next up previous contents
Next: Resources from MemoData Up: Lexical Semantic Resources Previous: GLDB - The Göteborg

Subsections

  
Wordnets

   
Introduction

WordNet1.5 is a generic lexical semantic network developed at Princeton University [Mil90a] structured around the notion of synsets. Each synset comprises one or more word senses with the same part of speech which are considered to be identical in meaning, further defined by a gloss, e.g.:

A synset represents a concept and semantic relations are expressed mostly between concepts (except for antonymy and derivational relations). The relations are similar to the lexical semantic relations between word senses as described by [Cru86] (see § 2.7).

EuroWordNet [Vos97a] is a multilingual database containing several monolingual wordnets structured along the same lines as Princeton WordNet1.5: synsets with basic semantic relations. The languages covered in EuroWordNet are: English, Dutch, German, Spanish, French, Italian, Czech and Estonian. In addition to the relations between the synsets of the separate languages there is also an equivalence relation for each synset to the closest concept from an Inter-Lingual-Index (ILI). The ILI contains all WordNet1.5 synsets extended with any other concept needed to establish precise equivalence relations across synsets. Via the ILI it is possible to match synsets from one wordnet to another wordnet (including WordNet1.5). Such mapping may be useful for cross-language Information-Retrieval, for transfer of information and for comparing lexical semantic structures across wordnets.

   
The Princeton WordNet1.5

General information on the size and coverage of WordNet1.5 is given in Table 3.6, taken from [Die96b]:


  Table 3.6: Numbers and figures for WordNet1.5 (* the synonymy relation is included in the notion of synset and is not counted here; ** the relations/sense is here calculated for synsets, because most relations apply to the synsets as a whole)
  All PoS Nouns Verbs Adjectives Adverbs Other
Number of Entries 126520 87642 14727 19101 5050 0
Number of Senses 168217 107484 25768 28762 6203 0
Senses/Entry 1.33 1.23 1.75 1.51 1.23  
Morpho-Syntax     Yes      
Synsets yes          
- Number of Synsets 91591 60557 11363 16428 3243 0
- Synonyms/Synset 1.84 1.77 2.27 1.75 1.91  
Sense Indicators Yes yes yes yes yes  
- Indicator Types 1 1 1 1 1  
- Indicator Tokens 76705 51253 8847 13460 3145  
- Indicators/Sense 0.46 0.48 0.34 0.47 0.51  
Semantic Network            
- Relation Types* 13 10 6 5 2  
- Relation Tokens 128313 80735 13321 30659 3598  
- Relations/Sense** 1.4 1.3 1.1 1.8 1.1  
- Number of Tops 584 11 573      
Semantic Features No          
Multilingual Relations No          
Argument Structure Yes          
- Semantic Roles No          
Semantic Frames Yes          
- Frame Types 35   35      
Selection Restrictions Yes          
- Restriction Types 2   2      
Domain Labels No          
Register Labels No          
 

The following relations are distinguished between synsets:

