Outline
introduction: compositional semantics,
GL and semantic space models.
denotation and collocation
 distribution of `magnitude’ adjectives
 hypotheses about adjective denotation
and collocation
 semi-productivity

Themes
semi-productivity: extending paper in
GL 2001 to phrases
 statistical and symbolic models
interacting
 generation as well as analysis
 computational account

Different branches of
computational semantics

compositional semantics: capture syntax, (some)
close-class words and (some) morphology



lexical semantics, e.g.,




every x [ dog’(x) -> bark’(x)]
large coverage grammars as testbed for GL (constructions,
composition, underspecification)
GL (interacts with compositional semantics)
WordNet
meaning postulates etc
semantic space models, e.g.,



LSA
Schütze (1995)
Lin (multiple papers), Pado and Lapata (2003)
semantic spaces



acquired from corpora
generally, collect vectors of words
which co-occur with the target
more sophisticated models incorporate
syntactic relationships
dog
bark
house
cat
dog
-
1
0
0
bark
1
-
0
0
Semantic space models and
compositional semantics?


do spaces correspond to predicates in compositional semantics?
e.g., bark’
attractions





problems




automatic acquisition
similarity metrics, priming
fuzziness, meaning variation, sense clustering
statistical approximation to real world knowledge? (but fallacy with
parse selection techniques)
classical lexical semantic relations (hyponymy etc) aren’t captured
well
can’t do inference
sensitivity to domain/corpus
role of collocation?
Denotation: assumptions

Truth-conditional, logically formalisable (in
principle), refers to `real world’ (extension)




Not necessarily decomposable: natural kinds (dog’
– canis familiaris), natural predicates
Naive physics, biology, etc
Computationally: specification of meaning
that interfaces with non-linguistic
components
Selectional restrictions?

bark’(x) -> dog’(x) or seal’(x) or ...
Collocation: assumptions

Significant co-occurrences of words in
syntactically interesting relationships




`syntactically interesting’: for this talk, attributive
adjectives and the nouns they immediately
precede
`significant’: statistically significant (but on what
assumptions about baseline?)
Compositional, no idiosyncratic syntax etc (as
opposed to multiword expression)
About language rather than the real world
Collocation versus denotation



Whether an unusually frequent word pair is a
collocation or not depends on assumptions about
denotation: fix denotation to investigate collocation
Empirically: investigations using WordNet synsets
(Pearce, 2001)
Anti-collocation: words that might be expected to go
together and tend not to


e.g., ? flawless behaviour (Cruse, 1986): big rain (unless
explained by denotation)
e.g., buy house is predictable on basis of denotation,
shake fist is not
Collocation and denotation
investigations




can this notion of collocation be made
precise, empirically testable?
assumptions about denotation determine
whether something is a collocation
semantic space models will include
collocational effects
initial, very preliminary, investigations with
magnitude adjectives


attributive adjectives: can get corpus data without
parsing
only one argument to consider
Distribution of `magnitude’
adjectives: summary



some very frequent adjectives have magnituderelated meanings (e.g., heavy, high, big, large)
basic meaning with simple concrete entities
extended meaning with abstract nouns, non-concrete
physical entities (high taxation, heavy rain)




extended uses more common than basic
not all magnitude adjectives – e.g. tall
nouns tend to occur with a limited subset of these
extended adjectives
some apparent semantic groupings of nouns which
go with particular adjectives, but not easily specified
Some adjective-noun
frequencies in the BNC
number
proportion quality problem
part
winds
rain
large
1790
404
0
10
533
0
0
high
92
501
799
0
3
90
0
big
11
1
0
79
79
3
1
heavy
0
0
1
0
1
2
198
Grammaticality judgments
number
proportion quality problem
large
*
high
heavy
?
*
big
?
?
*
*
part
?
winds
rain
*
*
*
*
*
More examples
impor
tance
success
majority
number
proport
ion
quality
role
problem
part
winds
support
rain
great
310
360
382
172
9
11
3
44
71
0
22
0
large
1
1
112
1790
404
0
13
10
533
0
1
0
high
8
0
0
92
501
799
1
0
3
90
2
0
major
62
60
0
0
7
0
272
356
408
1
8
0
big
0
40
5
11
1
0
3
79
79
3
1
1
strong
0
0
2
0
0
1
8
0
3
132
147
0
heavy
0
0
1
0
0
1
0
0
1
2
4
198
Judgments
impor
tance
success
majority
number
proporti
on
quality
role
problem
part
great
large
?
high
?
*
major
*
?
?
?
*
?
winds
?
strong
?
?
*
*
*
heavy
?
*
?
*
*
rain
?
*
*
*
?
*
?
big
support
?
?
*
*
*
?
Distribution




