division of semantic labor in the global wordnet grid piek vossen, vu university amsterdam german...

Division of semantic labor in

the Global WordNet Grid

Piek Vossen, VU University Amsterdam

German Rigau, University of the Basque Country

5th Global Wordnet Conference

Mumbai, India, Jan 30 – Feb 5, 2010

Overview

• KYOTO as a domain implementation of the Global Wordnet Grid

• Scope of knowledge integration• Division of linguistic labor• How to integrate resources?• How to make inferences?

KYOTO – some statistics• European-Asian project

• March 2008 – March 2011

• 7 countries (The Netherlands, Italy, Germany, Spain, Taiwan, Japan, Czech Republic)

• 12 sites

– Universities & research institutes: VUA, CNR-ILC, CNR-IIT, BBAW, EHU, AS, NICT, Masaryk

– Companies: Synthema, Irion

– User organizations: ECNC, WWF

• 7 languages (English, Italian, Japanese, Dutch, Spanish, Basque, Chinese)

Semantic & Syntacticrepresentation

Kyoto Annotation Format

Multilingual Knowledge Base

Fact Base Term Base

Linguistic Processor

12

Fact ExtractorKybot

Term Extractor

Tybot

Wiki Editor

Wikyoto

Wordnets & Ontology

Semantic & Syntacticrepresentation

Kyoto Annotation Format

Multilingual Knowledge Base

Fact Base Term Base

Linguistic Processor

1122

Fact ExtractorKybot

Fact ExtractorKybot

Term Extractor

Tybot

Term Extractor

Tybot

Wiki Editor

Wikyoto

Wordnets & Ontology

KYOTO – Overall architectureOverview of the KYOTO process

GWC2010, Mumbai

5Applying ontology mappings

GWC2010, Mumbai

6

Gobal Wordnet Grid

Domain

Domain

Domain

Domain

Domain

DomainDomain

Domain

OntologyBase concepts

Wn

Wn

WnWn

Wn

Wn

Wn

DOLCE/SUMOOntoWordnet

Domain

V

GWC2010, Mumbai

7

Available repositories in KYOTO

Environment domain• Term database: 500,000 terms per 1,000 documents per language

• Open data project:

– DBPedia: 2.6 million things, including at least 213,000 persons, 328,000 places, 57,000 music albums, 36,000 films, 20,000 companies. The knowledge base consists of 274 million pieces of information (RDF triples)

– GeoNames: 8 million geographical names and consists of 6.5 million unique features whereof 2.2 million populated places and 1.8 million alternate names

• Domain thesauri and taxonomies: Species 2000: 2,1 million species

• Wordnets for 7 languages: about 50,000 to 120,000 synsets per language

• Ontologies: SUMO, DOLCE, SIMPLE

GWC2010, Mumbai

8

Domain

T

Domain TV

Domain

TV

Domain

V

T

DomainT T

Domain

Species

Domain

Domain

Domain

Domain

Kyoto Knowledge Base

OntologyOntologyBase concepts

Wn

Wn

WnWn

Wn

Wn

Wn

DBPedia

Terms

500K

2,100K

DOLCE/SUMOOntoWordnet

Terms

500K

Species

2,100K

Domain

V

GWC2010, Mumbai

9

Species in the ontology

- Implies to store 2.1 million species twice in the ontology.

GWC2010, Mumbai

10

Should all knowledge be stored in the central ontology?

• Vocabularies are too large for full inferencing with current reasoners

• Vocabularies are linguistically too diverse to be represented in an ontology

• Inferencing capabilities of formal ontologies is not needed for all levels of knowledge

GWC2010, Mumbai

11

Modeling knowledge in a domain

• Knowledge needs to be divided over different lexical and ontological layers:– Precisely define the relations between lexical and

ontological layers– Precisely define the inferencing based on the

distributed knowledge layers

GWC2010, Mumbai

12

Division of linguistic labor principle

• Putnam 1975: – No need to know all the necessary and

sufficient properties to determine if something is "gold"

– Assume that there is a way to determine these properties and that domain experts know how to recognize instances of these concepts.

– Speakers can still use the word "gold" and communicate useful information

GWC2010, Mumbai

13

Division of semantic labor principle

• Digital version of Putnam (1975): – Computer does not need to have all the necessary and

sufficient properties to determine if something is a "European tree frog"

– Computer assumes that there is a way to determine this and that domain experts (people) know how to recognize instances of these concepts.

