Lecture14 Lin

Information about Lecture14 Lin

Published on October 16, 2007

Author: Mertice

Source: authorstream.com

Content

An Introduction to Information Retrieval and Question Answering:  An Introduction to Information Retrieval and Question Answering Jimmy Lin College of Information Studies University of Maryland Wednesday, December 8, 2004 The Information Retrieval Cycle:  The Information Retrieval Cycle Source Selection Search Selection Examination Delivery Query Formulation Supporting the Search Process:  Supporting the Search Process Source Selection Search Query Selection Ranked List Examination Documents Delivery Documents Query Formulation Resource Indexing Index Acquisition Collection Types of Information Needs:  Types of Information Needs Ad hoc retrieval: find me documents “like this” Question answering Who discovered Oxygen? When did Hawaii become a state? Where is Ayer’s Rock located? What team won the World Series in 1992? Identify positive accomplishments of the Hubble telescope since it was launched in 1991. Compile a list of mammals that are considered to be endangered, identify their habitat and, if possible, specify what threatens them. What countries export oil? Name U.S. cities that have a “Shubert” theater. Who is Aaron Copland? What is a quasar? “Factoid” “List” “Definition” IR is an Experimental Science!:  IR is an Experimental Science! Formulate a research question, the hypothesis Design an experiment to answer the question Perform the experiment Compare with a baseline “control” Does the experiment answer the question? Are the results significant? Report the results! Rinse, repeat… What experiments?:  What experiments? Example “questions”: Does morphological analysis improve retrieval performance? Does expanding the query with synonyms improve retrieval performance? Corresponding experiments: Build a “stemmed” index and compare against “unstemmed” baseline Expand queries with synonyms and compare against baseline unexpanded query. What’s missing here? IR Test Collections:  IR Test Collections Three components of a test collection: Collection of documents (corpus) Set of information needs (topics) Sets of documents that satisfy the information needs (relevance judgments) Metrics for assessing “performance” Precision Recall Other measures derived therefrom Where do they come from?:  Where do they come from? TREC = Text REtrieval Conferences Series of annual evaluations, started in 1992 Organized into “tracks” Test collections are formed by “pooling” Gather results from all participants Corpus/topics/judgments can be reused Roots of Question Answering:  Roots of Question Answering Information Retrieval (IR) Information Extraction (IE) Information Retrieval (IR):  Information Retrieval (IR) Can substitute “document” for “information” IR systems Use statistical methods Rely on frequency of words in query, document, collection Retrieve complete documents Return ranked lists of “hits” based on relevance Limitations Answers questions indirectly Does not attempt to understand the “meaning” of user’s query or documents in the collection Information Extraction (IE):  Information Extraction (IE) IE systems Identify documents of a specific type Extract information according to pre-defined templates Place the information into frame-like database records Templates = pre-defined questions Extracted information = answers Limitations Templates are domain dependent and not easily portable One size does not fit all! Weather disaster: Type Date Location Damage Deaths ... Central Idea of Factoid QA:  Central Idea of Factoid QA Determine the semantic type of the expected answer Retrieve documents