Question Answering (QA)
Lecture 2
© Johan Bos
April 2008
Lecture 1
•
•
•
•
•
•
•
•
What is QA?
Query Log Analysis
Challenges in QA
History of QA
System Architecture
Methods
System Evaluation
State-of-the-art
• Question Analysis
• Background Knowledge
• Answer Typing
Lecture 3
•
•
•
•
•
Query Generation
Document Analysis
Semantic Indexing
Answer Extraction
Selection and Ranking
Pronto QA System
parsing
question
answer
ccg
answer
reranking
boxing
drs
answer
selection
WordNet
NomLex
knowledge
© Johan Bos
April 2008
query
answer
typing
Indri
answer
extraction
Indexed Documents
Lecture 2
parsing
question
answer
ccg
answer
reranking
boxing
drs
answer
selection
WordNet
NomLex
knowledge
© Johan Bos
April 2008
query
answer
typing
Indri
answer
extraction
Indexed Documents
Question Answering (QA)
Lecture 2
© Johan Bos
April 2008
Question Analysis
• Background Knowledge
• Answer Typing
Question Analysis – Why?
© Johan Bos
April 2008
• The aim of QA is to output answers, not
documents
• We need question analysis to
– Determine the type of answer that we try
to find
– Estimate the number of answers that we
want to return
– Calculate the probability that an answer
is correct
Natural Language Processing
• We need ways to automate the process
of manipulating natural language
© Johan Bos
April 2008
– Punctuation
– The way words are composed
– The relationship between words
– The structure of phrases
– Represent meaning of phrases
• This is where NLP comes in!
– (NLP = Natural Language Processing)
How to use NLP tools?
© Johan Bos
April 2008
• There is a large set of tools available
on the web, most of it free for research
• Examples of integrated text processing
environments:
– GATE (University of Sheffield)
– TTT (University of Edinburgh)
– LingPipe
– For a general overview of NLP tools, see
http://registry.dfki.de/
– C&C (used by the Pronto QA system)
Architecture of PRONTO
parsing
question
answer
ccg
answer
reranking
boxing
drs
answer
selection
WordNet
NomLex
knowledge
© Johan Bos
April 2008
query
answer
typing
Indri
answer
extraction
Indexed Documents
Question Analysis
© Johan Bos
April 2008
•
•
•
•
•
Tokenisation
Part of speech tagging
Lemmatisation
Syntactic analysis (Parsing)
Semantic analysis (Boxing)
• Named entity recognition
• Anaphora resolution
Tokenisation
• Tokenisation is the task of splitting
words from punctuation
© Johan Bos
April 2008
– Semicolons, colons ; :
– exclamation marks, question marks
– commas and full stops . ,
– quotes “ ‘ `
!?
• Tokens are normally split by spaces
– In the following slides, we use |
Tokenisation: Example 1
• Input (9 tokens):
© Johan Bos
April 2008
When was the Buckingham Palace built in
London, England?
Tokenisation: Example 1
• Input (9 tokens):
© Johan Bos
April 2008
When | was | the | Buckingham | Palace |
built | in | London, | England?
Tokenisation: Example 1
• Input (9 tokens):
When | was | the | Buckingham | Palace |
built | in | London, | England?
© Johan Bos
April 2008
• Output (11 tokens):
When | was | the | Buckingham | Palace |
built | in | London | , | England | ?
Tokenisation: Example 2
• Input (7 tokens):
© Johan Bos
April 2008
What year did "Snow White" come out?
Tokenisation: Example 2
• Input (7 tokens):
© Johan Bos
April 2008
What | year | did | "Snow | White" | come
| out?
Tokenisation: Example 2
• Input (7 tokens):
What | year | did | "Snow | White" | come
| out?
© Johan Bos
April 2008
• Output (10 tokens):
What | year | did | “ | Snow | White | " | come
| out | ?
Tokenisation: combined words
• Combined words are split
– I’d
 I | ’d
– country’s  country | ’s
– won’t
 wo | n’t
– “don’t!”  “ do | n’t ! “
© Johan Bos
April 2008
• Some Italian examples
– gliel’ha detto
 glie | l’ | ha | detto
– posso prenderlo  posso | prender | lo
Difficulties with tokenisation
• Abbreviations, acronyms
– When was the U.S. invasion of Haiti?
© Johan Bos
April 2008
• In particular if the abbreviation or
acronym is the last word of a sentence
– Look at next word: if in uppercase, then
assume it is end of sentence
– But think of cases such as Mr. Jones
Why is tokenisation important?
© Johan Bos
April 2008
• Required for all subsequent stages of
processing
– Parsing
– Named entity recognition
– Lemmatisation
– To look up a word in an electronic
dictionary (such as WordNet)
Question Analysis
© Johan Bos
April 2008
• Tokenisation
Part of speech tagging
• Named Entity Recognition
• Lemmatisation
• Syntactic analysis (Parsing)
• Semantic analysis (Boxing)
Traditional parts of speech
• Verb
• Preposition
• Noun
• Pronoun
• Conjunction
© Johan Bos
April 2008
• Interjection
• Adjective
• Adverb
© Johan Bos
April 2008
Parts of speech in NLP
CLAWS1 (132 tags)
Penn Treebank (45 tags)
Examples:
Examples:
NN singular common noun (boy, pencil ... )
NN$ genitive singular common noun (boy's,
parliament's ... )
NNP singular common noun with word initial
capital (Austrian, American, Sioux,
Eskimo ... )
NNP$ genitive singular common noun with
word initial capital (Sioux', Eskimo's,
Austrian's, American's, ...)
