Chapter 4 : Semantic Analysis

(Part II)
Adama Science and Technology University
School of Electrical Engineering and Computing
Department of CSE
Dr. Mesfin Abebe Haile (2022)
 Outline

 Semantic Analysis
 Lexical semantics and word-sense disambiguation
 Compositional semantics
 Semantic Role Labeling and Semantic Parsing.

01/02/23 2
 Semantic Analysis

01/02/23 3
 Introduction

 Semantic analysis refers to analyzing the meanings of words,

fixed expressions, whole sentences, and utterances in context.
 In practice, this means translating original expressions into some
kind of semantic metalanguage.

 The major theoretical issues in semantic analysis therefore turn

on the nature of the metalanguage or equivalent
representational system.

01/02/23 4
 Introduction…

 For extended texts, specific NLP applications of semantic

analysis may include:
 Information retrieval,
 Information extraction,
 Text summarization,
 Data-mining, and
 Machine translation and translation aids.

01/02/23 5
 Introduction…

 Semantic analysis is also pertinent for much shorter texts, right

down to the single word level,
 For example, in understanding user queries and matching user
requirements to available data.

 Semantic analysis is also of high relevance in efforts to improve

Web ontologies and knowledge representation systems.

01/02/23 6
 Introduction…

 Various theories and approaches to semantic representation

can be roughly ranged along two dimensions:
 (1) formal vs. cognitive and
 (2) compositional vs. lexical

 Formal theories have been strongly advocated since the late

1960s while cognitive approaches have become popular in the
last three decades, driven also by influences from cognitive
science and psychology.

01/02/23 7
 Introduction…

 Compositional semantics is concerned with the bottom-up

construction of meaning, starting with the lexical items, whose
meanings are generally treated as given.

 Lexical semantics, on the other hand, aims at precisely

analyzing the meanings of lexical items, either by analyzing
their internal structure and content (decompositional
approaches) or by representing their relations to other elements
in the lexicon (relational approaches).

01/02/23 8
 Lexical Semantics and Word-Sense
Disambiguation
 Three Perspectives on Meaning
1. Lexical Semantics
 The meanings of individual words.
2. Formal Semantics (or Compositional Semantics or Sentential
Semantics)
 How those meanings combine to make meanings for individual
sentences or utterances.
3. Discourse or Pragmatics
 How those meanings combine with each other and with other facts
about various kinds of context to make meanings for a text or
discourse.
 Dialog or Conversation is often lumped together with Discourse.
01/02/23 9
 Lexical Semantics and Word-Sense
Disambiguation…
 Lexical Semantics
 Can be defined as the study of what individual lexical items mean,
why they mean, what they do, how we can represent all of this, and
where the combined interpretation for an utterance comes from.
 Lexical semantics is concerned with the identification and
representation of the semantics of lexical items.
 If we are to identify the semantics of lexical items, we have to be
prepared for the eventuality of a given word having multiple
interpretations = polysemy (cf. monosemy).
 Polysemy = the condition of a single lexical item having
multiple meanings.

01/02/23 10
 Lexical Semantics and Word-Sense
Disambiguation…
 Lexical Semantics…
 There is a traditional division made between lexical semantics and
supralexical semantics.
 Lexical semantics, which concerns itself with the meanings of words
and fixed word combinations,
 Supralexical (combinational, or compositional) semantics, which
concerns itself with the meanings of the indefinitely large number of
word combinations—phrases and sentences—allowable under the
grammar.
 While there is some obvious appeal and validity to this division, it
is increasingly recognized that word-level semantics and
grammatical semantics interact and interpenetrate in various ways.
01/02/23 11
 Lexical Semantics and Word-Sense
Disambiguation…
 Word-Sense Disambiguation
 Many tasks in natural language processing require
disambiguation of ambiguous words.
 Question Answering
 Information Retrieval
 Machine Translation
 Text Mining
 Phone Help Systems

 Understanding how people disambiguate words is an interesting

problem that can provide insight in psycholinguistics.
01/02/23 12
 Lexical Semantics and Word-Sense
Disambiguation…
 Word-Sense Disambiguation
 Task of determining the meaning of an ambiguous word in the
given context.
 Bank:
 Edge of a river
or
 Financial institution that accepts money

 Refers to the resolution of lexical semantic ambiguity and its goal

is to attribute the correct senses to words.

