Grammar Implementation with Lexicalized Tree Adjoining Grammars and - - PowerPoint PPT Presentation

▶
grammar implementation with lexicalized tree adjoining
SMART_READER_LITE
LIVE PREVIEW

Grammar Implementation with Lexicalized Tree Adjoining Grammars and - - PowerPoint PPT Presentation

Dec 14, 2023 15 likes •1.37k views

Grammar Implementation with Lexicalized Tree Adjoining Grammars and Frame Semantics Syntactic analyses Laura Kallmeyer, Timm Lichte, Rainer Osswald & Simon Petitjean University of Dsseldorf DGfS CL Fall School, September 12, 2017 SFB 991

slide-1
SLIDE 1

Grammar Implementation with Lexicalized Tree Adjoining Grammars and Frame Semantics

Syntactic analyses Laura Kallmeyer, Timm Lichte, Rainer Osswald & Simon Petitjean

University of Düsseldorf

DGfS CL Fall School, September 12, 2017

SFB 991

slide-2
SLIDE 2

The ideal grammar formalism

linguistically adequate:

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 2 2

slide-3
SLIDE 3

The ideal grammar formalism

linguistically adequate:

Phenomena: linearization, agreement, discontinuity, ellipsis, coordination (RNR), ...

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 3 2

slide-4
SLIDE 4

The ideal grammar formalism

linguistically adequate:

Phenomena: linearization, agreement, discontinuity, ellipsis, coordination (RNR), ... Generalization: valency, active/passive diathesis, sentence types, alternations, syntax-semantics interface, syntax-discourse interface

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 4 2

slide-5
SLIDE 5

The ideal grammar formalism

linguistically adequate:

Phenomena: linearization, agreement, discontinuity, ellipsis, coordination (RNR), ... Generalization: valency, active/passive diathesis, sentence types, alternations, syntax-semantics interface, syntax-discourse interface intuitive implementation

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 5 2

slide-6
SLIDE 6

The ideal grammar formalism

linguistically adequate:

Phenomena: linearization, agreement, discontinuity, ellipsis, coordination (RNR), ... Generalization: valency, active/passive diathesis, sentence types, alternations, syntax-semantics interface, syntax-discourse interface intuitive implementation

computationally adequate:

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 6 2

slide-7
SLIDE 7

The ideal grammar formalism

linguistically adequate:

Phenomena: linearization, agreement, discontinuity, ellipsis, coordination (RNR), ... Generalization: valency, active/passive diathesis, sentence types, alternations, syntax-semantics interface, syntax-discourse interface intuitive implementation

computationally adequate:

explicit/formalized

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 7 2

slide-8
SLIDE 8

The ideal grammar formalism

linguistically adequate:

Phenomena: linearization, agreement, discontinuity, ellipsis, coordination (RNR), ... Generalization: valency, active/passive diathesis, sentence types, alternations, syntax-semantics interface, syntax-discourse interface intuitive implementation

computationally adequate:

explicit/formalized decidable, maybe even tractable

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 8 2

slide-9
SLIDE 9

The ideal grammar formalism

linguistically adequate:

Phenomena: linearization, agreement, discontinuity, ellipsis, coordination (RNR), ... Generalization: valency, active/passive diathesis, sentence types, alternations, syntax-semantics interface, syntax-discourse interface intuitive implementation

computationally adequate:

explicit/formalized decidable, maybe even tractable bidirectional

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 9 2

slide-10
SLIDE 10

The ideal grammar formalism

linguistically adequate:

Phenomena: linearization, agreement, discontinuity, ellipsis, coordination (RNR), ... Generalization: valency, active/passive diathesis, sentence types, alternations, syntax-semantics interface, syntax-discourse interface intuitive implementation

computationally adequate:

explicit/formalized decidable, maybe even tractable bidirectional

psycholinguistically adequate:

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 10 2

slide-11
SLIDE 11

The ideal grammar formalism

linguistically adequate:

Phenomena: linearization, agreement, discontinuity, ellipsis, coordination (RNR), ... Generalization: valency, active/passive diathesis, sentence types, alternations, syntax-semantics interface, syntax-discourse interface intuitive implementation

computationally adequate:

explicit/formalized decidable, maybe even tractable bidirectional

psycholinguistically adequate:

strictly incremental derivations

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 11 2

slide-12
SLIDE 12

The ideal grammar formalism

linguistically adequate:

Phenomena: linearization, agreement, discontinuity, ellipsis, coordination (RNR), ... Generalization: valency, active/passive diathesis, sentence types, alternations, syntax-semantics interface, syntax-discourse interface intuitive implementation

computationally adequate:

explicit/formalized decidable, maybe even tractable bidirectional

psycholinguistically adequate:

strictly incremental derivations correct predictions wrt. processing complexity

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 12 2

slide-13
SLIDE 13

Outline of today’s course

1

The derivation tree

2

Design principles for elementary trees

3

Sample derivations NP and PP complements Sentential complements and long-distance dependencies Modifiers

4

Feature based TAG

5

Summary and outlook

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 13 3

slide-14
SLIDE 14

Outline of today’s course

1

The derivation tree

2

Design principles for elementary trees

3

Sample derivations NP and PP complements Sentential complements and long-distance dependencies Modifiers

