Loading presentation...

Present Remotely

Send the link below via email or IM


Present to your audience

Start remote presentation

  • Invited audience members will follow you as you navigate and present
  • People invited to a presentation do not need a Prezi account
  • This link expires 10 minutes after you close the presentation
  • A maximum of 30 users can follow your presentation
  • Learn more about this feature in our knowledge base article

Do you really want to delete this prezi?

Neither you, nor the coeditors you shared it with will be able to recover it again.


BSIM3004 Natural Language Processing

Tutorial 5

Kitman Chan

on 30 September 2015

Comments (0)

Please log in to add your comment.

Report abuse

Transcript of BSIM3004 Natural Language Processing

Tutorial 5:
Natural Language Search

Cutting edge technology
User friendly
Easy access to data
Requires companies to continue to improve
search speed and results:
Google vs. Bing
Voice recognition and prediction:
Apple vs. Samsung vs. Google
Ability to find and recommend:
Travel sites, social networks etc.
Convert the words listed in an enquiry into a meaning that matches the requester’s intention
Translate a sentence into values an algorithm can process
Match query meanings with meanings of documents
Using stemming, recognition of different grammatical forms, context,…
Rank documents in order of relevance
Syntactic & semantic interpretation techniques
What is Natural Language?
human language
in which the structure & rules have evolved from usage, usually over an extended period of time
what we use as an everyday means of communication among humans
Natural Language Search
Why is Natural Language Processing Important?
How it works?
What is Natural Language?
What is Natural Language Processing (NLP)?
What are the applications?
How to process?
Why is it important?
Natural Language Processing
Interaction between
human languages
IRS designed to handle input expressed in natural language
User may enter the query in the same form in which it would be spoken or written
Natural language query examples
“Why is sky blue?”
“Who is the US President?”

Combines the fields of:
computer science
Artificial Intelligence
Statistical NLP
Grammar framework
Assigns probabilities to parts of speech
Multiple approaches:
Stochastic context free grammar
Statistical parsing
Data-oriented parsing
N-gram Statistical Models
Hidden Markov model
Estimation theory
Uses of Natural Language Processing
Natural Language Search
Speech-to-text Translation
Apple Siri
Samsung S Voice
Google Voice
Challenges with Natural Language Processing
Different languages
Different grammatical format & rules

Long sentences processing
Words with multiple definitions
Syntactic Tree
known as parsing trees
Ordered, rooted tree
Represents the syntactic structure of a string according to a formal grammar (follow grammatical & structural rules)
Verb Phrase
Noun Phrase
All languages have ambiguity
Caused by words with multiple meanings, grammatical variance, etc.
Process the meaning of the text: sentence, passage, document
The success of natural language retrieval relies on the meaning (semantics) of the documents retrieved
But the meaning of a word or phrase does not always represent its literal meaning, e.g.:
‘a grain of salt’
'queer fish'
Semantic Analysis
Hypergraph Example
Numbers between words are string positions
(at least) two parse trees:
"with a mirror" can be attached to "saw" or "him"
Natural Language Search

search using regular spoken language
Natural Language Retrieval vs
Keywords–based Retrieval
Credit: Nicole Grieble
Full transcript