Synonyms:
members of the synset which are equal or very close in meaning, e.g.
{man 1, adult male}
Antonyms:
synsets which are opposite in meaning, e.g.
{man, adult male} ==> {woman, adult female}
Hyperonyms:
synsets which are the more general class of a synset, e.g.
{man, adult male} ==> {male, male person}
Hyponyms:
synsets which are particular kinds of a synset, e.g.
{weather, atmospheric condition, elements} ==> {cold weather, cold snap, cold wave, cold spell}
{fair weather, sunshine, temperateness}
{hot weather, heat wave, hot spell}
Holonyms:
synsets which are the whole of which a synset is a part.
[Part of] e.g.,
{flower, bloom, blossom} PART OF: {angiosperm, flowering plant}
[Member of] e.g.,
{homo, man, human being, human} MEMBER OF: {genus Homo}
[Substance of] e.g.,
{glass} SUBSTANCE OF: {glassware, glasswork}
{plate glass, sheet of glass}
Meronyms:
synsets which are the parts of a synset.
[Has Part] e.g.,
{perianth, floral envelope}
{flower, bloom, blossom} HAS PART: {stamen}
{pistil, gynoecium}
{carpel}
{ovary}
{floral leaf}
[Has Member] e.g.,
{womankind} HAS MEMBER: {womanhood, woman}
[Has Substance] {glassware, glasswork} HAS SUBSTANCE: {glass}
Entailments:
synsets which are entailed by the synset, e.g.
{walk, go on foot, foot, leg it, hoof, hoof it} {step, take a step}
Causes:
synsets which are caused by the synset, e.g.
{kill} {die, pip out, decease, perish, go, exit, pass away, expire}
Value of:
(adjectival) synsets which represent a value for a (nominal) target concept. e.g.
poor VALUE OF: {financial condition, economic condition}
Has Value:
(nominal) synsets which have (adjectival) concept as values, e.g.
size {large, big}
Also see:
Related synsets, e.g.
{cold} Also See $\rightarrow$ {cool, frozen}
Similar to:
Peripheral or Satellite adjective synset linked to the most central (adjectival) synset, e.g.
{damp, dampish, moist} SIMILAR TO: {wet}
Derived from:
Morphological derivation relation with a synset, e.g.
{coldly, in cold blood, without emotion} Derived from adj $\rightarrow$ {cold}
In the description of WordNet1.5 ([Fel90]), troponymy is discussed as a separate relation. It is restricted to verbs referring to specific manners of change. However, in the database is it is represented in the same way as hyponymy. In the description given here verb-hyponymy is thus eqquivalent to troponymy. Finally, multiple hyperonyms are possible but have not been encoded exhaustively.

Table 3.7 gives the distribution of the relations for each Part of Speech in terms of the number of synsets.

 
Table 3.7: Numbers and figures for WordNet1.5
Relation Nouns Verbs Adjectives Adverbs
Antonym 1713 1025 3748 704
Hyponym 61123 10817 0 0
Mero-member 11472 0 0 0
Mero-sub 366 0 0 0
Mero-part 5695 0 0 0
Entailment 0 435 0 0
Cause 0 204 0 0
Also-See 0 840 2686 0
Value of 1713 0 636 0
Similar to 0 0 20050 0
Derived from 0 0 3539 2894
Total 82082 13321 30659 3598
 

The semantic network is distributed over different files representing some major semantic clusters per parts-of-speech:

Within each of these files there may be one or more synsets which have no hyperonym and therefore represent the tops of the network. In the case of nouns there are only 11 tops or unique- beginners, in the case of verbs 573 tops.
Noun Tops in WordNet1.5
entity
something having concrete existence; living or nonliving
psychological feature
a feature of the mental life of a living organism
abstraction
a concept formed by extracting common features from examples
location, space
a point or extent in space
shape, form
the spatial arrangement of something as distinct from its substance
state
the way something is with respect to its main attributes; ``the current state of knowledge"; ``his state of health"; ``in a weak financial state"]
event
something that happens at a given place and time
act, humanaction, humanactivity
something that people do or cause to happen
group, grouping
any number of entities (members) considered as a unit
possession
anything owned or possessed
phenomenon
any state or process known through the senses rather than by intuition or reasoning

Whereas semantic relations such as hyponymy and synonymy are strictly paradigmatic relations, other relations such as meronymy and cause can be seen as syntagmatic relations imposing a preference relation between word senses:

Meronymy
the head of a lion
Cause
she died because he killed her

Finally, provisional argument-frames are stored for verbs. These frames provide the constituent structure of the complementation of a verb, where --s represents the verb and the left and right strings the complementation pattern:



Verb-frames in WordNet1.5

1.
Something --s
2.
Somebody --s
3.
It is --ing
4.
Something is --ing PP
5.
Something --s something Adjective/Noun
6.
Something --s Adjective/Noun
7.
Somebody --s Adjective
8.
Somebody --s something
9.
Somebody --s somebody
10.
Something --s somebody
11.
Something --s something
12.
Something --s to somebody
13.
Somebody --s on something
14.
Somebody --s somebody something
15.
Somebody --s something to somebody
16.
Somebody --s something from somebody
17.
Somebody --s somebody with something
18.
Somebody --s somebody of something
19.
Somebody --s something on somebody
20.
Somebody --s somebody PP
21.
Somebody --s something PP
22.
Somebody --s PP
23.
Somebody's (body part) --s
24.
Somebody --s somebody to INFINITIVE
25.
Somebody --s somebody INFINITIVE
26.
Somebody --s that CLAUSE
27.
Somebody --s to somebody
28.
Somebody --s to INFINITIVE
29.
Somebody --s whether INFINITIVE
30.
Somebody --s somebody into V-ing something
31.
Somebody --s something with something
32.
Somebody --s INFINITIVE
33.
Somebody --s VERB-ing
34.
It --s that CLAUSE
35.
Something --s INFINITIVE

