© Johan Bos November 2005 Carol Beer (Little Britain)

© J

oh

an B

os

No

vem

ber

200

5

Carol Beer (Little Britain)

© J

oh

an B

os

No

vem

ber

200

5

Computer says “no”

© J

oh

an B

os

No

vem

ber

200

5

Question Answering

• Lecture 1 (Today):Introduction; History of QA; Architecture of a QA system; Evaluation.

• Lecture 2 (Friday):Question Classification; NLP techniques for question analysis; POS-tagging; Parsing; Semantic analysis; WordNet.

• Lecture 3 (Next Monday):Retrieving Answers; Document pre-processing; Tokenisation; Stemming; Lemmatisation; Named Entity Recognition; Anaphora Resolution; Matching; Use of knowledge resources; Reranking; Sanity checking.

© J

oh

an B

os

No

vem

ber

200

5

What is Question Answering?

?

© J

oh

an B

os

No

vem

ber

200

5

Information Pinpointing

Information required: Average number of car accidents per year in Sweden.

Two ways of getting this information:- Ask Google or a similar search engine

(good luck!)- Ask a QA system the question:

What’s the rate of car accidents in Sweden?

© J

oh

an B

os

No

vem

ber

200

5

QA vs IR

• Traditional method for information access: IR (Information Retrieval)

– Think of IR as finding the “right book in a library”

– Think of QA as a “librarian giving you the book and opening it on the page with the information you’re looking for”

© J

oh

an B

os

No

vem

ber

200

5

QA vs IE

• Traditional method for information access: IE (Information Extraction)

– Think of IE as finding answers to a pre-defined question (i.e., a template)

– Think of QA as asking any question you like

© J

oh

an B

os

No

vem

ber

200

5

What is Question Answering?

• Questions in natural language, not queries!

• Answers, not documents!

© J

oh

an B

os

No

vem

ber

200

5

Why do we need QA?

• Information overload problem

• Accessing information using traditional methods such as IR and IE are limited

• QA increasingly important because:– Size of available information grows– There is duplicate information– There is false information– More and more “computer illiterates”

accessing electronically stored information

© J

oh

an B

os

No

vem

ber

200

5

Information Avalanche

• Available information is growing*:– 1999: 250MB pp for each person on earth– 2002: 800MB pp for each person on earth

• People want specific information

* source: M.de Rijke 2005

© J

oh

an B

os

No

vem

ber

200

5

People ask Questions*

* source: M.de Rijke 2005

© J

oh

an B

os

No

vem

ber

200

5

Why is QA hard? (1/3)

• Questions are expressed in natural language (such as English or Italian)

• Unlike formal languages, natural languages allow a great deal of flexibility

• Example:– What is the population of Rome?– How many people live in Rome?– What’s the size of Rome?– How many inhabitants does Rome have?

© J

oh

an B

os

No

vem

ber

200

5


• Answers are expressed in natural language (such as English or Italian)

• Unlike formal languages, natural languages allow a great deal of flexibility

• Example:…is estimated at 2.5 million residents…

… current population of Rome is 2817000…

…Rome housed over 1 million inhabitants…

© J

oh

an B

os

No

vem

ber

200

5


• Answers could be spread across different documents

• Examples:– Which European countries produce wine?

[Document A contains information about Italy, and document B about France]

– What does Bill Clinton’s wife do for a living?[Document A explains that Bill Clinton’s wife is Hillary Clinton, and Document B tells us that she’s a politician]

© J

oh

an B

os

No

vem

ber

200

5

History of QA (de Rijke & Webber 2003)

• QA is by no means a new area!• Simmons (1965) reviews 15

implemented and working systems• Many ingredients of today’s QA

systems are rooted in these early approaches

• Database oriented systems, domain independent, as opposed to today’s systems that work on large sets of unstructured texts

© J

oh

an B

os

No

vem

ber

200

5

Examples of early QA systems

• BASEBALL (Green et al. 1963)Answers English questions about scores, locations and dates of baseball games

• LUNAR (Woods 1977)Accesses chemical data on lunar material compiled during the Apollo missions

• PHLIQA1 (Scha et al. 1980)Answers short questions against a database of computer installations in Europe

© J

oh

an B

os

No

vem

ber

200

5

Recent work in QA

• Since the 1990s research in QA has by and large focused on open-domain applications

• Recently interest in restricted-domain QA has increased, in particular in commercial applications– Banking, entertainment, etc.

