Samenvatting
Uitgebreide samenvatting - Computationele Grammatica (LIX025B05)
Een uitgebreidere in-depth samenvatting voor het vak Computational Grammar of ook Computationele Grammatica.
[Meer zien]Voorbeeld 2 van de 7 pagina's
Voorbeeld 2 van de 7 pagina's
In winkelwagengesponsord bericht van onze partner
Verkoper
Volgen
Voorbeeld van de inhoud
Computational GrammarIntroduction to Natural Language Processing and
Context Free Grammars
1. Ambiguity
In NLP, ambiguity is the case of multiple (more than one) meaning or interpretation to a word,
phrase, or sentence. This causes errors in language understanding.
1.1 Lexical Ambiguity
One of the main types of ambiguity in NLP is called lexical ambiguity, which refers to semantic
ambiguity. In semantic ambiguity, a word can have more than one possible meaning, in the case of a
word having multiple definitions or in the case of having different meanings in different contexts.
An example of lexical ambiguity is “I saw her duck”, in which the word “duck” can be interpreted as
possibly a verb or a noun. Another example is “The bank is closed” in which the word “bank” as a
noun can have multiple definitions such as a 1. money bank or 2. the edge of a river.
1.2 Syntactic Ambiguity
Another main type of ambiguity is syntactic, this is when the structure of a sentence (i.e.: syntax) has
multiple (more than one) possible interpretation.
An example to this is again the sentence “I saw her duck” and differs from lexical ambiguity in the
case of “her duck” vs. “duck” being either a duck as noun being in possession of the person referred
to as “her” or this being a verb because of an action of the person. Another example is “I saw the
man with the telescope”, which can be interpreted as either a man with an accompanying telescope
or the person “I” using the telescope to watch the second person “man”.
1.3 Discourse Ambiguity
The last type of ambiguity we will discuss occurs when the meaning of a sentence is unclear context-
wise leading to multiple ways for interpretation.
An example of this is “John took his car to the garage” and “John met his friend at the garage”, in
which both “his car” and “his friend” can be connected to “John” or can be derived from somewhere
else in the discourse.
2. Levels of Analysis in NLP
In NLP there are five levels of analysis to understand the structure and meaning of language. These
levels are phonology, morphology, syntax, pragmatics, and discourse.
2.1 Phonology
Phonology concerns phonemes, which are the smallest units of sound that differentiates words in a
language. For example, “cat” and “bat” are differentiated by the phoneme /k/ and /b/, respectively.
2.2 Morphology
Morphology concerns the structure of words formed into smaller units called morphemes which
deals with the word formation. For example, “jumped” is formed from the root word “jump” and the
suffix “-ed”.
, 2.3 Syntax
Syntax is the structure of sentences concerning the rules for combining words to form sentences and
deals with the relationship of words and the order in the sentence. For example, “The dog chased
the cat” is formed in the specific order: subject – verb – object.
In syntax only the rules are considered, and the meaning is ignored, it’s strictly only structural. There
are two types of syntax, 1. The word order and 2. Inflection which is the correct formation of the
words such as singular/plural.
Another important concept to syntax is grammar and parsing. Grammar defines as the set of rules
which describe what syntactic structure of sentences are allowed. Parsing is the process of producing
the syntactic structure of a sentence.
2.4 Pragmatics
Pragmatics is the study of language use in context, dealing with the use of language by a speaker
and the interpretation of this language by the listener such as by speech acts. For example, “Can
you pass the salt?” is not only a request, but also a polite way of asking a person to pass the salt.
2.5 Discourse
Discourse concerns the study of how sentences are connected and how meaning is constructed
within a discourse. For example, “he” in a sentence can cause ambiguity, but this can be resolved by
tracing this back to its origin within a larger discourse in which it appears (see Discourse ambiguity).
3. Grammatical Structure
In NLP, there are two main approaches to modelling the structure of sentences: constituency
structure and dependency structure.
3.1 Constituency Structure
A way of representing the grammatical structure of a sentence as a tree (also known as a parse tree)
is called a constituency structure. Herein, a sentence is composed of smaller units called constituents
which are further broken down into smaller constituents or words. These constituents are identified
by their function in the sentence or type of sentence they form. They are often used in tasks
involving parsing with the goal to identify the syntactic structure of a sentence (syntax).
A word is a constituent when it 1. Can be replaced by another element, for example “She read it” can
be “He read it” or “She read that”, this is called replacement, 2. Can be moved around in the
sentence and keep its structural unity, for example “The teacher read a book in the library” can
become “In the library the teacher read a book”, this is called permutation, or 3. Can be coordinated,
for example “The new teacher ate an apple” and “The new teacher and the student ate an apple”
(addition of e.g.: and/or), this is called coordination.
In case of ambiguity in the parsing of a constituency tree we want to aim for the best, correct, and/or
intended one. We can do this by probabilistic parsing in which we assign scores to trees, in other
words we want to find P(x|y): the probability of x (a word), given y (the previous word). We refer to
this probability model over sentences as a (simple form of a) language model (LM). The equation is
P(x|y) = freq(y, x) / freq(y).
3.2 Dependency Structure
Another way of representing the grammatical structure of a sentence is by defining the relationship
between words represented as dependencies. Each word is a node in the tree and the
dependencies between words are directed edges, which indicate the grammatical relationship such
Voordelen van het kopen van samenvattingen bij Stuvia op een rij:
Verzekerd van kwaliteit door reviews
Stuvia-klanten hebben meer dan 700.000 samenvattingen beoordeeld. Zo weet je zeker dat je de beste documenten koopt!
Snel en makkelijk kopen
Je betaalt supersnel en eenmalig met iDeal, creditcard of Stuvia-tegoed voor de samenvatting. Zonder lidmaatschap.
Focus op de essentie
Samenvattingen worden geschreven voor en door anderen. Daarom zijn de samenvattingen altijd betrouwbaar en actueel. Zo kom je snel tot de kern!
Veelgestelde vragen
Wat krijg ik als ik dit document koop?
Je krijgt een PDF, die direct beschikbaar is na je aankoop. Het gekochte document is altijd, overal en oneindig toegankelijk via je profiel.
Tevredenheidsgarantie: hoe werkt dat?
Onze tevredenheidsgarantie zorgt ervoor dat je altijd een studiedocument vindt dat goed bij je past. Je vult een formulier in en onze klantenservice regelt de rest.
Van wie koop ik deze samenvatting?
Stuvia is een marktplaats, je koop dit document dus niet van ons, maar van verkoper s.2. Stuvia faciliteert de betaling aan de verkoper.
Zit ik meteen vast aan een abonnement?
Nee, je koopt alleen deze samenvatting voor €8,49. Je zit daarna nergens aan vast.
Is Stuvia te vertrouwen?
4,6 sterren op Google & Trustpilot (+1000 reviews)
Afgelopen 30 dagen zijn er 64438 samenvattingen verkocht
Opgericht in 2010, al 14 jaar dé plek om samenvattingen te kopen