conceptual coherence in the generation of referring expressions

33
conceptual coherence in the generation of referring expressions Albert Gatt & Kees van Deemter University of Aberdeen {agatt, kvdeemte}@csd.abdn.ac.uk

Upload: holland

Post on 06-Jan-2016

30 views

Category:

Documents


0 download

DESCRIPTION

conceptual coherence in the generation of referring expressions. Albert Gatt & Kees van Deemter University of Aberdeen {agatt, kvdeemte}@csd.abdn.ac.uk. - PowerPoint PPT Presentation

TRANSCRIPT

Page 1: conceptual coherence in the generation of referring expressions

conceptual coherence in the generation of referring expressions

Albert Gatt & Kees van Deemter

University of Aberdeen

{agatt, kvdeemte}@csd.abdn.ac.uk

Page 2: conceptual coherence in the generation of referring expressions

Gatt and Van Deemter 2007: “Lexical Choice and conceptual perspective in the generation of plural referring expressions”. Journal of Logic Language and Information (JoLLI) 16 (4), p.423-444.

Page 3: conceptual coherence in the generation of referring expressions

some received wisdom…

Choice is ultimately dependent on the perspective you decide to take on the referent (...).

Will it be more effective for me to refer to my sister as my sister or as that lady or as the physicist ? (Levelt `99, p. 226)

Page 4: conceptual coherence in the generation of referring expressions

the rest of this talk…

1. Generation of Referring Expressions2. Perspective and Conceptual Coherence

reference to sets experimental work

3. An algorithm evaluation

4. Extensions: local (Conceptual) Coherence in discourse

Page 5: conceptual coherence in the generation of referring expressions

Generation of Referring Expressions (GRE)

Part of micro-planning (Reiter/Dale `00)

At this stage, the content of a message is being determined, including descriptions of domain objects (Noun Phrases)

The task of GRE:– given a set of intended referents, look up properties of these

referents that will distinguish them from their distractors in a Knowledge Base

Page 6: conceptual coherence in the generation of referring expressions

Content determination strategies

Most algorithms inspired by the Gricean maxims (Grice `75)– especially Brevity (Dale `89, Gardent `02)

But compare:?? λx: professor(x) V plump(x)

?? λx: professor(x) V [plump(x) & man(x)]

λx: biologist(x) V physicist(x) Not all of these have an equally good ring to them.

entity base type occupation specialisation girth

e1 woman professor physicist plump

e2 woman lecturer geologist thin

e3 man lecturer biologist thin

e4 man postgraduate thin

Page 7: conceptual coherence in the generation of referring expressions

the Conceptual Coherence constraint

Sets (and disjunction): λx: p(x) V q(x) ‘the p and the q’– reference to a plurality suggests to the listener that there is a relationship

holding between elements of the pluralities– p and q should be related or “similar”– semantic relatedness allows the listener to conceptualise the plurality more

easily (Sanford and Moxey, `95)

Gatt and van Deemter (`02):– People’s preference for descriptions of this form were highly correlated to

the semantic similarity of disjuncts– Best results achieved with a distributional definition of similarity (Lin `98)– sim(w,w’) is a function of how often w and w’ occur in the same

grammatical relations in a corpus

Page 8: conceptual coherence in the generation of referring expressions

Lin’s definition of distributional similarity

Let w1, w2 be two words of the same grammatical category.

E.g. dog, cat GR contains information about a syntactic relation w occurs in:

– GR = <w, R, x, p>– w the target word, R the relation, x the co-argument of w– p is the probability of w and x occurring in this construction (as

mutual information).– Example: <dog, modified-by, stray, 0.002>

sim(w1, w2) is calculated using the GR triples that w1 and w2 share.