The distinction between human (Somebody) and non-human (Something) fillers of the frame-slots represents a shallow type of selection restriction.

The data in WordNet1.5. is stored in two separate files for each part of speech. The data file contains all the information for the synsets, where a file-offset position identifies the synset in the file. In the next example, the synset for entity is given: 


00002403 03 n 01 entity 0 013 

~ 00002728 n 0000 

~ 00003711 n 0000 

~ 00004473 n 0000 

~ 00009469 n 0000 

~ 01958400 n 0000 

~ 01959683 n 0000 

~ 02985352 n 0000 

~ 05650230 n 0000 

~ 05650477 n 0000 

~ 05763289 n 0000 

~ 05763845 n 0000 

~ 05764087 n 0000 

~ 05764262 n 0000 

| something having concrete existence; living or nonliving
The first line in this example starts with the file-offset number, which uniquely identifies a synset within a part-of-speech file. It is followed by a reference to the global semantic cluster (03 = noun.animal), the part-of-speech, the size of the synset, a verbal synset name, a sense number of the verbal synset name, and the number of relations. On the next lines the related synsets are given where the symbol indicates the type of relation, which is followed by a file-offset identifying the target synset, its part-of-speech and a number code for relations holding between synset members only. The final line contains the gloss. Verbal synsets may have an additional number for the verb-frames attached to it. A separate index-file then contains a list of lemmas with references to the synsets in which they occur: 

abacus n 2 1 @ 2 02038006 02037873  

abandoned\_infant n 0 1 @ 1 06098838  

abandoned\_person n 0 2 @ ~ 1 05912614  

abandoned\_ship n 0 1 @ 1 02038160  

abandonment n 0 2 @ ~ 3 00116137 00027945 00051049
Here, each word lemma is followed by a part-of-speech code, the polysemy rate (0 if only 1 sense), a number indicating the number of different relations a word has, a list of the relation types (@ ) and a number indicating the number of synsets. Finally, the actual synsets are listed as file-off-set positions.

  
EuroWordNet

As indicated in Table 3.8, the size of the wordnets in EuroWordNet will be (as it is still in development) between 15,000-30,000 synsets and 30,000-50,000 word senses per language. The vocabulary is limited to general language but some subvocabulary is included for demonstration purposes. The information is limited to nouns and verbs. Adjectives and adverbs are only included in so far they are related to the nouns and verbs. Since the wordnets are still under development we cannot complete quantitative data.

 
Table 3.8: Numbers and figures for EuroWordNet
  All PoS
Number of Entries 20.000
Number of Senses 50.000
Senses/Entry 2.5
Morpho-Syntax no
Synsets  
- Number of Synsets 30.000
- Synonyms/Synset 1.7
Sense Indicators  
- Indicator Types  
Semantic Network yes
- Relation Types 46
- Number of Tops 1
Semantic Features yes
- Feature Types 63
- Feature Tokens 1024
Multilingual Relations yes
- Relation Types 17
Argument Structure yes
- Semantic Roles yes
- Role Types 8
Semantic Frames no
Selection Restrictions no
Domain Labels yes
Register Labels yes
 

The data in EuroWordNet is divided into separate modules:

   
The Language Modules

The following information is then stored for each synset in the language-specific wordnets (the Language Modules):
Part of Speech
Noun, Verb, Adjective or Adverb
Synset
Set of synonymous word meanings (synset members)
Language-internal relations
to one or more target synsets
Language-external relations
to one or more ILI-records
Each of the synset-members represents a word sense for which further information can be specified: Most of this information for the synset-members or variants is optional.