4

Feature based TAG

5

Summary and outlook

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 14 4

slide-15
SLIDE 15

Example derivation

NP Adam VP VP∗ Adv always

S VP NP↓ V ate NP↓

NP NP∗ Det the NP apple

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 15 5

slide-16
SLIDE 16

Example derivation

VP VP∗ Adv always

S VP NP↓ V ate NP Adam

NP NP∗ Det the NP apple

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 16 5

slide-17
SLIDE 17

Example derivation

VP VP∗ Adv always

S VP NP apple V ate NP Adam

NP NP∗ Det the

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 17 5

slide-18
SLIDE 18

Example derivation

VP VP∗ Adv always

S VP NP NP apple Det the V ate NP Adam

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 18 5

slide-19
SLIDE 19

Example derivation

S VP VP NP NP apple Det the V ate Adv always NP Adam

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 19 5

slide-20
SLIDE 20

Derivation trees

TAG derivations are uniquely described by derivation trees. The derivation tree contains:

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 20 6

slide-21
SLIDE 21

Derivation trees

TAG derivations are uniquely described by derivation trees. The derivation tree contains: nodes for all elementary trees used in the derivation, and

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 21 6

slide-22
SLIDE 22

Derivation trees

TAG derivations are uniquely described by derivation trees. The derivation tree contains: nodes for all elementary trees used in the derivation, and edges for all adjunctions and substitutions performed through-

  • ut the derivation, and

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 22 6

slide-23
SLIDE 23

Derivation trees

TAG derivations are uniquely described by derivation trees. The derivation tree contains: nodes for all elementary trees used in the derivation, and edges for all adjunctions and substitutions performed through-

  • ut the derivation, and

edge labels indicating the target node of the rewriting opera- tion.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 23 6

slide-24
SLIDE 24

Derivation trees

TAG derivations are uniquely described by derivation trees. The derivation tree contains: nodes for all elementary trees used in the derivation, and edges for all adjunctions and substitutions performed through-

  • ut the derivation, and

edge labels indicating the target node of the rewriting opera- tion. Whenever an elementary tree γ rewrites the node at Gorn address p in the elementary tree γ ′, there is an edge from γ ′ to γ labeled with p. Note that in principle derivation trees are unordered. As a convention, daughters are ordered according to their addresses.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 24 6

slide-25
SLIDE 25

Derivation trees

For the node addresses of elementary trees, Gorn addresses are used: the root has address ϵ (or 0) the ith daughter of the node with address p has address pi.

2 22 221 21 212 211 1 13 12 121 11

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 25 7

slide-26
SLIDE 26

Derivation trees: example

adam:

NP Adam

always:

VP VP∗ Adv always

ate:

S VP NP↓ V ate NP↓

the:

NP NP∗ Det the

apple:

NP apple

Derivation tree:

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 26 8

slide-27
SLIDE 27

Derivation trees: example

VP VP∗ Adv always

S VP NP↓ V ate NP Adam

NP NP∗ Det the NP apple

Derivation tree:

ate adam

1

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 27 8

slide-28
SLIDE 28

Derivation trees: example

VP VP∗ Adv always

S VP NP apple V ate NP Adam

NP NP∗ Det the

Derivation tree:

ate adam apple

1 22

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 28 8

slide-29
SLIDE 29

Derivation trees: example

VP VP∗ Adv always

S VP NP NP apple Det the V ate NP Adam

Derivation tree:

ate adam apple the

1 22

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 29 8

slide-30
SLIDE 30

Derivation trees: example

Derived tree:

S VP VP NP NP apple Det the V ate Adv always NP Adam

Derivation tree:

ate adam apple the always

1 22 2

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 30 8

slide-31
SLIDE 31

Outline of today’s course

1

The derivation tree

2

Design principles for elementary trees

3

Sample derivations NP and PP complements Sentential complements and long-distance dependencies Modifiers

4

Feature based TAG

5

Summary and outlook

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 31 9

slide-32
SLIDE 32

Linguistic analyses with LTAG

What is an elementary tree, and what is its shape?

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 32 10

slide-33
SLIDE 33

Linguistic analyses with LTAG

What is an elementary tree, and what is its shape? elementary trees

?

⇐= syntactic/semantic properties of linguistic objects

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 33 10

slide-34
SLIDE 34

Linguistic analyses with LTAG

What is an elementary tree, and what is its shape? elementary trees

?

⇐= syntactic/semantic properties of linguistic objects ⇒ Syntactic design principles from Frank (2002):

Lexicalization Fundamental TAG Hypothesis (FTH) Condition on Elementary Tree Minimality (CETM) θ-Criterion for TAG

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 34 10

slide-35
SLIDE 35

Linguistic analyses with LTAG

What is an elementary tree, and what is its shape? elementary trees

?

⇐= syntactic/semantic properties of linguistic objects ⇒ Syntactic design principles from Frank (2002):

Lexicalization Fundamental TAG Hypothesis (FTH) Condition on Elementary Tree Minimality (CETM) θ-Criterion for TAG

⇒ Semantic design principles [Abeillé & Rambow (2000)]

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 35 10

slide-36
SLIDE 36

Linguistic analyses with LTAG

What is an elementary tree, and what is its shape? elementary trees

?

⇐= syntactic/semantic properties of linguistic objects ⇒ Syntactic design principles from Frank (2002):

Lexicalization Fundamental TAG Hypothesis (FTH) Condition on Elementary Tree Minimality (CETM) θ-Criterion for TAG

⇒ Semantic design principles [Abeillé & Rambow (2000)] ⇒ Design principle of economy

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 36 10

slide-37
SLIDE 37

Syntactic design principles (1): Lexicalization

Each elementary tree has at least one non-empty lexical item, its lexical anchor.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 37 11

slide-38
SLIDE 38

Syntactic design principles (1): Lexicalization

Each elementary tree has at least one non-empty lexical item, its lexical anchor. ⇒ All widely used grammar formalisms support some kind of lexicalization!