that have keywords from the question Look for named-entities of the proper type near keywords “Who won the Nobel Peace Prize in 1991?” is looking for a PERSON Retrieve documents that have the keywords “won”, “Nobel Peace Prize”, and “1991” Look for a PERSON near the keywords “won”, “Nobel Peace Prize”, and “1991” An Example:  An Example But many foreign investors remain sceptical, and western governments are withholding aid because of the Slorc's dismal human rights record and the continued detention of Ms Aung San Suu Kyi, the opposition leader who won the Nobel Peace Prize in 1991. The military junta took power in 1988 as pro-democracy demonstrations were sweeping the country. It held elections in 1990, but has ignored their result. It has kept the 1991 Nobel peace prize winner, Aung San Suu Kyi - leader of the opposition party which won a landslide victory in the poll - under house arrest since July 1989. The regime, which is also engaged in a battle with insurgents near its eastern border with Thailand, ignored a 1990 election victory by an opposition party and is detaining its leader, Ms Aung San Suu Kyi, who was awarded the 1991 Nobel Peace Prize. According to the British Red Cross, 5,000 or more refugees, mainly the elderly and women and children, are crossing into Bangladesh each day. Who won the Nobel Peace Prize in 1991? Generic QA Architecture:  Generic QA Architecture Question Analyzer Document Retriever Passage Retriever Answer Extractor NL question IR Query Documents Passages Answers Answer Type Question analysis:  Question analysis Question word cues Who  person, organization, location (e.g., city) When  date Where  location What/Why/How  ?? Head noun cues What city, which country, what year... Which astronaut, what blues band, ... Scalar adjective cues How long, how fast, how far, how old, ... Using WordNet:  Using WordNet wingspan length diameter radius altitude ceiling What is the service ceiling of an U-2? NUMBER Extracting Named Entities:  Extracting Named Entities Person: Mr. Hubert J. Smith, Adm. McInnes, Grace Chan Title: Chairman, Vice President of Technology, Secretary of State Country: USSR, France, Haiti, Haitian Republic City: New York, Rome, Paris, Birmingham, Seneca Falls Province: Kansas, Yorkshire, Uttar Pradesh Business: GTE Corporation, FreeMarkets Inc., Acme University: Bryn Mawr College, University of Iowa Organization: Red Cross, Boys and Girls Club More Named Entities:  More Named Entities Currency: 400 yen, $100, DM 450,000 Linear: 10 feet, 100 miles, 15 centimeters Area: a square foot, 15 acres Volume: 6 cubic feet, 100 gallons Weight: 10 pounds, half a ton, 100 kilos Duration: 10 day, five minutes, 3 years, a millennium Frequency: daily, biannually, 5 times, 3 times a day Speed: 6 miles per hour, 15 feet per second, 5 kph Age: 3 weeks old, 10-year-old, 50 years of age How do we extract NEs?:  How do we extract NEs? Heuristics and patterns Fixed-lists (gazetteers) Machine learning approaches Answer Type Hierarchy:  Answer Type Hierarchy Does it work?:  Does it work? Where do lobsters like to live? on a Canadian airline Where do hyenas live? in Saudi Arabia in the back of pick-up trucks Where are zebras most likely found? near dumps in the dictionary Why can't ostriches fly? Because of American economic sanctions What’s the population of Maryland? three Limitations?:  Limitations? Conclusion:  Conclusion Question answering is an exciting research area! Lies at the intersection of information retrieval and natural language processing A real-world application of NLP technologies The dream: a vast repository of knowledge we can “talk to” We’re a long way from there…