We use SketchEngine, a large-scale implementation of this theory, based on the BNC (Kilgarriff, `03)

Page 9: conceptual coherence in the generation of referring expressions

experiment 1: multimodal sentence completion

General idea:– To refer to a set, people will prefer to use a plural that

respects the conceptual coherence constraint– If this is impossible, then they will break down the set in

manageable parts. Experimental domains:

– 3 targets (a,b,c) + 1 distractor (d)– sim(a,b) could be high or low– sim(a,c) ≈ sim(b,c) = low

Expectation:– if 2 of the targets have semantically high-sim types, they will

be referred to in a plural description

Page 10: conceptual coherence in the generation of referring expressions

experiment 1: example domain

£5

£5 £5

£20

Complete the following by clicking on the pictures:

The _____________ and the _____________ cost £5.

The _____________ also costs £5.

Experimental domain:

1. Participants completed the sentences by clicking on the pictures.

2. Manipulation of similarity of two of the objects (a,b).

3. Hypothesis:

If {a,b} are similar, they are more likely to be referred to in the plural.

a

bc

d

Page 11: conceptual coherence in the generation of referring expressions

experiment 1: results

Proportion of plural references to designated targets {a,b} when:

{a,b} are semantically similar {a,b} are semantically dissimilar

Page 12: conceptual coherence in the generation of referring expressions

experiment 2: sentence continuation

Does similarity play a role in content determination?

Distinguishing properties: nouns (12) or adjectives (12 ). Expectation:

– Participants will select similar properties in the plural description

A university building was robbed last night. The police have detained three suspects for questioning, all of whom work or study at the university. 1. One of them is a postgraduate. He is a physicist. 2. Another is a Greek, an undergraduate. 3. Also among the suspects is a cleaner. He is an Italian. Both ______________________ were held in custody, but the physicist was released last night.

Page 13: conceptual coherence in the generation of referring expressions

experiment 2: results

Friedman 45.89, p < .001trend as expected

Friedman 36.3, p < .001trend in the opposite direction

Proportion of references using pairwise similar properties:

Nouns: Adjectives:

Page 14: conceptual coherence in the generation of referring expressions

summary of findings so far

In referential situations, people prefer to produce plural descriptions if the entities can be conceptualised under the same perspective.

This holds for types, but not modifiers– Types correspond to “concepts”, and are the way we carve

up the world and categorise objects– Modifiers correspond to properties of those objects.

Results have been corroborated in other experiments

Page 15: conceptual coherence in the generation of referring expressions

Aloni (2002): answers to questions “wh x?” must conceptualise the different x using one and the same perspective (relevant given hearer’s information state and the context)

Our experiments confirm that this idea is on the right track …

Page 16: conceptual coherence in the generation of referring expressions

The challenge for an algorithm:

Complete coherence is often not possible “the Italian, the Greek and the Spaniard” –

But what if there are 5 Spaniards? “the Italian, the Greek and ?” – What if you

don’t know the person’s nationality? “the table, the chair and the plant” – What if

you need to refer to an object that’s of different kind of the other two?

Page 17: conceptual coherence in the generation of referring expressions

a GRE algorithm

The algorithm should try to find the most coherent description possible. Assumption: this should be done even at the cost of brevity!

Main knowledge source:– The relation sim (Kilgarriff `03)

Input:– Knowledge Base– Target referents (R )

Page 18: conceptual coherence in the generation of referring expressions

step 1

1. Lexicalise properties in the KB2. Identify types (nominal properties) and modifiers The set of types and the similarity relation define a

semantic space S = <T, sim>

Definition 1: PerspectiveA perspective P is a convex subset of S, i.e.:

∀ t, t’, t’’ T: ∈t, t’ ∈ P & sim(t, t’’) ≥ sim(t, t’) t’’ P∈

Computed using a clustering algorithm (Gatt `06), which recursively groups together semantic nearest neighbours.

Page 19: conceptual coherence in the generation of referring expressions

perspective graph

T: {lecturer, professor, postgraduate}

T: {woman, man}M: {plump, thin}

T: {geologist, physicist,biologist, chemist}

32

1

1 0.6

1

Aim: find a description for R that minimises the distance between perspectives from which properties are selected.