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 38 11

slide-39
SLIDE 39

Syntactic design principles (1): Lexicalization

Each elementary tree has at least one non-empty lexical item, its lexical anchor. ⇒ All widely used grammar formalisms support some kind of lexicalization! ⇒ TAG → LTAG: Lexicalized Tree-Adjoining Grammar

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 39 11

slide-40
SLIDE 40

Syntactic design principles (1): Lexicalization

Each elementary tree has at least one non-empty lexical item, its lexical anchor. ⇒ All widely used grammar formalisms support some kind of lexicalization! ⇒ TAG → LTAG: Lexicalized Tree-Adjoining Grammar

[Schabes & Joshi (1990); Joshi & Schabes (1991)]

Recall: reasons for lexicalization Formal properties: A finite lexicalized grammar provides finitely many analyses for each string (finitely ambiguous). Linguistic properties: Syntactic properties of lexical items can be accounted for more directly. Parsing: The search space during parsing can be delimited (grammar filtering).

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 40 11

slide-41
SLIDE 41

Syntactic design principles (2): FTH

Fundamental TAG Hypothesis (FTH); [Frank (2002)] Every syntactic dependency is expressed locally within an elemen- tary tree.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 41 12

slide-42
SLIDE 42

Syntactic design principles (2): FTH

Fundamental TAG Hypothesis (FTH); [Frank (2002)] Every syntactic dependency is expressed locally within an elemen- tary tree.

“syntactic dependency” valency/subcategorization binding filler-gap constructions ...

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 42 12

slide-43
SLIDE 43

Syntactic design principles (2): FTH

Fundamental TAG Hypothesis (FTH); [Frank (2002)] Every syntactic dependency is expressed locally within an elemen- tary tree.

“syntactic dependency” valency/subcategorization binding filler-gap constructions ... “expressed within an elementary tree” terminal leaf (i.e. lexical anchor) nonterminal leaf (substitution node and footnode) marking an inner node for obligatory adjunction

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 43 12

slide-44
SLIDE 44

Syntactic design principles (2): FTH

Examples of ill-formed elementary trees:

S VP V persuades NP↓ S VP NP ε V ate NP↓

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 44 13

slide-45
SLIDE 45

Complex primitives

Joshi (2004): Complicate locally, simplify globally. “[...] start with complex (more complicated) primitives, which capture directly some crucial linguistic properties and then introduce some general operations for composing these complex structures (primitive or derived). What is the nature of these complex primitives? In the conventional approach the primitive structures (or rules) are kept as simple as possible. This has the consequence that information (e.g., syntactic and semantic) about a lexical item (word) is distributed over more than one primitive structure. Therefore, the information associated with a lexical item is not captured locally, i.e., within the domain of a primitive structure.” [Joshi (2004)]

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 45 14

slide-46
SLIDE 46

Syntactic design principles (3): CETM

Condition on Elementary Tree Minimality (CETM); ; [Frank (2002)] The syntactic heads in an elementary tree and their projections must form the extended projection of a single lexical head.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 46 15

slide-47
SLIDE 47

Syntactic design principles (3): CETM

Condition on Elementary Tree Minimality (CETM); ; [Frank (2002)] The syntactic heads in an elementary tree and their projections must form the extended projection of a single lexical head. Examples of ill-formed elementary trees:

S VP VP V arrived AP↓ NP↓ S VP S VP NP↓ V to eat NP ε NP↓ V persuades NP↓

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 47 15

slide-48
SLIDE 48

Syntactic design principles (4): θ-Criterion for TAG

Thematic role (θ-role) the semantic relationship of an argument with its predicate is ex- pressed through the assignment of a role by the predicate to the

  • argument. Different theta-roles have different labels, such as Agent,

Theme, Patient, Goal, Source, Experiencer etc.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 48 16

slide-49
SLIDE 49

Syntactic design principles (4): θ-Criterion for TAG

Thematic role (θ-role) the semantic relationship of an argument with its predicate is ex- pressed through the assignment of a role by the predicate to the

  • argument. Different theta-roles have different labels, such as Agent,

Theme, Patient, Goal, Source, Experiencer etc. example: Bart kicked the ball.

kicked predicate Bart Agent ball Theme/Patient

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 49 16

slide-50
SLIDE 50

Syntactic design principles (4): θ-Criterion for TAG

Thematic role (θ-role) the semantic relationship of an argument with its predicate is ex- pressed through the assignment of a role by the predicate to the

  • argument. Different theta-roles have different labels, such as Agent,

Theme, Patient, Goal, Source, Experiencer etc. example: Bart kicked the ball.

kicked predicate Bart Agent ball Theme/Patient

The ball was kicked by Bart.

kicked predicate Bart Agent ball Theme/Patient

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 50 16

slide-51
SLIDE 51

Syntactic design principles (4): θ-Criterion for TAG

θ-Criterion (TAG version)

  • a. If H is the lexical head of an elementary tree T, H assigns all of

its θ-roles in T.

  • b. If A is a frontier non-terminal of elementary tree T, A must be

assigned a θ-role in T. [Frank (2002)] =⇒ Valency/subcategorization is expressed only with nonterminal leaves!

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 51 17

slide-52
SLIDE 52

Syntactic design principles (4): θ-Criterion for TAG

θ-Criterion (TAG version)

  • a. If H is the lexical head of an elementary tree T, H assigns all of

its θ-roles in T.