– Computers can still reason with semantics and do useful stuff with textual data

GWC2010, Mumbai

14

What does the computer need to know?

• Distinction between rigid and non-rigid (Welty & Guarino 2002):– being a "cat" is essential to individual's existence and

therefore rigid– being a "pet" is a temporarily role and therefore non-

rigid; a cat can become a pet and stop being a pet without ceasing to exist

– Felix is born as a cat and will always be a cat, but during some period Felix can become a pet and stop being a pet while he continuous to exist as a cat

• All 2.1 million species are rigid concepts

GWC2010, Mumbai

15

What does the computer need to know?

• Roles and processes in documents have more information value than the defining properties of species:– Species defined in terms of physical properties already

known to expert;– Roles such as "invasive species", "migration species",

"threatened species" express THE important properties of instances of species

• Roles are typically the terms we learn from the text not the species!

GWC2010, Mumbai

16

Wordnet-ontology-relations• Rigid synset relations to ontology:

– Synset:Endurant(Object); Synset:Perdurant(Event); Synset:Quality:

– sc_equivalenceOf (= relation in WN-SUMO) or sc_subclassOf (+ relation in WN-SUMO)

• Non-rigid synset relations to ontology:– Synset:Role; Synset:Endurant(Object); Synset:Perdurant(Event)– sc_domainOf: range of ontology types that restricts a role– sc_playRole: role that is being played– sc_participantOf: the process in wich the role is played

• Rigidity can be detected automatically (Rudify, 80% precision, IAG 80%) and is stored in wordnets as attributes to synsets

Global Wordnet Grid Model

perdurant

change-of-location

migration

endurant

object

organism

bird

role

done-by

has-sourcehas-destinationhas-path

some

hashas

bird_1_N sc_equivalentOf birdrigid

English Wordnet in WN-LMF KYOTO Ontology in OWL-DL(Extension of DOLCE LT)

migration_bird_1_N sc_domainOf bird non-rigid sc_playRole done-by

sc_participantOf migration

migration_4_N sc_equivalentOf migrationmigrate_1 _V sc_equivalentOf migration

duck_1_N, rigid

hyponymsubclass

subclass

subclass

subclass

subclass

subclass

Global Wordnet Grid Model

perdurant

change-of-location

migration

endurant

object

organism

bird

role

done-by

has-sourcehas-destinationhas-path

some

hashas

bird_1_N sc_equivalentOf birdrigid

English Wordnet in WN-LMF KYOTO Ontology in OWL-DL(Extension of DOLCE LT)

migration_bird_1_N sc_domainOf bird non-rigid sc_playRole done-by

sc_participantOf migration

migration_4_N sc_equivalentOf migrationmigrate_1 _V sc_equivalentOf migration

duck_1_N, rigid

subclass

subclass

subclass

subclass

subclass

subclass

Dutch Wordnet migrerende dieren_1_N sc_domainOf organism

(migrating species) sc_playRole done-by non-rigid sc_participantOf migration

equivalent_hypernym eng-30-02356039-n (bird) eend_1_N (duck)

equivalent eng-30-01254614-n (duck)

Spanish Wn, Basque WnItalian Wn, Japanese WnChinese Wn ....