Weight of a description, w(D): the sum of distances between perspectives represented in D.

– w( ‘the professor and the plump man’ ) = 1– w( ‘the biologist and the physicist’ ) = 0

Page 20: conceptual coherence in the generation of referring expressions

descriptive coherence

Definition 2: Maximal coherence

D is maximally coherent if there is no D’ coextensive with D such that w(D’) < w(D)

– implies finding a shortest connection network in the perspective graph (intractable!)

Definition 3: Local coherenceD is locally coherent if there is no D’ coextensive with D s.t.:1. D’ is obtained by replacing a perspective in D 2. w(D’) < w(D)

Page 21: conceptual coherence in the generation of referring expressions

N ∅ //the perspectives represented in D root perspective with most referents in its extension starting from root do:

– Check types and modifiers. – If a property excludes distractors:

add it to D add the perspective to N

– If R is not distinguished, go to the next perspective, which is

search procedure

NuVPPuw ),(min

(V is the set of perspectives).

Page 22: conceptual coherence in the generation of referring expressions

evaluation

Do people prefer coherence over brevity?– (Two Gricean maxims: “Be brief” vs. “Be orderly”)

Method: subjects (N = 39) shown 6 discourses. – Each discourse introduces 3 entities– Followed by 2 possible continuations– Subjects had to indicate their preferred continuation

Each of the 6 discourses represented a condition: – Brevity: descriptions equally (in-)coherent, but one is brief– Coherence: descriptions equally (non-)brief; only one is

coherent– Trade-off: coherent description is non-brief

Page 23: conceptual coherence in the generation of referring expressions

Example: the domain

Three old manuscripts were auctioned at Sotheby’s:

e1: One of them is a book, a biography of a composer

e2: The second, a sailor’s journal, was published in the form of a pamphlet. It is a record of a voyage.

e3: The third, another pamphlet, is an essay by Hume

Page 24: conceptual coherence in the generation of referring expressions

Intuitively, this is about texts– of different genres (e.g., essay)– published in different forms (e.g., pamphlet)

Of course our corpus-based model doesn’t use these concepts …

Page 25: conceptual coherence in the generation of referring expressions

Example: continuations:

(+c,-b) The biography, the journal and the essay were sold to a collector

(+c,+b) The book and the pamphlets were sold to a collector

(-c,+b) The biography and the pamphlets were sold to a collector

(-c,-b) The book, the record and the essay were sold to a collector

Page 26: conceptual coherence in the generation of referring expressions

results: no preference for brevity

both descriptions coherentx2 = .023, p = .8

both descriptions non-coherentx2 = .64, p = .4

Page 27: conceptual coherence in the generation of referring expressions

results: preference for coherence

both descriptions minimalx2 = 16.03, p < .001

both descriptions non-minimalx2 = 13.56, p < .001

Page 28: conceptual coherence in the generation of referring expressions

results: trade-off

x2 = 39.0, p < .001

Finally, (+c,-b) preferred over (-c,+b)

In other words Coherence was more important than brevity In fact, brevity made no difference at all!

– we did not confirm that +b is preferred over –b

Page 29: conceptual coherence in the generation of referring expressions

Conclusion

When it’s impossible to use the same perspective, use perspectives that are similar

A version of Grice’s maxim “be orderly”?

Page 30: conceptual coherence in the generation of referring expressions

Methodology

Many experiments were done– to find a suitable notion of similarity/coherence– to discover how coherence and brevity relate

Different algorithmic interpretations would be possible

Algorithms are almost always under-determined by the empirical evidence

Page 31: conceptual coherence in the generation of referring expressions

A limitation

Ambiguity/polysemy is not taken into account For example, we might generate

– “the river and the/its bank”

These issues investigated in Imtiaz Khan’s PhD project

One remark: “river” might disambiguate “bank”

Page 32: conceptual coherence in the generation of referring expressions

An open question

Why doesn’t coherence play the same role for modifiers as for types?

Page 33: conceptual coherence in the generation of referring expressions