Download presentation
Presentation is loading. Please wait.
2
Christel Kemke 2007/08 COMP 4060 Natural Language Processing Feature Structures and Unification
3
2007/08 Christel Kemke Feature Structures - General Feature structures describe linguistic attributes or features like number, person associated with words or syntactic constituents like noun phrase. Feature structures are sets of features and values, e.g. hat[Numbersing ] buys[Person 3 ] [Numbersing ]
4
2007/08 Christel Kemke Feature Structures - Agreement Feature structures can be collected in one ‘variable’ called agreement. buys agreement [Person 3] [Number sing]
5
2007/08 Christel Kemke Feature Structures, Grammar, Parsing Feature Structures describe additional syntactic-semantic information, like category, person, number, e.g. goes specify feature structure constraints (agreements) as part of the grammar rules during parsing, check agreements of feature structures (unification) example S → NP VP =
6
2007/08 Christel Kemke Feature Structures as Constraints Ungrammatical sentences like “He go” or “We goes” can be excluded using feature constraints. example S → NP VP = =
7
2007/08 Christel Kemke Add to feature structure category cat: buys cat verb agreement [Person 3 ] [Number sing] Feature Structures and Categories
8
2007/08 Christel Kemke Compare and combine feature structures: he buys buys cat verb agreement [Person 3] [Number sing] he cat noun agreement [Person 3] [Number sing] Feature Structures and Unification
9
2007/08 Christel Kemke Unification of Feature Structures Agreement is checked by the unification operation according to the following rules: [feature i value i ] |_| [feature i value i ] = [feature i value i ] [feature i value i ] |_| [feature i value j ] = failif value i value j [feature i value i ] |_| [feature i undef.] = [feature i value i ] [feature i value i ] |_| [feature j value j ] = feature i value i feature j value j if feature i feature j
10
2007/08 Christel Kemke "Inheritance" of Feature Structures Feature structures are "inherited" during parsing or generation, using the feature structure of the head of a phrase: NP det Nom NP i=1,..,n det Nom i=1,..,n Nom i=1,..,n pre-Nom Nom i=1,..,n post-Nom Complex feature structures are often referenced through identifying numbers. Constraints on feature structures can be checked using these references; and the same feature structure can be used in different parts of the parse tree through reference. head
11
2007/08 Christel Kemke Features and Subcategorization 1 NP modifiers or Verb complements central noun + modifiers + agreement central verb + complements + agreements “... the man who chased the cat out of the house...” “... the man chased the barking dog who bit him...” Agreements are passed on / inherited within phrases, e.g. agreement of VP derived from Head-Verb of VP: determined by
12
2007/08 Christel Kemke Features and Subcategorization 2 NP modifiers: central noun + modifiers + agreement “... the man who chased the cat out of the house...” NP - determined by man - Verb complements: central verb + complements + agreements “... the man chased the barking dog who bit him...” VP- determined by chased-
13
Christel Kemke 2007/08 Semantics
14
2007/08 Christel Kemke Semantics – What do we need? Distinguish between surface structure (syntactic structure) and deep structure (semantic structure) of sentences. Different forms of semantic representation: logic formalisms ontology / semantic representation languages Case Frame Structures (Filmore) Conceptual Dependy Theory (Schank) Description Logic (DL) and similar KR languages Ontologies
15
2007/08 Christel Kemke Constructing a Semantic Representation General approach: Start with surface structure derived from parser. Map surface structure to semantic structure: Use phrases as sub-structures. Find concepts and representations for central phrases (e.g. VP, NP, then PP) Assign phrases to appropriate roles around central concepts (e.g. bind PP into VP representation).
