100% found this document useful (1 vote)
2K views5 pages

NLP UNIT IV Notes

Download as docx, pdf, or txt
Download as docx, pdf, or txt
Download as docx, pdf, or txt
You are on page 1/ 5

Natural Language Processing

R18 B.Tech. CSE (AIML) III & IV Year JNTU Hyderabad

Prepared by
K SWAYAMPRABHA
Assistance Professor
UNIT IV

Predicate-Argument Structure, Meaning Representation Systems, Software.

Predicate-Argument Structure
Verbal predicates that demand just a subject argument (e.g. sleep, work, relax) are
intransitive, verbal predicates that demand an object argument as well (e.g. like, fry,
help) are transitive, and verbal predicates that demand two object arguments are
ditransitive (e.g. give, lend).

A thing that refers to the type of event or state we are dealing with is termed a
predicate, while the things that refer to the participants in the event/state are called
the arguments of the predicate.
Expressiveness -- Ability to express wide range of subject matter. The ideal
situation: a single meaning representation language that could adequately represent
the meaning of any sensible natural language utterance. Although this ideal situation
may not be possible, but the first order predicate calculus (FOPC) is expressive
enough to handle a lot of things. In fact, it is claimed that anything can be
representable with other three representation language, it can be also representable
with FOPC. We will concentrate on FOPC, but other representation languages are
also used. For example, Text Meaning Representation (TMR) used in the machine
translation system of NMSU is a frame based representation

Predicate-Argument Structure
All natural languages have a form of predicate-argument arrangement at the core of
their semantic structure. Specific relations hold among the constituent words and
phrases of the sentence. (predicate and its arguments)Our meaning representation
should support the predicate-argument structure induced by the language. In fact,
there is a relation between syntactic frames and semantic frames. We will try to find
these relations between syntactic frames and semantic frames
.Example:Want(somebody,something) -- Want is predicate with two arguments

Syntactic Structures:I want Turkish food. NP want NPI want to spend less than
five dollars. NP want InfVPI want it to be close by here. NP want NP Inf VP Verb sub-
categorization rules allow the linking of the arguments of syntactic structures with
the semantic roles of these arguments in the semantic representation of that
sentence. The study of semantic roles associated with verbs is known as thematic
role.In syntactic structures, there are restrictions on the categories of their
arguments. Similarly, there are also semantic restrictions on the arguments of the
predicates. The selectional restrictions specify semantic restrictions on the
arguments of verbs.
Other objects (other than verbs) in natural languages may have predicate-argument
structure. A Turkish restaurant under fifteen dollars. Under(TurkishRestaurant,
$15)meaning representation is associated with the preposition under. The
preposition under can be characterized by a two-argument predicate. Make a
reservation for this evening for a table for two persons at
8.Reservation(Hearer,Today,8PM,2)meaning representation is associated with the
noun reservation (not with make).Our meaning representation should
support :variable arity predicate-argument structuresthe semantic labeling of
arguments to predicatessemantic constraints on the fillers of argument roles.
First Order Predicate Calculus (FOPC)
First Order Predicate Calculus (FOPC) is a flexible, well-understood, and
computationally tractable approach. So, FOPC satisfies the most of the things that
we expect from a meaning representation language. FOPC provides a sound
computational basis for verifiability, inference, and expressiveness requirements.
The most attractive feature of FOPC is that it makes very few specific commitments
for how things should be represented.
FOPC ExampleI only have five dollars and I don’t have a lot of time.
Have(Speaker,LotOfTime) Have(Speaker,FiveDollars)
A restaurant that serves Turkish food near Bilkent.
Serves(x,TurkishFood) x Restaurant(x) Near(LocationOf(x),LocationOf(Bilkent))
All vegetarian restaurants serve vegetarian food.
Serves(x,VegetarianFood)x VegetarianRestuarant(x) 

Semantics of FOPC The truth value of each FOPC formula can be computed using
meanings of the elements of FOPC.
   Truth tables for  Meanings of Assigned meanings to Predicates, Constant,
Functions in an interpretation. The truth values of our examples:
Have(Speaker,LotOfTime) Have(Speaker,FiveDollars)  Serves(x,TurkishFood) x
Restaurant(x) Near(LocationOf(x),LocationOf(Bilkent)) Serves(x,VegetarianFood)x
VegetarianRestuarant(x) 
An argument expressed with sentences in predicate logic is valid if and only if the
conclusion is true in every interpretation in which all the premises are true.
Meaning Representation Systems

In NLP training, the senses through which we experience the world, are referred to
as representational systems.