The language-internal relations

The most basic semantic relations, such as synonymy, hyponymy and meronymy, have been taken over from WordNet1.5. Some relations have been added to capture less-clear cases of synonymy, to be able to relate equivalences across parts-of-speech (so-called XPOS- relations), to deal with meronymy-relations between events (SUBEVENT), and to express role-relations between nouns and verbs (ROLE/INVOLVED relations):

Paradigmatic relations in EuroWordNet

Syntagmatic relations in EuroWordNet

The syntagmatic relations in the above list can be seen as specification of a potential semantic context for a word, where especially the role-relations may coincide with grammatical contexts as well. As is the case for WordNet1.5, multiple hyperonyms are also allowed in EuroWordNet,just as the other relations (multiple meronyms, holonyms, causes, subevents, etc.). Furthermore, relations can be augmented with specific features to differentiate the precise semantic implication expressed:

The language external relations

The equivalence relations are used to link the language-specific synset to the Inter-Lingual-Index or ILI. The relations parallel the language-internal relations:

Eq_synonym is the most important relation to encode direct equivalences. However, when there is no direct equivalence the synset is linked to the most informative and closest concept using one of the complex equivalence relations. Eq_near_synonym is used when a single synset links to multiple but very similar senses of the same target word (this may be the result of inconsistent sense-differentiation across resources). Has_eq_hyperonym and has_eq_hyponym are typically used for gaps, when the closest target synsets are too narrow or too broad. The other relations are only used when the closest target concept cannot be related by one of the previous relations.

Below is an example of the Dutch synset (aanraking; beroering: touch as a Noun) in the EuroWordNet database import format: 


0 WORD_MEANING

  1 PART_OF_SPEECH "n"

  1 VARIANTS

    2 LITERAL "aanraking"

      3 SENSE 1

      3 DEFINITION "het aanraken"

        4 FEATURE "Register"

          5 FEATURE_VALUE "Usual"

      3 EXTERNAL_INFO

        4 CORPUS_ID 1

          5 FREQUENCY 1026

        4 SOURCE_ID 1

          5 NUMBER_KEY 1336

    2 LITERAL "beroering"

      3 SENSE 2

        4 FEATURE "Date"

          5 FEATURE_VALUE "Old-fashioned"

      3 EXTERNAL_INFO

        4 CORPUS_ID 1

          5 FREQUENCY 238

        4 SOURCE_ID 1

          5 NUMBER_KEY 401472

  1 INTERNAL_LINKS

    2 RELATION "XPOS_NEAR_SYNONYM"

      3 TARGET_CONCEPT

        4 PART_OF_SPEECH "v"

        4 LITERAL "aanraken"

          5 SENSE 1

      3 SOURCE_ID 1001

    2 RELATION "HAS_HYPERONYM"

      3 TARGET_CONCEPT

        4 PART_OF_SPEECH "n"

        4 LITERAL "beweging"

          5 SENSE 1

      3 SOURCE_ID 1001

    2 RELATION "CAUSES"

      3 TARGET_CONCEPT

        4 PART_OF_SPEECH "n"

        4 LITERAL "contact"

          5 SENSE 1

      3 SOURCE_ID 1001

    2 RELATION "XPOS_NEAR_SYNONYM"

      3 TARGET_CONCEPT

        4 PART_OF_SPEECH "v"

        4 LITERAL "raken"

          5 SENSE 2

      3 SOURCE_ID 1001

  1 EQ_LINKS

    2 EQ_RELATION "EQ_SYNONYM"

      3 TARGET_ILI

        4 PART_OF_SPEECH "n"

        4 FILE_OFFSET 69655

      3 SOURCE_ID 1002

   
The Inter-Lingual-Index

The ILI is not internally structured: no lexical semantic relations are expressed between the ILI-records. In this respect it should not be seen as a language-neutral ontology but only as a linking-index between wordnets3.1. The Inter-Lingual-Index is thus basically a list of ILI-records, with the only purpose to provide a matching across wordnets. In addition, it also provides access to the language-neutral modules by listing the Top Concepts and Domain labels that may apply to it.

