Please enable JavaScript.
Coggle requires JavaScript to display documents.
Semantic Features and Selection Restrictions by Paducheva - Coggle Diagram
Semantic Features and Selection Restrictions by Paducheva
Lexical Database of the System
LEXICOGRAPllER: expert system
Designed for natural language conservation purposes
Consists of the following components:
lexical database (LBD), bibliographic database (BBD).
Work on the project is being
carried out by a group of researchers. M.Filipenko
e.a.; cf. Paduchcva, Rakhilina 1989.
The user can obtain information about morphology, syntactic characteristics, semantic features, prosody, etc.
There are characteristics such as: [ + speech act verb], [ + performative verb], [ + movement verb], [ + kinship term]
You can also compile lists of words that have a common
characteristic
compile alphabetical lists of words and characteristics mentioned in the BBD
Unlike all existing bibliographic catalogues, our BBD contains bibliographic information on individual lexemes
Semantic Feature According to U.Weinreich
U. Weinreich (1967) proposed a useful distinction between a paradigmatic semantic feature and a transfer function.
This distinction made it possible to use the notion of semantic feature.
There is a broader meaning than in transformational grammar
(TG)
In Weinreieh's conception: Purposes
Is considered as a basis for semantic agreement
Explains deviant and metaphorical readings
Adds provisional semantic contents to a
potentially ambiguous word
As a result, the word house is interpreted as an event,
for example, as an event of someone's perception of a house
Weinreich divides semantics into: categorical and transitive.
Verbs= movement (categorical)
Transitive = requires two or more participants, example: the tree bore fruit.
Semantic Features In Systems Of Natural Language Processing (NLP)
Semantic features are crucial for understanding meaning in NLP
Verb Complementation
Verbs require specific types of adverbials based on semantics (e.g., purpose adverbials only with controlled actions).
Coordinate Constructions:
Semantic agreement guides how to expand a conjunction.
Word Disambiguation
Homonyms
: Categorical features help choose the intended meaning based on context (e.g., "hotel" vs. "boy").
Nouns
: Transfer features of the predicate help identify the correct meaning.
Anaphora Resolution
Features help identify the referent of pronouns (e.g., "spilled" requiring a liquid subject).
Syntactic Parsing
Ensure correct structure by verifying semantic agreement between verb and arguments (categories).
Literal vs. Figurative Language
Features can distinguish literal interpretations from metaphors.
Historical Context
Early NLP (1960s):
Dominant tool for semantic analysis.
Modern NLP:
Less prominent due to advancements in semantic decomposition.
Semantic Features and Selection Restrictions in Lexicon and Grammar
Semantic features play a key role in how words can be combined grammatically (selection restrictions)
Historical Context:
Early NLP (1960s):
Semantic features were the dominant tool for semantic analysis.
Modern NLP:
Semantic decomposition gained prominence, diminishing the role of features.
Modern Relevance
Semantic features represent components of lexical meaning.
Examples
Semantic features explain traditionallysyntactic restrictions:
Neg-Raising predicates
Conjunction use in Russian
On Semantic In-Variant of the class of words with Genitive Subject
Semantic invariant
defines when a
genitive subject is
used and connects
syntactic behavior
of verbs to their
semantic components.
Main Concepts
Genitive Subject Construction
Example Sentences
Ответа не принятo
(No answer accepted)
Alternative with Nominative
Ответ не принят
(Answer not accepted)
Мороз не чувствуется
(Cold not felt)
Ketexty Thesis
List of verbs
(200+ items)
predicts genitive subject
Semantic affinity
is insufficient for
reliable prediction.
Apresjan 1985 Thesis
determines case
of subject.
Stylistic Variants
Aggressive Nominative
Example
Готовность не достигнута
vs
Готовности не достигнуто
Semantic Invariant
Characterization
Exclusion Conditions
Factors
Referential vs. Non-referential
Topic-focus
Animate vs. Inanimate
Observer presence
Nominative Alternative
Examples
Ответ принят
=
Ответ не принят
(presupposed)
Ответа не принято
(not presupposed)
Semantic Explanation
Genitive Verbs
Group I
Component
X exists
Negation
X does not exist
Group II
Component
X in observer's vision
Negation
X not in vision