Cross-lingual equivalence mappings are expressed

through wordnet mappings

Cross-lingual equivalence mappings are expressed

through wordnet mappings

Wordnet to ontology mappings

{create, produce, make}Verb, English

-> sc_ equivalenceOf construction

{artifact, artefact}Noun, English

-> sc_domainOf physical_object

-> sc_playRole result-existence

-> sc_participantOf construction

{kunststof}Noun, Dutch // lit. artifact substance

-> sc_domainOf amount_of_matter

-> sc_playRole result-existence

-> sc_participantOf construction

{meat}Noun, English

-> sc_domainOf cow, sheep, pig

-> sc_playRole patient

-> sc_participantOf eat

{ 名肉 , 食物 , 餐 }Noun, Chinese

-> sc_domainOf animal



{ , طعام, لحم Noun, Arabic{غذاء

-> sc_domainOf cow, sheep



Wordnet to ontology mappings

{teacher}Noun, English

-> sc_domainOf human

-> sc_playRole done-by

-> sc_participantOf teach

{leraar}Noun, Dutch // lit. male teacher

-> sc_domainOf man



{lerares}Noun, Dutch // lit. female teacher

-> sc_domainOf woman



Wordnet-LMF<LexicalEntry id="footmark">

<Lemma writtenForm="footmark" partOfSpeech="n"/>

<Sense id="footmark_1" synset="eng-30-06645039-n">

<MonolingualExternalRefs>

<MonolingualExternalRef externalSystem="Wordnet3.0" externalReference="" />

</MonolingualExternalRefs>

</Sense>

</LexicalEntry>

<Synset/>

<SenseAxis/>

<SenseAxis id="sa_ita16-eng30_001" relType="eq_synonym">

<Target ID="ita-16-1251-n" />

<Target ID="eng-30-13480848-n"/>

</SenseAxis>

WN-LMF Synset relations<Synset id="eng-30-06645039-n" baseConcept="0"> 

<Definition gloss="mark of a foot or shoe on a surface">

<Statement example="the police made casts of the footprints in the soft earth outside the window" />

</Definition>

<OntologicalMetaProperties rigidValue=”1”>

<rigid score=”0.57” author=”Rudify1.0” date="2008-07-01">

<non-rigid score=”0.09” author=”Rudify1.0” date="2008-07-01">

</OntologicalMetaProperties>

<SynsetRelations/>


<MonolingualExternalRef externalSystem="SUMO" reference="superficialPart" relType="at"/>

<MonolingualExternalRef externalSystem="KYO" reference="mark" relType="sc_subclassOf"/>


</Synset>

WN-LMF Synset relations<Synset id="eng-30-02356039-n" baseConcept="0"> 

<Definition gloss="birds that migrate in winter to warmer regions"/>

<OntologicalMetaProperties rigidValue=”0”>

<rigid score=”0.00” author=”Rudify1.0” date="2008-07-01">

<non-rigid score=”0.69” author=”Rudify1.0” date="2008-07-01">

</OntologicalMetaProperties>

<SynsetRelations/>


<Statement>

<MonolingualExternalRef externalSystem="KYO" reference="bird" relType="sc_domainOf"/>

<MonolingualExternalRef externalSystem="KYO" reference="done-by" relType="sc_playRole"/>

<MonolingualExternalRef externalSystem="KYO" reference="migration" relType="sc_participantOf"/>

</Statement>


</Synset>

GWC2010, Mumbai

24

Division of labor in knowledge sources

Eleutherodactylus augusti

Eleutherodactylus

Leptodactylidae

Anura

Amphibia

Chordata

Animalia

Eleutherodactylus atrabracus barking frog

frog:1, toad:1, toad frog:1, anuran:1, batrachian:1, salientian:1

amphibian:3

vertebrate:1,craniate:1

chordate:1

animal:1Base Concept

2.1 million species 100,000 synsets 2,000 types

endurant

physical-object

organism

endemic frogendangered frogpoisonous frogalien frog

500,000 terms

Skos database Wordnet-LMF Ontology-OWL-DL

Term database

perdurant

endanger

GWC2010, Mumbai

25

How to make inferences?

• Sparql queries to large Virtuoso databases: Aligned Species 2000, DBPedia

• Sql queries to term database• Graph matching on wordnets stored in

DebVisDic• Reasoning on a small ontology

KYOTO Project meeting, Jan 13-14th 2010, PolyU Hong Kong

26

Ontotagger applied to KAF Apply WSD to every term in the KAF representation of a text For each term in KAF representation of a text:

(a)If wordnet synset (WSD) then check for ontology mappings, if none traverse wordnet hierarchy to find first mapping

(b)Else check the SKOS database for wordnet mapping, if necessary traverse broader relations up to the first wordnet mapping and go to a.)

(c)Else check the term database for wordnet mappings, if necessary traverse parent relations up to the first wordnet mapping and go to a.)

Collect all mappings from the ontology and all (relevant) ontological implications and insert them into the KAF representation of the text.