Simple ILI-records contain the following information fields:

Most information is optional. The Top Concepts and Domains linked to an ILI-record can be transferred to the synsets in the local wordnets that are linked to the same ILI-record, as is illustrated in the next schema: 


ES wordnet language-specific-word-meaning

   |___

    ___> eq_synonym-> ILI-record -has_top_concept-> Top Concept

   |

IT wordnet: language-specific-word-meaning

In addition to the Simple ILI-records, there are Complex ILI-records which group closely related meanings. These groupings are based on systematic polysemy relations between meanings, such as specialization of more general meanings, metonymy (§2.7, 3.10.2) and diathesis alternations (§2.6.2). Complex ILI-records are needed to provide a better linking between the wordnets. Inconsistent sense-differentiation across resources often makes it very difficult to find exact equivalences across the resources. By linking different meaning realizations (e.g. university as a building and as the institute) to same complex ILI-records it is still possible to find the closely related meanings.

Below is an example of a complex ILI-record in which specific meanings of car are grouped by a new generalized meaning: 


0 ILI_RECORD

  1 PART_OF_SPEECH "n"

  1 NEW_ILI_ID 1234

  1 GLOSS "a 4-wheeled vehicle"

  1 VARIANTS

    2 LITERAL "car"

      3 SENSE 2

    2 LITERAL "automobile"

      3 SENSE 1

  1 EQ_RELATION "eq_generalization"

    2 ILI_RECORD

      3 FILE_OFFSET 54321

    2 ILI_RECORD

      3 NEW_ILI_ID 9876

Here, eq_generalization expresses the relation that holds with two more specific ILI-records, identified by FILE_OFFSET and NEW_ILI_ID respectively. The former indicates that it originates from WordNet1.5, the latter that it has been added as a new concept in EuroWordNet. These sense-groupings apply cross-linguistically, although the lexicalization of these meanings can differ from language to language.

  
Top Ontology

The EuroWordNet top ontology contains 63 concepts. It is developed to classify a set of so- called Base Concepts extracted from the Dutch, English, Spanish and Italian wordnets that are being developed. These Base Concepts have most relations and occupy high positions in the separate wordnets, as such making up the core of the semantic networks. The Base Concepts are specified in terms of WordNet1.5 synsets in the ILI.

The top-ontology incorporates the top-levels of WordNet1.5, ontologies developed in EC-projects Acquilex (BRA 3030, 7315) and Sift (LE-62030)[Vos96], Qualia-structure [Pus95a], Aktions-Art distinctions [Ven67], [Ver72], [Ver89], [Pus91b],and entity orders [Lyo77]. Furthermore, the ontology has been adapted to group the Base Concepts into coherent semantic clusters. The ontology combines notions described in §2.2, 2.7, and 2.5. Important characteristics of the ontology are:

The Top Concepts are more like semantic features than like common conceptual classes. We typically find Top Concepts for Living and for Part but we do not find a Top Concept Bodypart, even though this may be more appealing to a non-expert. BCs representing body parts are now cross-classified by two feature-like Top Concepts Living and Part. The main reason for this is that a more flexible system of features is needed to deal with the diversity of the Base Concepts.

The top-concepts are structured according to the hierarchy shown in Fig 3.2.

  
Figure 3.2: Hierachy of Top Concepts in EuroWordNet

Following [Lyo77], the first level the ontology is differentiated into 1stOrderEntity, 2ndOrderEntity, 3rdOrderEntity. According to Lyons, 1stOrderEntities are publicly observable individual persons, animals and more or less discrete physical objects and physical substances. They can be located at any point in time and in, what is at least psychologically, a three-dimensional space. The 2ndOrderEntities are events, processes, states-of-affairs or situations which can be located in time. Whereas 1stOrderEntities exist in time and space 2ndOrderEntities occur or take place, rather than exist. The 3rdOrderEntities are propositions, such as ideas, thoughts, theories, hypotheses, that exist outside space and time and which are unobservable. They function as objects of propositional attitudes, and they cannot be said to occur or be located either in space or time. Furthermore, they can be predicated as true or false rather than real, they can be asserted or denied, remembered or forgotten, they may be reasons but not causes.