The words used to describe these systems are called predicates.

Hence, you have several different systems working for you:

Visual predicates (for things you see),

auditory predicates (for things you hear),

kinaesthetic predicates (things you feel or tactile


sensations), AD or self talk predicates (also called
labelling system),

olfactory predicates (things you smell), and


gustatory predicates (things you taste).
The Neuro Linguistic Programming community says that we all have a system
that we prefer to another and hence, process most information through it. This is, of
course, a generalisation, yet it will give you a clue as to what might be going on right
inside someone else’s’ internal map of the world. This preferred system can be
determined through physiology and predicates used in a conversation. Let’s look at
an example:
-The way I look at this, is that it is still rather unclear. (Almost totally visual
predicates).
-I really don’t feel comfortable with this. (Tactile/Kinaesthetic predicates).
-This doesn’t sound right to me. (Auditory predicates).
Knowing the preferred representational system (and therefore the predicates they
prefer) of someone is useful in virtually any context. Imagine you go to Spain and
communicate in English. Chances are, the general gist of what you want to say will
get across (hopefully!), yet the finer distinctions of what you are saying are lost. It is
the same with representational systems. If you present information in somebody
else’s preferred system, this information will be virtually irresistible to them!
Below is a chart which outlines the types of words/predicates people will use more
often than not in a preferred system. Remember the preferred system can change
depending on the context they are in.
Predicates Chart
Visual
Memorize by seeing pictures and are less distracted by noise. Often have trouble
remembering and are bored by long verbal ‘instructions because their mind may
wander. They are interested by how the program looks.
Auditory
Typically are easily distracted by noise. They can repeat things back to you easily &
learn by listening. They like music and like to talk on the phone. Tone of voice and
the words used can be important.
Kinaesthetic
Often they talk slowly and breathy. They respond to physical rewards & touching.
They memorize by doing or walking through something. They will be interested in a
program that feels right or gives them a gut feeling.
AD/Labelling
They spend a fair amount of time talking to themselves. They memorize by steps,
procedures, sequences. They will want to know the program makes sense. They can
also sometimes exhibit characteristics of other rep systems.
 See  Hear  Feel  Sense
 Look  Listen  Touch  Experience
 View  Sound(s)  Grasp  Understand
 Appear  Make music  Get hold of  Think
 Show  Harmonize  Slip through  Learn
 Dawn  Tune in/out  Catch on  Process
 Reveal  Be all ears  Tap into  Decide
 Envision  Rings a bell  Make contact  Motivate
 Illuminat  Silence  Throw out  Consider
e  Be heard  Turn around  Change
 Imagine  Resonate  Hard  Perceive
 Clear  Deaf  Unfeeling  Insensitive
 Foggy  Mellifluous  Concrete  Distinct
 Focused  Dissonance  Conceive
 Hazy  Question  Know
 Crystal  Unhearing
 picture

V Visual
People who are visual often stand or sit with their heads and/or bodies erect, with
their eyes up. They will be breathing from the top of their lungs. They often sit
forward in their chair and tend to be organized, neat, well-groomed and orderly.
They memorize by seeing pictures, and are less distracted by noise. They often
have trouble remembering verbal instructions because their minds tend to wander.
A visual person will be interested in how your program LOOKS. Appearances are
important to them. They are often thin and wiry.
A Auditory
People who are auditory will move their eyes sideways (remember Richard Nixon?).
They breathe from the middle of their chest. They typically talk to themselves, and
are easily distracted by noise. (some even move their lips when they talk to
themselves.) They can repeat things back to you easily, they learn by listening, and
usually like music and talking on the phone. They memorize by steps, procedures,
and sequences (sequentially). The auditory person likes to be TOLD how they’re
doing, and responds to a certain tone of voice or set of words. They will be
interested in what you have to say about your program.
K Kinaesthetic
People who are kinaesthetic will typically be breathing from the bottom of their
lungs, so you’ll see their stomach go in and out when they breathe. They often
move and talk verrry slooowly. They respond to physical rewards, and touching.
They also stand closer to people than a visual person. They memorize by doing or
walking through something. They will be interested in your program if it “feels
right”.
Ad Auditory Digital (self talk, or labelling system)
This person will spend a fair amount of time talking to themselves. They will want to
know if your program “makes sense”. The auditory digital person can exhibit
characteristics of the other major representational systems.

You might also like