Investigated the distribution of heavy, high, big,
large, strong, great, major with the most common
co-occurring nouns in the BNC
Nouns tend to occur with up to three of these
adjectives with high frequency and low or zero
frequency with the rest
My intuitive grammaticality judgments correlate but
allow for some unseen combinations and disallow a
few observed but very infrequent ones
big, major and great are grammatical with many
nouns (but not frequent with most), strong and
heavy are ungrammatical with most nouns, high and
large intermediate
heavy: groupings?
magnitude: dew, rainstorm, downpour, rain, rainfall,
snowfall, fall, snow, shower: frost, spindrift: clouds,
mist, fog: flow, flooding, bleeding, period, traffic:
demands, reliance, workload, responsibility, emphasis,
dependence: irony, sarcasm, criticism: infestation,
soiling: loss, price, cost, expenditure, taxation, fine,
penalty, damages, investment: punishment, sentence:
fire, bombardment, casualties, defeat, fighting:
burden, load, weight, pressure: crop: advertising: use,
drinking:
magnitude of verb: drinker, smoker:
magnitude related? odour, perfume, scent, smell,
whiff: lunch: sea, surf, swell:
high: groupings?
magnitude: esteem, status, regard, reputation,
standing, calibre, value, priority; grade, quality, level;
proportion, degree, incidence, frequency, number,
prevalence, percentage; volume, speed, voltage,
pressure, concentration, density, performance,
temperature, energy, resolution, dose, wind; risk, cost,
price, rate, inflation, tax, taxation, mortality, turnover,
wage, income, productivity, unemployment, demand
magnitude of verb: earner
heavy and high
50 nouns in BNC with the extended
magnitude use of heavy with frequency
10 or more
 160 such nouns with high
 Only 9 such nouns with both adjectives:
price, pressure, investment, demand,
rainfall, cost, costs, concentration,
taxation

Basic adjective denotation
with simple concrete objects:
high’(x) => zdim(x) > norm(zdim,type(x),c)
heavy’(x) => wt(x) > norm(wt,type(x),c)
where zdim is distance on vertical, wt is weight
(measure functions, MF)
norm(MF,class,context) is some standard for
MF for class in context
(high’ also requires selectional restriction – not animate)
Metaphor

Different metaphors for different nouns (cf., Lakoff et
al)



Empirical account of distribution?



`high’ nouns measured with an upright scale: e.g.,
temperature: temperature is rising
`heavy’ nouns metaphorically like burden: e.g., workload:
her workload is weighing on her
predictability of noun classes? high volume? high and heavy
taxation
adjective denotation for inference etc? via literal denotation?
Discussed again at end of talk
Possible empirical accounts of
distribution
1. Difference in denotation between
`extended’ uses of adjectives
2. Grammaticized selectional
restrictions/preferences
3. Lexical selection
•
stipulate Magn function with nouns (MeaningText Theory)
4. Semi-productivity / collocation
•
plus semantic back-off
Computational semantics
perspective
Require workable account of
denotation: not too difficult to acquire,
not over-specific
 Require account of distribution for
generation
 Robustness and completeness
 Can’t assume pragmatics / real world
knowledge does the difficult bits!