List of Top Ontology concepts in EuroWordNet with definitions
Top
all
1stOrderEntity
Any concrete entity (publicly) perceivable by the senses and located at any point in time, in a three-dimensional space.
2ndOrderEntity
Any Static Situation (property, relation) or Dynamic Situation, which cannot be grasped, heard, seen, felt as an independent physical thing. They can be located in time and occur or take place rather than exist; e.g. continue, occur, apply.
3rdOrderEntity
An unobservable proposition which exists independently of time and space. They can be true or false rather than real. They can be asserted or denied, remembered or forgotten, e.g. idea, thought, information, theory, plan.
Origin
Considering the way concrete entities are created or come into existence.
Natural
Anything produced by nature and physical forces as opposed to artifacts.
Living
Anything living and dying including objects, organic parts or tissue, bodily fluids; e.g. cells; skin; hair, organism, organs.
Human
e.g. person, someone.
Creature
Imaginary creatures; e.g. god, Faust, E.T..
Animal
e.g. animal, dog.
Plant
e.g. plant, rice.
Artifact
Anything manufactured by people as opposed to natural.
Form
Considering the shape of concrete entities, fixed as an object or a-morf as a substance
Substance
all stuff without boundary or fixed shape, considered from a conceptual point of view not from a linguistic point of view; e.g. mass, material, water, sand, air.
Solid
Substance which can fall, does not feel wet and you cannot inhale it; e.g. stone, dust, plastic, ice, metal
Liquid
Substance which can fall, feels wet and can flow on the ground; e.g. water, soup, rain.
Gas
Substance which cannot fall, you can inhale it and it floats above the ground; e.g. air, ozone.
Object
Any conceptually-countable concrete entity with an outer limit; e.g. book, car, person, brick.
Composition
Considering the composition of concrete entities in terms of parts, groups and larger constructs
Part
Any concrete entity which is contained in an object, substance or a group; head, juice, nose, limb, blood, finger, wheel, brick, door.
Group
Any concrete entity consisting of multiple discrete objects (either homogeneous or heterogeneous sets), typically people, animals, vehicles; e.g. traffic, people, army, herd, fleet.
Function
Considering the purpose, role or main activity of a concrete entity. Typically it can be used for nouns that can refer to any substance, object which is involved in a certain way in some event or process; e.g. remains, product, threat.
Vehicle
e.g. car, ship, boat.
Software
e.g. computer programs and databases.
Representation
Any concrete entity used for conveying a message; e.g. traffic sign, word, money.
Place
Concrete entities functioning as the location for something else; e.g. place, spot, centre, North, South.
Occupation
e.g. doctor, researcher, journalist, manager.
Instrument
e.g. tool, machine, weapon
Garment
e.g. jacket, trousers, shawl
Furniture
e.g. table, chair, lamp.
Covering
e.g. skin, cloth, shield.
Container
e.g. bag, tube, box.
Comestible
food and drinks, including substances, liquids and objects.
Building
e.g. house, hotel, church, office.
MoneyRepresentation
Physical Representations of value, or money; e.g. share, coin.
LanguageRepresentation
Physical Representations conveyed in language (spoken, written or sign language); e.g. text, word, utterance, sentence, poem.
ImageRepresentation
Physical Representations conveyed in a visual medium; e.g. sign language, traffic sign, light signal.
SituationType
Considering the predicate-inherent Aktionsart properties of Situations: dynamicity and boundedness in time. Subclasses are disjoint, every 2ndOrderEntity has only 1 SituationType.
Static
Situations (properties, relations and states) in which there is no transition from one eventuality or situation to another: non- dynamic; e.g. state, property, be.
Relation
Static Situation which applies to a pair of concrete entities or abstract Situations, and which cannot exist by itself without either one of the involved entities; e.g. relation, kinship, distance, space.
Property
Static Situation which applies to a single concrete entity or abstract Situation; e.g. colour, speed, age, length, size, shape, weight.
Dynamic
Situations implying either a specific transition from a state to another (Bounded in time) or a continuous transition perceived as an ongoing temporally unbounded process; e.g. event, act, action, become, happen, take place, process, habit, change, activity.
UnboundedEvent