Related presentations


Other presentations created by Mertice

Lec11 Algae
01. 01. 2008
0 views

Lec11 Algae

carter200603
04. 10. 2007
0 views

carter200603

Jutta Immanen Poyry
09. 10. 2007
0 views

Jutta Immanen Poyry

symbiosis2
11. 10. 2007
0 views

symbiosis2

street children denver handouts
12. 10. 2007
0 views

street children denver handouts

252b lecture1
15. 10. 2007
0 views

252b lecture1

CROP Cosmic Ray History 2005
15. 10. 2007
0 views

CROP Cosmic Ray History 2005

BMI705 Lecture1
15. 10. 2007
0 views

BMI705 Lecture1

Chapter31
22. 10. 2007
0 views

Chapter31

posters zwickel
22. 10. 2007
0 views

posters zwickel

Chain reactions
16. 10. 2007
0 views

Chain reactions

HistoryTalk4
23. 10. 2007
0 views

HistoryTalk4

73
24. 10. 2007
0 views

73

ps 22
09. 10. 2007
0 views

ps 22

Matthew Bick Panama Canal
25. 10. 2007
0 views

Matthew Bick Panama Canal

1 Introduction Y050823
02. 11. 2007
0 views

1 Introduction Y050823

TORNADO DRAGON
02. 10. 2007
0 views

TORNADO DRAGON

qcd3
27. 09. 2007
0 views

qcd3

T Maruyama
09. 10. 2007
0 views

T Maruyama

Cloud Formation
05. 01. 2008
0 views

Cloud Formation

2002interimresults
07. 01. 2008
0 views

2002interimresults

Islamic Empires 2
07. 01. 2008
0 views

Islamic Empires 2

Module54 Aggregate Project Plan
07. 01. 2008
0 views

Module54 Aggregate Project Plan

change prof
27. 09. 2007
0 views

change prof

InformationSystem
17. 10. 2007
0 views

InformationSystem

AFS explanation
04. 01. 2008
0 views

AFS explanation

Crovella
05. 10. 2007
0 views

Crovella

dwyer SAC 2005 v2
23. 10. 2007
0 views

dwyer SAC 2005 v2

loi entreprise
23. 10. 2007
0 views

loi entreprise

sovietep
31. 12. 2007
0 views

sovietep

CROP Ionizing Detectors
13. 10. 2007
0 views

CROP Ionizing Detectors

RUSSIA Sergey Tikhonov
26. 11. 2007
0 views

RUSSIA Sergey Tikhonov

510 shortPRESEN
10. 10. 2007
0 views

510 shortPRESEN

leethao
19. 02. 2008
0 views

leethao

SelfInjury
20. 02. 2008
0 views

SelfInjury

MCOR 384 Presentation Okinawa
26. 02. 2008
0 views

MCOR 384 Presentation Okinawa

Lecture19 Uranus Neptune
15. 11. 2007
0 views

Lecture19 Uranus Neptune

Concrete
29. 02. 2008
0 views

Concrete

Chapter 1 Intro
04. 03. 2008
0 views

Chapter 1 Intro

Forum Casablanca
24. 10. 2007
0 views

Forum Casablanca

fulbright3
10. 03. 2008
0 views

fulbright3

Lecture One
13. 03. 2008
0 views

Lecture One

Objective1
20. 03. 2008
0 views

Objective1

Estrategia Nacional de Comercio
22. 10. 2007
0 views

Estrategia Nacional de Comercio

TRANS WP29 132 inf09e
25. 03. 2008
0 views

TRANS WP29 132 inf09e

prohibiteddogs
19. 11. 2007
0 views

prohibiteddogs

Ch 08
09. 04. 2008
0 views

Ch 08

phase out
11. 04. 2008
0 views

phase out

nossdav99pk
16. 04. 2008
0 views

nossdav99pk

qos3
17. 04. 2008
0 views

qos3

RACEM
23. 11. 2007
0 views

RACEM

StuMbr Benefits
18. 04. 2008
0 views

StuMbr Benefits

Tim SBS
22. 04. 2008
0 views

Tim SBS

Module12
28. 04. 2008
0 views

Module12

5 1
11. 10. 2007
0 views

5 1

Be an RT in NC
02. 05. 2008
0 views

Be an RT in NC

452 lecture7
02. 05. 2008
0 views

452 lecture7

5 10
30. 10. 2007
0 views

5 10

Ito
09. 10. 2007
0 views

Ito

sae intro
29. 12. 2007
0 views

sae intro

lecture001
29. 09. 2007
0 views

lecture001

UMDNJ Connell oct29
21. 10. 2007
0 views

UMDNJ Connell oct29

060619 OIF Foisel iPOP2006 01
09. 10. 2007
0 views

060619 OIF Foisel iPOP2006 01

For Every Season 2003
13. 11. 2007
0 views

For Every Season 2003

AMCARO Mineral Ltd
22. 10. 2007
0 views

AMCARO Mineral Ltd

casablanca 1
23. 10. 2007
0 views

casablanca 1

Mito Cinese
24. 10. 2007
0 views

Mito Cinese

CharetteAASL
03. 10. 2007
0 views

CharetteAASL

nfb social impact fr
11. 03. 2008
0 views

nfb social impact fr

Development aid
23. 12. 2007
0 views

Development aid

SFDVpresentation
30. 10. 2007
0 views

SFDVpresentation

ITEK Alvento
02. 11. 2007
0 views

ITEK Alvento

Paraphrasing New
15. 10. 2007
0 views

Paraphrasing New

yurtdisi sunu
23. 11. 2007
0 views

yurtdisi sunu

soldering progress
12. 10. 2007
0 views

soldering progress

Agustin Carstens
10. 04. 2008
0 views

Agustin Carstens

small mediumschools
08. 10. 2007
0 views

small mediumschools

Congres RMRA 07 Marrakech 2007
24. 10. 2007
0 views

Congres RMRA 07 Marrakech 2007

overview 2004 MF Meeting v2
29. 10. 2007
0 views

overview 2004 MF Meeting v2