Denotation account of
distribution
Denotation of adjective simply prevents it being possible with
the noun.
 heavy and high have different denotations
heavy’(x) => MF(x) > norm(MF,type(x),c) & precipitation(x) or
cost(x) or flow(x) or consumption(x)...

(where rain(x) -> precipitation(x) and so on)

But: messy disjunction or multiple senses, open-ended, unlikely
to be tractable.



e.g., heavy shower only for rain sense, not bathroom sense
Not falsifiable, but no motivation other than distribution.
Dictionary definitions can be seen as doing this (informally), but
none account for observed distribution.
Selectional restrictions and
distribution


Assume the adjectives have the same denotation
Distribution via features in the lexicon






e.g., literal high selects for [ANIMATE false ]
approach used in the LinGO ERG for in/on in temporal
expressions
grammaticized, so doesn’t need to be determined by
denotation (though assume consistency)
can utilise qualia structure
Problem: can’t find a reasonable set of cross-cutting
features!
Stipulative approach possible, but unattractive.
Lexical selection
MTT approach
 noun specifies its Magn adjective



in Mel’čuk and Polguère (1987), Magn is a
function, but could modify to make it a set,
or vary meanings
stipulative: if we’re going to do this,
why not use a corpus directly?
Collocational account of
distribution


all the adjectives share a denotation corresponding
to magnitude (more details later), distribution
differences due to collocation, soft rather than hard
constraints
linguistically:





adjective-noun combination is semi-productive
denotation and syntax allow heavy esteem etc, but speakers
are sensitive to frequencies, prefer more frequent phrases
with same meaning
cf morphology and sense extension: Briscoe and Copestake
(1999)
blocking (but weaker than with morphology)
anti-collocations as reflection of semi-productivity
Collocational account of
distribution

computationally, fits with some current
practice:
filter adjective-noun realisations according
to n-grams (statistical generation – e.g.,
Langkilde and Knight)
 use of co-occurrences in WSD


back-off techniques
Collocational vs denotational
differences
high
heavy
Denotation
difference
low
Collocation difference
Back-off and analogy


back-off: decision for infrequent noun with no corpus
evidence for specific magnitude adjective
based on productivity of adjective: number of nouns
it occurs with


default to big
back-off also sensitive to word clusters



e.g., heavy spindrift because spindrift is semantically similar
to snow
semantic space models: i.e., group according to distribution
with other words
hence, adjective has some correlation with semantics of the
noun
Metaphor again
extended metaphor idea is consistent
with idea that clusters for backoff are
based on semantic space
 words cluster according to how they cooccur



e.g., high words cluster with rise words?
but this doesn’t require that we
interpret high literally and then coerce
More details: denotation of
extended adjective uses

mass: e.g., rain, and some plural e.g.,
casualties



cf much, many
inherent measure: e.g., grade, percentage,
fine
other: e.g., rainstorm, defeat, bombardment


attribute in qualia has Magn – heavy rainstorm
equivalent to storm with heavy rain
also heavy drinker etc
More details



Different uses cross-cut adjective distinction
and domain categories
Want to have single extended sense and
some form of co-composition
Further complications: nouns with temporal
duration



heavy rain – not the same as persistent rain
heavy fighting but heavy drinking
how much of this do we have to encode
specifically?
Connotation

heavy often has negative connotations
heavy fine but not ? heavy reward etc
 heavy taxation versus high taxation


consistent with the semantic cluster /
extended metaphor idea
Necessary experiments





None of this is tested yet!
Specify denotation, check for accuracy
Implement semi-productivity model with
back-off
Determine predictability of adjective based on
noun alone
Extension to other adjectives? Magnitude
adjectives may be more lexical than others.
Conclusions
Testing collocational account of
distribution requires fixing denotation
 Magnitude adjectives: assume same
denotation



more complex denotations would need
different experiments
Semi-productivity at the phrasal level

Back-off account is crucial
Some final comments
denotation, selectional restriction,
collocation: choice between
mechanisms?
 ngrams for language models for speech
recognition
 variants of semantic space models that
are less sensitive to collocation effects?


can we `remove’ collocation?