  • b. If A is a frontier non-terminal of elementary tree T, A must be

assigned a θ-role in T. [Frank (2002)] =⇒ Valency/subcategorization is expressed only with nonterminal leaves!

S VP V sleeps NP↓ S VP S∗ V try NP↓

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 52 17

slide-53
SLIDE 53

Further design principles

Semantic design principles Predicate-argument co-occurrence: Each elementary tree associated with a predicate contains a non- terminal leaf for each of its arguments.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 53 18

slide-54
SLIDE 54

Further design principles

Semantic design principles Predicate-argument co-occurrence: Each elementary tree associated with a predicate contains a non- terminal leaf for each of its arguments. Semantic anchoring: Elementary trees are not semantically void (to, that.)

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 54 18

slide-55
SLIDE 55

Further design principles

Semantic design principles Predicate-argument co-occurrence: Each elementary tree associated with a predicate contains a non- terminal leaf for each of its arguments. Semantic anchoring: Elementary trees are not semantically void (to, that.) Compositional principle: An elementary tree corresponds to a single semantic unit.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 55 18

slide-56
SLIDE 56

Further design principles

Semantic design principles Predicate-argument co-occurrence: Each elementary tree associated with a predicate contains a non- terminal leaf for each of its arguments. Semantic anchoring: Elementary trees are not semantically void (to, that.) Compositional principle: An elementary tree corresponds to a single semantic unit. Design principle of economy The elementary trees are shaped in such a way, that the size of the elementary trees and the size of the grammar is minimal.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 56 18

slide-57
SLIDE 57

Modification and functional elements

How to insert modifiers (e.g. easily) and functional elements (complementizers, determiners, do-auxiliaries, ...)?

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 57 19

slide-58
SLIDE 58

Modification and functional elements

How to insert modifiers (e.g. easily) and functional elements (complementizers, determiners, do-auxiliaries, ...)? either as co-anchor in the elementary tree of the lexical item they are associated with

S S VP sleeps NP↓ Comp that S VP AP easily VP sleeps NP↓

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 58 19

slide-59
SLIDE 59

Modification and functional elements

How to insert modifiers (e.g. easily) and functional elements (complementizers, determiners, do-auxiliaries, ...)? either as co-anchor in the elementary tree of the lexical item they are associated with

S S VP sleeps NP↓ Comp that S VP AP easily VP sleeps NP↓

  • r by separate auxiliary trees (e.g., XTAG grammar)

S S∗ Comp that VP AP easily VP∗

⇒ Footnodes/Adjunctions indicate both complementation and modification. ⇒ Enhancement of the CETM: [see Abeillé & Rambow (2000)]

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 59 19

slide-60
SLIDE 60

Outline of today’s course

1

The derivation tree

2

Design principles for elementary trees

3

Sample derivations NP and PP complements Sentential complements and long-distance dependencies Modifiers

4

Feature based TAG

5

Summary and outlook

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 60 20

slide-61
SLIDE 61

Sample derivations: NP and PP complements

(1) Adam gave Eve the apple.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 61 21

slide-62
SLIDE 62

Sample derivations: NP and PP complements

(1) Adam gave Eve the apple. Elementary trees:

S VP NP↓ NP↓ V gave NP↓ NP N Adam NP N Eve NP N apple NP NP∗ Det the

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 62 21

slide-63
SLIDE 63

Sample derivations: NP and PP complements

(1) Adam gave Eve the apple. Elementary trees:

S VP NP↓ NP↓ V gave NP↓ NP N Adam NP N Eve NP N apple NP NP∗ Det the

Derivation tree:

gave adam eve apple the

1 22 23 Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 63 21

slide-64
SLIDE 64

Sample derivations: NP and PP complements

(2) Adam gave the apple to Eve.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 64 22

slide-65
SLIDE 65

Sample derivations: NP and PP complements

(2) Adam gave the apple to Eve. Elementary trees:

S VP PP NP↓ P to NP↓ V gave NP↓ NP N Adam NP N Eve NP N apple NP NP∗ Det the

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 65 22

slide-66
SLIDE 66

Sample derivations: NP and PP complements

(2) Adam gave the apple to Eve. Elementary trees:

S VP PP NP↓ P to NP↓ V gave NP↓ NP N Adam NP N Eve NP N apple NP NP∗ Det the

Derivation tree:

gave adam apple eve the

1 22 232 Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 66 22

slide-67
SLIDE 67

Sample derivations: Sentential complements

(3) Adam hopes that Eve comes. Elementary trees:

S VP S∗ V hopes NP↓ S VP V comes NP↓ NP N Adam NP N Eve S S∗ Comp that

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 67 23

slide-68
SLIDE 68

Sample derivations: Sentential complements

(3) Adam hopes that Eve comes. Elementary trees:

S VP S∗ V hopes NP↓ S VP V comes NP↓ NP N Adam NP N Eve S S∗ Comp that

Derivation tree:

comes adam eve hopes that

1 1 Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 68 23

slide-69
SLIDE 69

Sample derivations: long-distance dependency

(4) Whati did Adam say (that) Eve ate _i?