16
2007/08 Christel Kemke Semantic Representation Semantic Representations are based on some form of (formal) Representation Language. Semantics Networks Conceptual Dependency Graphs Case Frames Ontologies DL and similar KR languages
17
2007/08 Christel Kemke Ontology (Interlingua) approach Ontology: a language-independent classification of objects, events, relations A Semantic Lexicon, which connects lexical items to nodes (concepts) in the ontology An analyzer that constructs Interlingua representations and selects (an?) appropriate one (based on Steven Helmreich's 419 Class, Nov 2003)
18
2007/08 Christel Kemke Semantic Lexicon Provides a syntactic context for the appearance of the lexical item Provides a mapping for the lexical item to a node in the ontology (or more complex associations) Provides connections from the syntactic context to semantic roles and constraints on these roles
19
2007/08 Christel Kemke Input: John makes tools Syntactic Analysis: Basic Semantic Dependency - Example catverb root make tensepresent subject root john catnoun-proper object root tool catnoun numberplural
20
2007/08 Christel Kemke John-n1 syn-struc rootjohn catnoun-proper sem-struc human name john gender male tool-n1 syn-struc roottool catn sem-struc tool Lexicon Entries for John and tool
21
2007/08 Christel Kemke Relevant extract from the specification of the ontological concept used to describe the appropriate meaning of make: manufacturing-activity... agenthuman themeartifact … Ontological Representation - Example
22
2007/08 Christel Kemke John-n1 syn-struc rootjohn catnoun-proper sem-struc human name john gendermale tool-n1 syn-struc roottool catn sem-struc tool Relevant parts of the (appropriate senses of the) lexicon entries for John and tool
23
2007/08 Christel Kemke The basic semantic dependency component of the TMR for John makes tools manufacturing-activity-7 agentuman-3 themeset-1 element tool cardinality> 1 … Semantic Dependency Component
24
try-v3 syn-struc root try cat v subj root $var1 cat n xcomp root $var2 cat v formOR infinitive gerund sem-struc set-1element-typerefsem-1 cardinality>=1 refsem-1 semevent agent^$var1 effectrefsem-2 modality modality-typeepiteuctic modality-scoperefsem-2 modality-value< 1 refsem-2value^$var2 semevent
25
2007/08 Christel Kemke Constructing an IL representation For each syntactic analysis: Access all semantic mappings and contexts for each lexical item. Create all possible semantic representations. Test them for coherency of structure and content.
26
“Why is Iraq developing weapons of mass destruction?”
27
2007/08 Christel Kemke Wordsense Disambiguation Methods Constraint checking make sure the constraints imposed on context are met Graph traversal is-a links are inexpensive other links are more expensive the “cheapest” structure is the most coherent one Hunter-gatherer processing find (hunt) and eliminate (kill) unlikely interpretations collect (gather) remaining interpretations
28
Christel Kemke 2007/08 Logic Formalisms Lambda Calculus
29
2007/08 Christel Kemke Semantics - Lambda Calculus 1 Logic representations often involve Lambda( )-Calculus: -expressions represent central phrases (e.g. VP) They are like functions which can be applied to terms We replace variables in -expression with semantic representations of complements or modifier phrases x, y: loves (x, y)FOPLsentence x y loves (x, y) -expression x y loves (x, y) (John) y loves (John, y) function
30
2007/08 Christel Kemke Semantics - Lambda Calculus 2 Transform sentence into lambda-expression: “AI Caramba is close to ICSI.” specific: close-to (AI Caramba, ICSI) general: x,y: close-to (x, y) x=AI Caramba y=ICSI Lambda Conversion: x y: close-to (x, y) (AI Caramba) Lambda Reduction: y: close-to (AI Caramba, y) close-to (AI Caramba, ICSI)
31
2007/08 Christel Kemke Semantics - Lambda Calculus 3 Lambda-expressions can be constructed from central expression (VP), inserting semantic representations for complement phrases verb serves { x y e IS-A(e, Serving) Server(e,y) Served(e,x)} represents general semantics for the verb serve. Fill in appropriate expressions for x, y, derived from the noun in a NP which is complement to verb serve. For example,“AI Caramba serves meat.” - use meat for y.
32
2007/08 Christel Kemke References Jurafsky, D. & J. H. Martin, Speech and Language Processing, Prentice-Hall, 2000. (Chapters 9 and 10) Helmreich, S., From Syntax to Semantics, Presentation in the 74.419 Course, November 2003.
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.