27

Examples

1. Migration birds in the Humber Estuary.

2. The migration of birds to the Humber Estuary

3. Bird migration in the Humber Estuary

4. Birds that migrate to the Humber Estuary

Annotation of ontological implications in KAF



<term lemma=“migration bird” pos=”N.pl”>

<externalReference> 

<externalRef resource=“ontology" relation=”sc_domainOf” reference=“bird"/>

<externalRef resource=“ontology" relation=“sc_participantOf” reference=“migration"/>

<externalRef resource=“ontology” relation=“sc_playRole" reference=“done-by"/>

<externalRef resource=“ontology" relation=“implied” reference=“ done-by" some=”physical-plurality”/>

<externalRef resource=“ontology" relation=“implied” reference=“ has-destination" some=”particular”/>

<externalRef resource=“ontology" relation=“implied” reference=“ has-source" some=”particular”/>

<externalRef resource=“ontology" relation=“implied” reference=“ has-path" some=”particular”/>

</externalReference>

</term>

<term lemma=”in” pos=”P”/>

<term lemma = “Humber Estuary”><externalRef resource=“ontology” reference=“location"/>

<term lemma=“bird” pos = “N.pl”>


<externalRef resource=“ontology" relation=”sc_equivalentOf” reference=“bird"/>


</term>

<term lemma=“migration” pos=”N”>


<externalRef resource=“ontology" relation=“sc_equivalentOf” reference=“migration"/>






</term>

<term lemma=”in”/>

<term lemma=“bird” pos=”N.pl”>


<externalRef resource=“ontology" relation=”sc_equivalentOf” reference=“bird"/>


</term>

<term lemma=“migrate” pos=”V”>


<externalRef resource=“ontology" relation=“sc_equivalentOf” reference=“migration"/>






</term>

<term lemma=”to”/>


31

Kybot profiles

IF <! bird migration to HE>

T1 + to + T2 &

T1.impliedType="change_of_location" & T1.impliedRole="has-target" &

T2.Type="location"

THEN

<location-target, T1, T2>

IF <! species migration from HE>

T1 + from + T2 &

T1.impliedType="change_of_location" & T1.impliedRole="has-source" &

T2.Type="location"

THEN

<location-source, T1, T2>

Kybot Knowledge Patterns <events>

<event eid="e1" target="t2" lemma="feed" pos="V" tense="PAST"

aspect="NONE" polarity="POS"/>

<event eid="e2" target="t20" lemma="migrate" pos="V" tense="PRESENT"

aspect="NONE" polarity="POS"/>

<role rid="r1" event="e1" target="t1" rtype="agent"/>

<role rid="r2" event="e1" target="t3" rtype="patient"/>

<role rid="r3" event="e1" target="t9" rtype="theme"/>

<role rid="r3" event="e2" target="t21" rtype="agent"/>

<role rid="r4" event="e2" target="t22" rtype="source"/>

<role rid="r5" event="e2" target="t24" rtype="goal"/>

</events>

GWC2010, Mumbai

33

Conclusion: Should all knowledge be stored in the central ontology?

• Vocabularies are too large for full inferencing• Vocabularies are linguistically too diverse to be represented in an

ontology• Inferencing capabilities of formal ontologies is not needed for all

levels of knowledge• A model of division of labor (along the lines of Putnam 1975) in

which knowledge is stored in 3 layers: – SKOS vocabularies and term databases– wordnet (WN-LMF)– ontology (OWL-DL),

• Each layer supports different types of inferencing ranging from Sparql queries, graph algorithms to reasoning.

• Mapping relations that support the division of labour and different types of inferencing and that allow for the encoding of language-specific lexicalizations and restrictions.

Conclusions• Ontologies are abstract and minimal and lexicons are large and rich

• Semantic relations in lexicons are complementary to ontological relations

• Semantic relations expressed in a language system should be compatible with ontologies

• Large vocabularies of types (rigid things in the world) can be mapped to the ontology through combinations of lexical relations and basic ontological mappings

• Lexicalizations of contextual and subjective concepts need to be expressed through more complex relations

• Equivalences across languages partially through ontological expressions and partially across lexicons

Applying WSD to terms

<terms> <term tid=”t1” type=”open” lemma=”bird” pos=”N”> <span id=”w1”/> <externalReferences> 

<externalRef resource="wn30g" ref="eng-30-01855672-n" conf="0.38"/><externalRef resource="wn30g" ref="eng-30-10157744-n" conf="0.31"/><externalRef resource="wn30g" ref="eng-30-07646821-n" conf="0.30"/>

</externalReferences> </term>

</terms>