NP N what S S VP S∗ V say NP↓ Aux did NP N Adam NP N Eve S S S VP NP ϵi V ate NP↓ COMP ϵ NPi↓

Derivation tree:

ate what eve did_say adam

221 1 2 21 Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 69 24

slide-70
SLIDE 70

Sample derivations: Modifiers

(5) The good student participated in every course during the semester. Elementary trees:

S VP PP NP↓ P in V part. NP↓ NP N student NP N course NP N semester NP NP∗ Det the NP NP∗ Det every N N∗ AP good VP PP NP↓ P during VP∗

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 70 25

slide-71
SLIDE 71

Sample derivations: Modifiers

Derivation tree:

part_in stud course during the good every semester the

1 222 2 1 22 Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 71 26

slide-72
SLIDE 72

Outline of today’s course

1

The derivation tree

2

Design principles for elementary trees

3

Sample derivations NP and PP complements Sentential complements and long-distance dependencies Modifiers

4

Feature based TAG

5

Summary and outlook

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 72 27

slide-73
SLIDE 73

Feature structures

Idea: Instead of atomic categorial symbols, feature structures are used as non-terminal nodes.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 73 28

slide-74
SLIDE 74

Feature structures

Idea: Instead of atomic categorial symbols, feature structures are used as non-terminal nodes. Two reasons: generalizing agreement and case marking (via underspecifica- tion) modelling adjunction constraints (TAG specific) ⇒ smaller grammars that are easier to maintain

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 74 28

slide-75
SLIDE 75

Feature structures

Idea: Instead of atomic categorial symbols, feature structures are used as non-terminal nodes. Two reasons: generalizing agreement and case marking (via underspecifica- tion) modelling adjunction constraints (TAG specific) ⇒ smaller grammars that are easier to maintain case assignment:

Joe saw her. / *Joe saw she. Joe expected her to come. (ECM) / *Joe expected she to come.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 75 28

slide-76
SLIDE 76

Feature structures

Idea: Instead of atomic categorial symbols, feature structures are used as non-terminal nodes. Two reasons: generalizing agreement and case marking (via underspecifica- tion) modelling adjunction constraints (TAG specific) ⇒ smaller grammars that are easier to maintain case assignment:

Joe saw her. / *Joe saw she. Joe expected her to come. (ECM) / *Joe expected she to come.

person/number agreement:

You sing. / *You sings. She sings. / *She sing. This woman sings. / *This woman sing. These women sing. / *These women sings.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 76 28

slide-77
SLIDE 77

Feature structures

Idea: Instead of atomic categorial symbols, feature structures are used as non-terminal nodes. Two reasons: generalizing agreement and case marking (via underspecifica- tion) modelling adjunction constraints (TAG specific) ⇒ smaller grammars that are easier to maintain case assignment:

Joe saw her. / *Joe saw she. Joe expected her to come. (ECM) / *Joe expected she to come.

person/number agreement:

You sing. / *You sings. She sings. / *She sing. This woman sings. / *This woman sing. These women sing. / *These women sings.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 77 28

slide-78
SLIDE 78

Features structures

a list of features (e.g. case) and values (e.g. nom)

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 78 29

slide-79
SLIDE 79

Features structures

a list of features (e.g. case) and values (e.g. nom) feature structures are ofen represented as atribute value matrices (AVM)

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 79 29

slide-80
SLIDE 80

Features structures

a list of features (e.g. case) and values (e.g. nom) feature structures are ofen represented as atribute value matrices (AVM) sings:

            cat V vform finite agr       num sg pers 3                  

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 80 29

slide-81
SLIDE 81

Features structures

a list of features (e.g. case) and values (e.g. nom) feature structures are ofen represented as atribute value matrices (AVM) sings:

            cat V vform finite agr       num sg pers 3                  

feature values:

atomic (e.g. for vform) feature structures (e.g. for agr)

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 81 29

slide-82
SLIDE 82

Features structures

a list of features (e.g. case) and values (e.g. nom) feature structures are ofen represented as atribute value matrices (AVM) sings:

            cat V vform finite agr       num sg pers 3                  

feature values:

atomic (e.g. for vform) feature structures (e.g. for agr)

combining constituents ⇒ unify their feature structures

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 82 29

slide-83
SLIDE 83

Unification

unification is a (partial) operation on feature structures

intuitively: the operation of combining two feature structures such that the new feature structure contains all the information

  • f the original two, and nothing more

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 83 30

slide-84
SLIDE 84

Unification

unification is a (partial) operation on feature structures

intuitively: the operation of combining two feature structures such that the new feature structure contains all the information

  • f the original two, and nothing more

e.g.

       cat vp agr

  • num

pl

      ⊔        cat vp agr

  • pers

3

      =           cat vp agr       num pl pers 3                

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 84 30

slide-85
SLIDE 85

Unification

unification is a (partial) operation on feature structures

intuitively: the operation of combining two feature structures such that the new feature structure contains all the information

  • f the original two, and nothing more

e.g.

       cat vp agr

  • num

pl

      ⊔        cat vp agr

  • pers

3

      =           cat vp agr       num pl pers 3                

partial operation ⇒ unification can fail e.g.

      cat np num sg       ⊔       cat np num pl       = FAIL

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 85 30

slide-86
SLIDE 86

Unification

unification is a (partial) operation on feature structures

intuitively: the operation of combining two feature structures such that the new feature structure contains all the information

  • f the original two, and nothing more

e.g.

       cat vp agr

  • num

pl

      ⊔        cat vp agr

  • pers

3

      =           cat vp agr       num pl pers 3                

partial operation ⇒ unification can fail e.g.

      cat np num sg       ⊔       cat np num pl       = FAIL

underspecified feature values e.g.

      cat np case nom | acc       ⊔       cat np case acc       =       cat np case acc      

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 86 30

slide-87
SLIDE 87

Unification: definition

Unification (F ⊔ G) The unification of two feature structures F and G is (if it exists) the smallest feature structure that is subsumed by both F and G.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 87 31

slide-88
SLIDE 88

Unification: definition

Unification (F ⊔ G) The unification of two feature structures F and G is (if it exists) the smallest feature structure that is subsumed by both F and G. That is, (if it exists) F ⊔ G is the feature structure with the following three properties:

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 88 31

slide-89
SLIDE 89

Unification: definition

Unification (F ⊔ G) The unification of two feature structures F and G is (if it exists) the smallest feature structure that is subsumed by both F and G. That is, (if it exists) F ⊔ G is the feature structure with the following three properties: (1) F ⊑ (F ⊔ G)

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 89 31

slide-90
SLIDE 90

Unification: definition

Unification (F ⊔ G) The unification of two feature structures F and G is (if it exists) the smallest feature structure that is subsumed by both F and G. That is, (if it exists) F ⊔ G is the feature structure with the following three properties: (1) F ⊑ (F ⊔ G) (2) G ⊑ (F ⊔ G)

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 90 31

slide-91
SLIDE 91

Unification: definition

Unification (F ⊔ G) The unification of two feature structures F and G is (if it exists) the smallest feature structure that is subsumed by both F and G. That is, (if it exists) F ⊔ G is the feature structure with the following three properties: (1) F ⊑ (F ⊔ G) (2) G ⊑ (F ⊔ G) (3) If H is a feature structure such that F ⊑ H and G ⊑ H, then (F ⊔ G) ⊑ H. If there is no smallest feature structure that is subsumed by both F and G, then we say that F ⊔ G is undefined.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 91 31

slide-92
SLIDE 92

Unification: definition

Unification (F ⊔ G) The unification of two feature structures F and G is (if it exists) the smallest feature structure that is subsumed by both F and G. That is, (if it exists) F ⊔ G is the feature structure with the following three properties: (1) F ⊑ (F ⊔ G) (2) G ⊑ (F ⊔ G) (3) If H is a feature structure such that F ⊑ H and G ⊑ H, then (F ⊔ G) ⊑ H. If there is no smallest feature structure that is subsumed by both F and G, then we say that F ⊔ G is undefined. Subsumption (F1 ⊑ F2) A feature structure F1 subsumes (⊑) another feature structure F2, iff all the information that is contained in F1 is also contained in F2.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 92 31

slide-93
SLIDE 93

Reentrancies

the paths that both lead to the same node ⇒ to the same value

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 93 32

slide-94
SLIDE 94

Reentrancies

the paths that both lead to the same node ⇒ to the same value ⇒ hence, they share that value

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 94 32

slide-95
SLIDE 95

Reentrancies

the paths that both lead to the same node ⇒ to the same value ⇒ hence, they share that value this property of sharing value(s) is called reentrancy

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 95 32

slide-96
SLIDE 96

Reentrancies

the paths that both lead to the same node ⇒ to the same value ⇒ hence, they share that value this property of sharing value(s) is called reentrancy in AVMs: expressed by coindexing the shared values (boxed numbers)

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 96 32

slide-97
SLIDE 97

Reentrancies

the paths that both lead to the same node ⇒ to the same value ⇒ hence, they share that value this property of sharing value(s) is called reentrancy in AVMs: expressed by coindexing the shared values (boxed numbers) within feature structures:       attr1

1

attr2

1

            attr1

1 val1

attr2

1

     

  • attr1

1

  • attr2

1

  • Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf)

97 32

slide-98
SLIDE 98

Reentrancies

FTAG uses acyclic reentrancies!

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 98 33

slide-99
SLIDE 99

Reentrancies

FTAG uses acyclic reentrancies! between feature structures (in a tree):

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 99 33

slide-100
SLIDE 100

Reentrancies

FTAG uses acyclic reentrancies! between feature structures (in a tree):

  • attr1

1

  • attr1

1

  • Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf)

100 33

slide-101
SLIDE 101

Reentrancies

FTAG uses acyclic reentrancies! between feature structures (in a tree):

  • attr1

1

  • attr1

1

  • Note that

feature structues in FTAG are untyped. (In contrast to frames, see tomorrow’s course.) the feature geometry is such that there is only a finite number

  • f possible feature structures

therefore, FTAG can be shown to be strongly equivalent to TAG without feature structures

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 101 33

slide-102
SLIDE 102

Unification: examples

        agr

  • num

sg

  • subj
  • agr
  • num

sg

       ⊔

  • subj
  • agr
  • pers

3

  • =

           agr

  • num

sg

  • subj

       agr       num sg pers 3                        

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 102 34

slide-103
SLIDE 103

Unification: examples

        agr

  • num

sg

  • subj
  • agr
  • num

sg

       ⊔

  • subj
  • agr
  • pers

3

  • =

           agr

  • num

sg

  • subj

       agr       num sg pers 3                        

       agr

1

  • num

sg

  • subj
  • agr

1

      ⊔

  • subj
  • agr
  • pers

3

  • =

          agr

1

      num sg pers 3       subj

  • agr

1

        

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 103 34

slide-104
SLIDE 104

Unification: examples

        agr

  • num

sg

  • subj
  • agr
  • num

sg

       ⊔

  • subj
  • agr
  • pers

3

  • =

           agr

  • num

sg

  • subj

       agr       num sg pers 3                        

       agr

1

  • num

sg

  • subj
  • agr

1

      ⊔

  • subj
  • agr
  • pers

3

  • =

          agr

1

      num sg pers 3       subj

  • agr

1

        

for any feature structure F: F ⊔ [ ] = [ ] ⊔ F = F

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 104 34

slide-105
SLIDE 105

Unification: examples

        agr

  • num

sg

  • subj
  • agr
  • num

sg

       ⊔

  • subj
  • agr
  • pers

3

  • =

           agr

  • num

sg

  • subj

       agr       num sg pers 3                        

       agr

1

  • num

sg

  • subj
  • agr

1

      ⊔

  • subj
  • agr
  • pers

3

  • =

          agr

1

      num sg pers 3       subj

  • agr

1

        

for any feature structure F: F ⊔ [ ] = [ ] ⊔ F = F the empty feature structure is the identity element for unifi- cation

