Artificial Intelligence

(CS F407)

Understanding Natural Languages

Chittaranjan Hota
Professor, Computer Science & Information Systems Department
BITS Pilani Hyderabad Campus, Hyderabad
[email protected]
 Natural Language Processing

Make computers learn our language rather than we learn

theirs.
Yahoo, Google, Microsoft  Info. Retrieval

आप कह ाँ ज रहे  Machine Translation

Ask Jeeves  Question Answering

Myspace, Facebook Processing of User-Generated Content

All “Big Guys” have (several) strong NLP research labs:

IBM, Microsoft, AT&T, Xerox, Sun, etc.
Where does it fit in the CS taxonomy?

Computer Science

Databases Artificial Intelligence Systems prog. Networking

ML Natural Language Processing Search

Information Machine Language

Retrieval Translation Analysis

Semantics Parsing
 Information Extraction (IE)

• Identify phrases in language that refer to specific

types of entities and relations in the text.
• Named entity recognition is the task of identifying
names of people, places, organizations, etc. in text.
• Rajesh Gopinathan is the CEO of TCS who lives in
Mumbai.
• Relation extraction identifies specific relations
between entities.
• Rajesh Gopinathan is the CEO of TCS who lives in
Mumbai.
Text Summarization

Summarized Text
BITS conducts the all-India
computerized entrance examination,
BITSAT (BITS Admission Test).

Birla Institute of Technology & Science,

Pilani (shortened BITS Pilani or BITS)
is an Indian institute of higher
education and a deemed university
under Section 3 of the UGC Act 1956.

http://textsummarization.net/text-summarizer
 Machine Translation (MT)

• Translate a sentence from one natural language to

another.

• आप कह ाँ ज रहे
• Where are you going

• बिट् स है दर ि द कह हे
• BITS Hyderabad said,

http://translate.google.com/#
Knowledge of Language

• Phonology – concerns how words are related to the sounds

that realize them. (e.g, thumb, dumb)

• Morphology – concerns how words are constructed from more

basic meaning units called morphemes. nonperishable

• Syntax - concerns the proper ordering of

3 morphemes: non-
words and it’s effect on meaning. , perish, and -able
• The dog bit the boy.
• The boy bit the dog.
• Bit boy dog the the.
 Continued…
 Semantics concerns the (literal) meaning of words, phrases,
and sentences.
 “plant” is a manufacturing facility
 “plant” is the act of sowing

 Pragmatics concerns the overall communicative and social

context and its effect on interpretation.
 John thinks mushrooms.

 Discourse – concerns how the immediately preceding

sentences affect the interpretation of the next sentence.

 World Knowledge – includes general knowledge about the

world.
Ambiguity (I made him duck)
• Some interpretations:
• I cooked duck for him.
• I cooked duck belonging to him.
• I created a toy duck which he owns.
• I caused him to quickly lower his head or body.
• I bowled him out without scoring any runs.
• I used magic and turned him into a duck.
• duck – morphologically and syntactically
ambiguous: noun or verb.
• make – semantically ambiguous: cook, bowl, or
create.
 Syntactic user
stems
Analysis
Morphologic
al Analysis
Surface
form

Semantic
Analysis lexicon
Pragmatic
Analysis Perform
action

Internal Discourse Resolve

references
representation
Analysis

Lexicon is a vocabulary data bank, that contains the

Source: H. Al-Muhtaseb
language words and their linguistic information.
 Surface form stems

I want to print I (pronoun)

Ali’s .init file want (verb)
to (prep)
to(infinitive)
print (verb)
Ali (noun)
‘s (possessive)
.init (adj)
file (noun)
file (verb)
Source: H. Al-Muhtaseb
 S

stems NP VP
I (pronoun) V S
want (verb) PRO Parse
to (prep) NP VP tree
I
to(infinitive) want
PRO V NP
print (verb)
Ali (noun) ADJ NP
‘s (possessive) I print
.init (adj) ADJ N
file (noun) Ali’s
.init file
file (verb)
Source: H. Al-Muhtaseb
Parsing: Another example
s --> np, vp. det -->[a]. det --> [an].
np --> det, noun. det --> [the].
np --> proper_noun. noun --> [apple].
vp --> v, np. noun --> [orange].
vp --> v. proper_noun --> [john].
proper_noun --> [mary].
v --> [eats]. v --> [loves].
Eg. john eats an apple.

proper_noun det noun

v np
np vp

s
 Semantic Net
I Ali
who

who Who’s
S
NP want print file
VP what what
type
PRO V S
.init
NP VP
I
want
PRO V NP

ADJ NP
I print
ADJ N
Parse tree Ali’s
.init file
Source: H. Al-Muhtaseb
Semantic analysis
 To whom the pronoun ‘I’
refers
To whom the proper
noun ‘Ali’ refers
What are the files to be
printed
I Ali
who

who Who’s

want print file

what what
type
Semantic Net
Execute the command
.init
lpr /ali/stuff.init
•Who/when/where/what … are involved in an event?
•How to connect the semantic representations of
•Practical usage of language: what a
different sentences?
•What is the cause of an event and what is the sentence means in practice?
consequence of an action?
Thank You!