01/02/23 13
 Lexical Semantics and Word-Sense
Disambiguation…
 Word-Sense Disambiguation…
 Given
 A word in context,
 A fixed inventory of potential word senses.
 Decide which sense of the word this is:
 English-to-Spanish MT
 Inventory is the set of Spanish translations
 Speech Synthesis
 Inventory is homographs with different pronunciations like bass
and bow.
 Automatic indexing of medical articles
 MeSH (Medical Subject Headings) thesaurus entries.
01/02/23 14
 Lexical Semantics and Word-Sense
Disambiguation…
 Word-Sense Disambiguation…
 Approaches
 Supervised
 Semi-supervised
 Unsupervised
» Dictionary-based techniques
» Selectional association
 Lightly supervised
» Bootstrapping
» Preferred Selectional Association

01/02/23 15
 Lexical Semantics and Word-Sense
Disambiguation…
 Word-Sense Disambiguation…
 WSD Approaches
 Disambiguation based on manually created rules,

 Disambiguation using machine readable dictionaries,

 Disambiguation using thesauri,

 Disambiguation based on unsupervised machine learning with

corpora.

01/02/23 16
 Lexical Semantics and Word-Sense
Disambiguation…
 Word-Sense Disambiguation…
 Lexical Ambiguity
 Most words in natural languages have multiple possible meanings.
– “pen” (noun)
» The dog is in the pen.
» The ink is in the pen.
– “take” (verb)
» Take one pill every morning.
» Take the first right past the stoplight.
 Syntax helps distinguish meanings for different parts of speech of an
ambiguous word.
– “conduct” (noun or verb)
» John’s conduct in class is unacceptable.
» John will conduct the orchestra on Thursday.
01/02/23 17
 Lexical Semantics and Word-Sense
Disambiguation…
 Word-Sense Disambiguation…
 Evaluation of WSD
 “In vitro”:
 Corpus developed in which one or more ambiguous words are
labeled with explicit sense tags according to some sense inventory.
 Corpus used for training and testing WSD and evaluated using
accuracy (percentage of labeled words correctly disambiguated).
» Use most common sense selection as a baseline.

 “In vivo”:
 Incorporate WSD system into some larger application system, such
as machine translation, information retrieval, or question answering.
 Evaluate relative contribution of different WSD methods by
measuring performance impact on the overall system on final task
(accuracy of MT, IR, or QA results).
01/02/23 18
 Lexical Semantics and Word-Sense
Disambiguation…
 Word-Sense Disambiguation…
 Issues in WSD
 What is the right granularity of a sense inventory?
 Integrating WSD with other NLP tasks:
 Syntactic parsing
 Semantic role labeling
 Semantic parsing

 Does WSD actually improve performance on some real end-user

task?
 Information retrieval
 Information extraction
 Machine translation
 Question answering
01/02/23 19
 Compositional Semantics

 Compositional semantics is concerned with the bottom-up

construction of meaning, starting with the lexical items, whose
meanings are generally treated as given.

 Compositional semantics: the construction of meaning

(generally expressed as logic) based on syntax.

01/02/23 20
 Semantic Role Labeling and
Semantic Parsing…
 Semantic Parsing:
 Traditional sentence parsing is often performed as a method of
understanding the exact meaning of a sentence or word,
sometimes with the aid of devices such as sentence diagrams.
 It usually emphasizes the importance of grammatical divisions
such as subject and predicate.
 Within computational linguistics parsing is used to refer to the
formal analysis by a computer of a sentence or other string of
words into its constituents, resulting in a parse tree showing their
syntactic relation to each other.
 Semantic parsing is the extension of broad-coverage probabilistic
parsers to represent sentence meaning.
01/02/23 21
 Question & Answer

01/02/23 22
 Thank You !!!

01/02/23 23
 Individual Assignment - Three

 Write a Term Paper on one of the following topics : (It will be a

substitute for the mid exam – 20%)
 There might be a presentation (the main body of the term paper
should not be less than 20 pages)
 Sentiment Analysis,
 Text Classification,
 Chatbots and Virtual Assistants,
 Text Summarization,
 Email Filtering,
 Advertisement to Targeted Audience

01/02/23 24
 Individual Assignment - Two

 Write a Term Paper on one of the following topics : (It will be a

substitute for the mid exam – 20%)
 There might be a presentation (the main body of the term paper
should not be less than 20 pages)
 Social Media Analytics,
 Plagiarism Tools,
 Retrieval System,
 Question and Answering,
 Speech Recognition,
 Fake News Detection

01/02/23 25