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 105 34

slide-106
SLIDE 106

TAG with feature structures

idea: feature structures as non-terminal nodes

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 106 35

slide-107
SLIDE 107

TAG with feature structures

idea: feature structures as non-terminal nodes at substitution/adjunction the feature structures of the partici- pating nodes are unified

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 107 35

slide-108
SLIDE 108

TAG with feature structures

idea: feature structures as non-terminal nodes at substitution/adjunction the feature structures of the partici- pating nodes are unified

            cat np agr       num sg pers 3       case nom            

she

  • cat

s

         cat vp agr

1

      num sg pers 3                

sings

         cat np agr

1

case nom         

↓

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 108 35

slide-109
SLIDE 109

TAG with feature structures

idea: feature structures as non-terminal nodes at substitution/adjunction the feature structures of the partici- pating nodes are unified

  • cat

s

         cat vp agr

1

      num sg pers 3                

sings

            cat np agr

1

      num sg pers 3       case nom            

she

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 109 35

slide-110
SLIDE 110

FTAG

Feature-structure based TAG (FTAG Vijay-Shanker & Joshi 1988):

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 110 36

slide-111
SLIDE 111

FTAG

Feature-structure based TAG (FTAG Vijay-Shanker & Joshi 1988): annotate each substitution node with one and each other node with two feature structures

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 111 36

slide-112
SLIDE 112

FTAG

Feature-structure based TAG (FTAG Vijay-Shanker & Joshi 1988): annotate each substitution node with one and each other node with two feature structures adjunction splits the feature structures

top features: the relation of the node to the tree above it botom features: the relation of the node to the tree below it

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 112 36

slide-113
SLIDE 113

FTAG

Feature-structure based TAG (FTAG Vijay-Shanker & Joshi 1988): annotate each substitution node with one and each other node with two feature structures adjunction splits the feature structures

top features: the relation of the node to the tree above it botom features: the relation of the node to the tree below it

FTAG description of node η

  • 1. The relation of η to its supertree is called feature structure tη.
  • 2. The relation of η to its descendants is called feature structure bη.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 113 36

slide-114
SLIDE 114

FTAG

Feature-structure based TAG (FTAG Vijay-Shanker & Joshi 1988): annotate each substitution node with one and each other node with two feature structures adjunction splits the feature structures

top features: the relation of the node to the tree above it botom features: the relation of the node to the tree below it

FTAG description of node η

  • 1. The relation of η to its supertree is called feature structure tη.
  • 2. The relation of η to its descendants is called feature structure bη.

at the final derived tree top and botom features are unified for all nodes

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 114 36

slide-115
SLIDE 115

FTAG: Substitution

Substitution in FTAG The top features of the root of the tree to substitute unify with the top features of the substitution node. Y[t1]

[b]

X . . . Y↓[t2] ⇒ X . . . Y[t1]⊔[t2]

[b]

substitution nodes (Y↓) have only top features

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 115 37

slide-116
SLIDE 116

FTAG: Adjunction

Adjunction in FTAG The top features of the root of the auxiliary tree unify with the top features of the adjunction node, and the botom features of the footnode of the auxiliary tree unify with the botom features of the adjunction node. X Y

[ty] [by]

. . . Y[tr]

[br]

Z Y∗[tf ]

[bf ]

⇒ X Y

[ty]⊔[tr] [br]

Z Y

[tf ] [by]⊔[bf ]

. . .

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 116 38

slide-117
SLIDE 117

Adjunction constraints

Modeling adjunction constraints with features:

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 117 39

slide-118
SLIDE 118

Adjunction constraints

Modeling adjunction constraints with features: SA: top and botom are unifiable

  • cat

vp

  • cat

vp

  • Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf)

118 39

slide-119
SLIDE 119

Adjunction constraints

Modeling adjunction constraints with features: SA: top and botom are unifiable

  • cat

vp

  • cat

vp

  • OA + SA: feature mismatch between top and botom

      cat vp mode ind             cat vp mode ger      

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 119 39

slide-120
SLIDE 120

Adjunction constraints

Modeling adjunction constraints with features: SA: top and botom are unifiable

  • cat

vp

  • cat

vp

  • OA + SA: feature mismatch between top and botom

      cat vp mode ind             cat vp mode ger      

NA: top and botom are unifiable, but there is no auxiliary tree in the grammar that can be unified with them

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 120 39

slide-121
SLIDE 121

FTAG example for OA

(6) John is singing.

NP[]

[agr=[pers=3, num=sg]]

‘John’ S NP[agr= 1 ] VP[agr= 1 , mode=ind]

[mode=ger]

V ‘singing’ VP[agr= 2 , mode= 3 ] V[mode= 3 ind]

[agr= 2 [pers=3, num=sg]]

VP∗[mode=ger] ‘is’ The features are inspired by the XTAG grammar (XTAG Research Group 2001). The cat feature is taken to be special, in particular it is usually the same in top and botom. We therefore notate it as the main category of a node,

  • utside the feature structures.

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 121 40

slide-122
SLIDE 122

FTAG example for OA

(6) John is singing.