© J

oh

an B

os

No

vem

ber

200

5

Architecture of a QA system

IRQuestion Analysis

query

Document Analysis

Answer Extraction

question

answer-type

question representation

documents/passages

passage representation

corpus

answers

© J

oh

an B

os

No

vem

ber

200

5

Question Analysis

• Input:Natural Language Question

• Output:Expected Answer Type(Formal) Representation of Question

• Techniques used:Machine learning, parsing

© J

oh

an B

os

No

vem

ber

200

5

Document Analysis

• Input:Documents or Passages

• Output:(Formal) Representation of Passages that might contain the answer

• Techniques used:Tokenisation, Named Entity Recognition, Parsing

© J

oh

an B

os

No

vem

ber

200

5

Answer Retrieval

• Input:Expected Answer TypeQuestion (formal representation)Passages (formal representation)

• Output:Ranked list of answers

• Techniques used:Matching, Re-ranking, Validation

© J

oh

an B

os

No

vem

ber

200

5

Example Run

IRQuestion Analysis

query

Document Analysis

Answer Extraction

question

answer-type


documents/passages


corpus

answers

© J

oh

an B

os

No

vem

ber

200

5

Example Run

IRQuestion Analysis

query

Document Analysis

Answer Extraction

question

answer-type


documents/passages


corpus

answers

How long is the river Thames?

© J

oh

an B

os

No

vem

ber

200

5

Example Run

IRQuestion Analysis

query

Document Analysis

Answer Extraction

question

answer-type


documents/passages


corpus

answers

length river thames

© J

oh

an B

os

No

vem

ber

200

5

Example Run

IRQuestion Analysis

query

Document Analysis

Answer Extraction

question

answer-type


documents/passages


corpus

answers

MEASURE

© J

oh

an B

os

No

vem

ber

200

5

Example Run

IRQuestion Analysis

query

Document Analysis

Answer Extraction

question

answer-type


documents/passages


corpus

answers

Answer(x) & length(y,x) & river(y) & named(y,thames)

© J

oh

an B

os

No

vem

ber

200

5

Example Run

IRQuestion Analysis

query

Document Analysis

Answer Extraction

question

answer-type


documents/passages


corpus

answers

A: NYT199802-31B: APW199805-12C: NYT200011-07

© J

oh

an B

os

No

vem

ber

200

5

Example Run

IRQuestion Analysis

query

Document Analysis

Answer Extraction

question

answer-type


documents/passages


corpus

answers

A: 30(u) & mile(u) & length(v,u) & river(y)

B: 60(z) & centimeter(z) & height(v,z) & dog(z)

C: 230(u) & kilometer(u) & length(x,u) & river(x)

© J

oh

an B

os

No

vem

ber

200

5

Example Run

IRQuestion Analysis

query

Document Analysis

Answer Extraction

question

answer-type


documents/passages


corpus

answers

C: 230 kilometerA: 30 milesB: 60 centimeter

© J

oh

an B

os

No

vem

ber

200

5

Evaluating QA systems

• International evaluation campaigns for QA systems (open domain QA):– TREC (Text Retrieval Conference)

http://trec.nist.gov/– CLEF (Cross Language Evaluation Forum)

http://clef-qa.itc.it/– NTCIR (NII Test Collection for IR Systems)

http://www.slt.atr.jp/CLQA/

© J

oh

an B

os

No

vem

ber

200

5

TREC-QA (organised by NIST)

• Annual event, started in 1999• Difficulty of the QA task increased over

the years:– 1999: Answers in snippets, ranked list of

answers;– 2005: Exact answers, only one answer.

• Three types of questions:– Factoid questions– List questions– Definition questions

© J

oh

an B

os

No

vem

ber

200

5

QA@CLEF

• CLEF is the “European edition” of TREC• Monolingual (non-English) QA

– Bulgarian (BG), German (DE), Spanish (ES), Finnish (FI), French (FR), Italian (IT), Dutch (NL), Portuguese (PT)

• Cross-Lingual QA – Questions posed in source language, answer

searched in documents of target language – All combinations possible

© J

oh

an B

os

No

vem

ber

200

5

Open-Domain QA

• QA at TREC is considered “Open-Domain” QA – Document collection is Acquint Corpus

(over a million documents)– Questions can be about anything

• Restricted-Domain QA– Documents described a specific domain– Detailed questions– Less redundancy of answers!

© J

oh

an B

os

No

vem

ber

200

5

TREC-type questions

• Factoid questions– Where is the Taj Mahal?

• List questions– What actors have played Tevye in

`Fiddler on the Roof'?

• Definition/biographical questions– What is a golden parachute?– Who is Vlad the Impaler?

© J

oh

an B

os

No

vem

ber

200

5

What is a correct answer?

• Example Factoid Question– When did Franz Kafka die?

