Lecture 1___ Introduction to Building NLP Applications (1).pdf

(1428 KB) Pobierz
INTRODUCTION TO
“BUILDING NLP
APPLICATIONS”
Wlodek Zadrozny
Fall Lectures 2014
Why are we building NLP applications?
1.
Great challenge.
NLP is an AI-
complete problem
Image
sources:
http://en.wiki
pedia.org/wi
ki/Artificial_i
ntelligence;
http://www.ro
botappstore.
com/Robope
dia/Artificial-
intelligence
Why are we building NLP applications?
1.
Great challenge. NLP is
an AI-complete problem
2.
It is a next step in
searching for information
•
Question answering:
knowledge when it is
needed
•
Semantic search: searching
by meaning
Image sources:
http://en.wikipedia.org/wiki/Artifici
al_intelligence;
http://singularityhub.com/
Recent headlines
•
Scientists develop lie
•
The Story Behind the SAT
detector for social media
Overhaul
•
Advanced Text Mining
•
CRM Watchlist 2014
Winners: Customer
Engagement Hottest
Market
•
Diving deep into data to
Improves Medicare
Advantage Coding
•
Capital Market Exchange
crack the gene code on
disease
to Discuss Transforming
Investor Opinions into
Actionable Trading Signals
at Upcoming Sentiment
Analysis Symposium
Why are we building NLP applications?
1.
Great challenge. NLP is an
AI-complete problem
2.
Next step in searching for
information
•
Question answering:
knowledge when it is needed
•
Semantic search: searching by
meaning
3.
Part of “big data”. Novel
and/or highly profitable
applications
Image sources:
Wikipedia,
http://publicintelligence.net/cia-cto-big-data/
http://www.flickr.com/photos/marc_smith/8924975889/

Plik z chomika:

Inne pliki z tego folderu:

Inne foldery tego chomika:

Zgłoś jeśli naruszono regulamin