NP[]

[agr=[pers=3, num=sg]]

‘John’ S NP[agr= 1 ] VP[agr= 1 , mode=ind]

[mode=ger]

V ‘singing’ VP[agr= 2 , mode= 3 ] V[mode= 3 ind]

[agr= 2 [pers=3, num=sg]]

VP∗[mode=ger] ‘is’

Result of derivation:

S NP[agr= 1 ]

[agr=[pers=3, num=sg]]

VP[agr= 1 ,mode=ind]

[agr= 2 , mode= 3 ]

‘John’ V[mode= 3 ind]

[agr= 2 [pers=3, num=sg]]

VP[mode=ger]

[mode=ger]

‘is’ V ‘singing’

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 122 40

slide-123
SLIDE 123

FTAG example for OA

(6) John is singing.

NP[]

[agr=[pers=3, num=sg]]

‘John’ S NP[agr= 1 ] VP[agr= 1 , mode=ind]

[mode=ger]

V ‘singing’ VP[agr= 2 , mode= 3 ] V[mode= 3 ind]

[agr= 2 [pers=3, num=sg]]

VP∗[mode=ger] ‘is’

Afer top-botom unifications:

S NP[agr = 1 ] VP[agr = 1 [pers = 3, num =sg]

mode =ind ]

‘John’ V[agr = 1

mode=ind]

VP[mode =ger] ‘is’ V ‘singing’

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 123 40

slide-124
SLIDE 124

Case assignment

nouns carry the case, which is ‘checked’

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 124 41

slide-125
SLIDE 125

Case assignment

nouns carry the case, which is ‘checked’ noun case is checked against the case value assigned by the verb during the unification

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 125 41

slide-126
SLIDE 126

Case assignment

nouns carry the case, which is ‘checked’ noun case is checked against the case value assigned by the verb during the unification features of case-assignment:

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 126 41

slide-127
SLIDE 127

Case assignment

nouns carry the case, which is ‘checked’ noun case is checked against the case value assigned by the verb during the unification features of case-assignment:

case with values: nom | acc | gen | none ⇒ Ns, NPs

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 127 41

slide-128
SLIDE 128

Case assignment

nouns carry the case, which is ‘checked’ noun case is checked against the case value assigned by the verb during the unification features of case-assignment:

case with values: nom | acc | gen | none ⇒ Ns, NPs assign-case with values: nom | acc | none ⇒ case assigners (prepositions, verbs) and S, VP, PP nodes that dominate them

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 128 41

slide-129
SLIDE 129

Case assignment

(7) a. she laughed

  • b. *her laughed

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 129 42

slide-130
SLIDE 130

Case assignment

(7) a. she laughed

  • b. *her laughed

NP[ ]

      

agr = [num = sg,pers = 3] case = nom

      

she NP[ ]

      

agr = [num = sg,pers = 3] case = acc

      

her

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 130 42

slide-131
SLIDE 131

Case assignment

(7) a. she laughed

  • b. *her laughed

NP[ ]

      

agr = [num = sg,pers = 3] case = nom

      

she NP[ ]

      

agr = [num = sg,pers = 3] case = acc

      

her S VP

  • assign-case = 1 nom
  • tense = past
  • laughed

NP

  • case = 1
  • ↓

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 131 42

slide-132
SLIDE 132

Outline of today’s course

1

The derivation tree

2

Design principles for elementary trees

3

Sample derivations NP and PP complements Sentential complements and long-distance dependencies Modifiers

4

Feature based TAG

5

Summary and outlook

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 132 43

slide-133
SLIDE 133

Summary & outlook

Summary derivation tree vs. derived tree design principles: syntactic / semantic / economy example derivations: the base cases TAG + features structures

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 133 44

slide-134
SLIDE 134

Summary & outlook

Summary derivation tree vs. derived tree design principles: syntactic / semantic / economy example derivations: the base cases TAG + features structures Tomorrow further linguistic applications (syntax) in particular: extraction and long-distance dependencies grammar factorization the syntax-semantics interface

Kallmeyer, Lichte, Osswald & Petitjean (HHU Düsseldorf) 134 44

slide-135
SLIDE 135

References

Abeillé, Anne & Owen Rambow. 2000. Tree adjoining grammar: An overview. In Anne Abeillé & Owen Rambow (eds.), Tree Adjoining Grammars: Formalisms, linguistic analyses and processing (CSLI Lecture Notes 107), 1–68. Stanford, CA: CSLI Publications. Frank, Robert. 2002. Phrase structure composition and syntactic dependencies. Cambridge,MA: MIT Press. Joshi, Aravind K. 2004. Starting with complex primitives pays off: complicate locally, simplify

  • globally. Cognitive Science 28. 637–668.

Joshi, Aravind K. & Yves Schabes. 1991. Tree-Adjoining Grammars and lexicalized grammars.

  • Tech. Rep. MS-CIS-91-22 Department of Computer and Information Science, University of
  • Pennsylvania. http://repository.upenn.edu/cis_reports/445/.

Schabes, Yves & Aravind K. Joshi. 1990. Parsing with lexicalized tree adjoining grammar. Tech.

  • Rep. MS-CIS-90-11 Department of Computer and Information Science, University of
  • Pennsylvania. http://repository.upenn.edu/cis_reports/542/.

Vijay-Shanker, K. & Aravind K. Joshi. 1988. Feature structures based tree adjoining grammar. In Proceedings of coling, 714–719. Budapest. XTAG Research Group. 2001. A Lexicalized Tree Adjoining Grammar for English. Tech. rep. Institute for Research in Cognitive Science, University of Pennsylvania Philadelphia, PA.