• Possible Answers:– Kafka died in 1923.– Kafka died in 1924.– Kafka died on June 3, 1924 from

complications related to Tuberculosis.– Ernest Watz was born June 3, 1924.– Kafka died on June 3, 1924.

© J

oh

an B

os

No

vem

ber

200

5





Incorrect

© J

oh

an B

os

No

vem

ber

200

5





Inexact(under-informative)

© J

oh

an B

os

No

vem

ber

200

5


• Example Question– When did Franz Kafka die?



Inexact(over-informative)

© J

oh

an B

os

No

vem

ber

200

5





Unsupported

© J

oh

an B

os

No

vem

ber

200

5





Correct

© J

oh

an B

os

No

vem

ber

200

5

Answer Accuracy

# correct answers

Answer Accuracy = ---------------------------

# questions

© J

oh

an B

os

No

vem

ber

200

5

Correct answers to list questions

System A:

France

Italy

Example List Question

Which European countries produce wine?

System B:

Scotland France

Germany Italy

Spain Iceland

Greece

the Netherlands

Japan

Turkey

Estonia

© J

oh

an B

os

No

vem

ber

200

5

Evaluation metrics for list questions

• Precision (P): # answers judged correct & distinct P = ---------------------------------------------- # answers returned

• Recall (R): # answers judged correct & distinct R = ------------------------------------------------ # total answers

• F-Score (F): 2*P*R F = ------------ P+R

© J

oh

an B

os

No

vem

ber

200

5

Correct answers to list questions

System A:

France

Italy

Example List Question

Which European countries produce wine?

System B:

Scotland France

Germany Italy

Spain Iceland

Greece

the Netherlands

Japan

Turkey

Estonia

P = 1.00

R = 0.25

F = 0.40

P = 0.64

R = 0.88

F = 0.74

© J

oh

an B

os

No

vem

ber

200

5

Other evaluation metrics

System A: Ranked answers (Accuracy = 0.2) Q1 Q2 Q3 Q4 Q6 Q7 Q8 Q9 …. Qn

A1 W W C W C W W W …. W

A2 W W W W W W W W …. W



A5 W C W W W C W W …. W

System B: Ranked answers (Accuracy = 0.1) Q1 Q2 Q3 Q4 Q6 Q7 Q8 Q9 …. Qn

A1 W W W W C W W W …. W

A2 C W C W W C C W …. C

A3 W C W W W W W W …. W

A4 W W W C W W W W …. W


© J

oh

an B

os

No

vem

ber

200

5

Mean Reciprocal Rank (MRR)

• Score for an individual question:– The reciprocal of the rank at which

the first correct answer is returned – 0 if no correct response is returned

• The score for a run: – Mean over the set of questions in the test

© J

oh

an B

os

No

vem

ber

200

5

MRR in action

System A: MRR = (.2+1+1+.2)/10 = 0.24 Q1 Q2 Q3 Q4 Q6 Q7 Q8 Q9 …. Qn

A1 W W C W C W W W …. W




A5 W C W W W C W W …. W

System B: MRR = (.5+.33+.5+.25+1+.5+.5+.5)/10=0.42 Q1 Q2 Q3 Q4 Q6 Q7 Q8 Q9 …. Qn

A1 W W W W C W W W …. W

A2 C W C W W C C W …. C

A3 W C W W W W W W …. W

A4 W W W C W W W W …. W


© J

oh

an B

os

No

vem

ber

200

5

Open-Domain Question Answering

• TREC QA Track– Factoid questions– List questions– Definition questions

• State-of-the-Art– Hard problem– Only few systems with

good results

Accuracy TREC 2004 (n=28)

0 10

0.0-0.1

0.2-0.3

0.4-0.5

0.6-0.7

© J

oh

an B

os

No

vem

ber

200

5

Friday

• QA Lecture 2:– Question Classification– NLP techniques for question analysis– POS-tagging– Parsing– Semantic analysis– Use of lexical resources such as WordNet

© J

oh

an B

os

No

vem

ber

200

5Question Classification

(preview)• How many islands does Italy have?• When did Inter win the Scudetto?• What are the colours of the Lithuanian flag?• Where is St. Andrews located?• Why does oil float in water?• How did Frank Zappa die?• Name the Baltic countries.• Which seabird was declared extinct in the 1840s?• Who is Noam Chomsky?• List names of Russian composers.• Edison is the inventor of what?• How far is the moon from the sun?• What is the distance from New York to Boston?• How many planets are there?• What is the exchange rate of the Euro to the Dollar?• What does SPQR stand for?• What is the nickname of Totti?• What does the Scottish word “bonnie” mean?• Who wrote the song “Paranoid Android”?