Dynamic Situations occurring during a period of time and composed of a sequence of (micro-)changes of state, which are not perceived as relevant for characterizing the Situation as a whole; e.g. grow, change, move around, live, breath, activity, hobby, sport, education, work, performance, fight, love, caring, management.
BoundedEvent
Dynamic Situations in which a specific transition from one Situation to another is implied; Bounded in time and directed to a result; e.g. to do, to cause to change, to make, to create.
SituationComponent
Considering the conceptual components that play a role in Situations. Situations can be cross-classified by any number of Situation Components
Cause
Situations involving causation of Situations (both Static and Dynamic); result, effect, cause, prevent.
Stimulating
Situations in which something elicits or arouses a perception or provides the motivation for some event, e.g. sounds, such as song, bang, beep, rattle, snore, views, such as smell, appetizing, motivation.
Phenomenal
Situations that occur in nature controlled or uncontrolled or considered as a force; e.g. weather, chance.
Agentive
Situations in which a controlling agent causes a dynamic change; e.g. to kill, to do; to act.
Usage
Situations in which something (an instrument, substance, time, effort, force, money) is or can be used; e.g. to use, to spent, to represent, to mean, to be about, to operate, to fly, drive, run, eat , drink, consume.
Time
Situations in which duration or time plays a significant role; Static e.g. yesterday, day, pass, long, period, Dynamic e.g. begin, end, last, continue.
Social
Situations related to society and social interaction of people: Static e.g. employment, poor, rich, Dynamic e.g. work, management, recreation, religion, science.
Quantity
Situations involving quantity and measure ; Static e.g. weight, heaviness, lightness; changes of the quantity of first order entities; Dynamic e.g. to lessen, increase, decrease.
Purpose
Situations which are intended to have some effect.
Possession
Situations involving possession; Static e.g. have, possess, possession, contain, consist of, own; Dynamic changes in possession, often to be combined which changes in location as well; e.g. sell, buy, give, donate, steal, take, receive, send.
Physical
Situations involving perceptual and measurable properties of first order entities; either Static e.g. health, a colour, a shape, a smell; or Dynamic changes and perceptions of the physical properties of first order entities; e.g. redden, thicken, widen, enlarge, crush, form, shape, fold, wrap, thicken, to see, hear, notice, smell.
Modal
Situations (only Static) involving the possibility or likelihood of other situations as actual situations; e.g. abilities, power, force, strength.
Mental
Situations experienced in mind, including emotional and attitudinal situations; a mental state is changed; e.g. invent, remember, learn, think, consider.
Manner
Situations in which the way or manner plays a role. This may be Manner incorporated in a dynamic situation, e.g. ways of movement such as walk, swim, fly, or the static Property itself: e.g. manner, sloppy, strongly, way.
Location
Situations involving spatial relations; static e.g. level, distance, separation, course, track, way, path; something changes location, irrespective of the causation of the change; e.g. move, put, fall, drop, drag, glide, fill, pour, empty, take out, enter.
Experience
Situations which involve an experiencer: either mental or perceptual through the senses.
Existence
Situations involving the existence of objects and substances; Static states of existence e.g. exist, be, be alive, life, live, death; Dynamic changes in existence; e.g. kill, produce, make, create, destroy, die, birth.
Condition
Situations involving an evaluative state of something: Static, e.g. health, disease, success or Dynamic e.g. worsen, improve.
Communication
Situations involving communication, either Static, e.g. be about or Dynamic (Bounded and Unbounded); e.g. speak, tell, listen, command, order, ask, state, statement, conversation, call.

The Top Concepts have been applied to the 1024 Base Concepts, distributed as shown in Table 3.9.

 
Table 3.9: Distribution of Base Concepts in EuroWordNet
  Nouns Verb Total
1stOrderEntities 491   491
2ndOrderEntities 272 228 500
3rdOrderEntities 33   33
Total 796 228 1024
 

As suggested above the Base Concepts are typically classified in terms of several top concepts.

The ontology thus should be seen as a partial lattice in which distinctions can be combined. In total 450 clusters of features have been used to classify 1014 Base Concepts. Below are examples of top concept conjunctions for the Base Concepts (note that some classifications may seem odd because they apply to rather specific senses of words):