NNPS plural common noun with word initial
capital (Americans, ... )
NNPS$ genitive plural common noun with
word initial capital (Americans‘, …)
NNS plural common noun (pencils, skeletons,
days, weeks ... )
NNS$ genitive plural common noun (boys',
weeks' ... )
NNU abbreviated unit of measurement
unmarked for number (in, cc, kg …)
JJ adjective (green, …)
JJR adjective, comparative (greener,…)
JJS adjective, superlative (greenest, …)
MD modal (could, will, …)
NN noun, singular or mass (table, …)
NNS noun plural (tables, …)
NNP proper noun, singular (John, …)
NNPS proper noun, plural (Vikings, …)
PDT predeterminer (both the boys)
POS possessive ending (friend's)
PRP personal pronoun (I, he, it, …)
PRP$ possessive pronoun (my, his, …)
RB adverb (however, usually, naturally, here,
good, …)
RBR adverb, comparative (better, …)
© Johan Bos
April 2008
POS tagged example
What
year
did
“
Snow
White
"
come
out
?
© Johan Bos
April 2008
POS tagged example
What
year
did
“
Snow
White
"
come
out
?
WP
NN
VBD
“
NNP
NNP
“
VB
IN
.
Why is POS-tagging important?
• To disambiguate words
• For instance, to distinguish “book” used
as a noun from “book” used as a verb
© Johan Bos
April 2008
– Where can I find a book on cooking?
– Where can I book a room?
• Prerequisite for further processing
stages, such as parsing
Question Analysis
© Johan Bos
April 2008
• Tokenisation
• Part of speech tagging
Lemmatisation
• Syntactic analysis (Parsing)
• Semantic analysis (Boxing)
Lemmatisation
• Lemmatising means
© Johan Bos
April 2008
– grouping morphological variants of words under a
single headword
• For example, you could group the words
am,
was,
are,
is,
were, and
been together under the word be
Lemmatisation
• Lemmatising means
© Johan Bos
April 2008
– grouping morphological variants of words under a
single headword
• For example, you could group the words
am,
was,
are,
is,
were, and
been together under the word be
Lemmatisation
© Johan Bos
April 2008
• Using linguistic terminology, the variants
taken together form the lemma of a lexeme
• Lexeme: a “lexical unit”, an abstraction over
specific constructions
• Other examples:
dying, die, died, dies  die
car, cars  car
man, men  man
Question Analysis
© Johan Bos
April 2008
• Tokenisation
• Part of speech tagging
• Lemmatisation
Syntactic analysis (Parsing)
• Semantic analysis (Boxing)
© Johan Bos
April 2008
What is Parsing
• Parsing is the process of assigning a
syntactic structure to a sequence of
words
• The syntactic structure is defined using
a grammar
• A grammar contains of a set of symbols
(terminal and non-terminal symbols)
and production rules (grammar rules)
• The lexicon is built over the terminal
symbols (i.e., the words)
Syntactic Categories
© Johan Bos
April 2008
• The non-terminal symbols correspond to
syntactic categories
–
–
–
–
–
–
–
–
–
–
Det (determiner)
N (noun)
IV (intransitive verb)
TV (transitive verb)
PN (proper name)
Prep (preposition)
NP (noun phrase)
PP (prepositional phrase)
VP (verb phrase)
S (sentence)
the car
at the table
saw a car
Mia likes Vincent
Example Grammar
© Johan Bos
April 2008
Lexicon
Det: which, a, the,…
N: rock, singer, …
IV: die, walk, …
TV: kill, write,…
PN: John, Lithium, …
Prep: on, from, to, …
Grammar Rules
S  NP VP
NP  Det N
NP  PN
NNN
N  N PP
VP  TV NP
VP  IV
PP  Prep NP
VP  VP PP
© Johan Bos
April 2008
The Parser
• A parser automates the process of
parsing
• The input of the parser is a string of
words (annotated with POS-tags)
• The output of a parser is a parse tree,
connecting all the words
• The way a parse tree is constructed is
also called a derivation
© Johan Bos
April 2008
Derivation Example
Which
rock singer wrote Lithium
© Johan Bos
April 2008
Lexical stage
Det
Which
N
N
TV
PN
rock singer wrote Lithium
Use rule: NP  Det N
© Johan Bos
April 2008
NP
Det
Which
N
N
TV
PN
rock singer wrote Lithium
Use rule: NP  PN
© Johan Bos
April 2008
NP
Det
Which
NP
N
N
TV
PN
rock singer wrote Lithium
Use rule: VP  TV NP
VP
© Johan Bos
April 2008
NP
Det
Which
NP
N
N
TV
PN
rock singer wrote Lithium
Backtracking
VP
© Johan Bos
April 2008
NP
Det
Which
NP
N
N
TV
PN
rock singer wrote Lithium
Use rule: N  N N
VP
© Johan Bos
April 2008
N
Det
Which
NP
N
N
TV
PN
rock singer wrote Lithium
Use rule: NP  Det N
NP
© Johan Bos
April 2008
N
Det
Which
VP
NP
N
N
TV
PN
rock singer wrote Lithium
Use rule S  NP VP
S
NP
© Johan Bos
April 2008
N
Det
Which
VP
NP
N
N
TV
PN
rock singer wrote Lithium
Wide coverage parsers
• Normally expect tokenised and
POS-tagged input
© Johan Bos
April 2008
• Example of wide-coverage parsers:
– Charniak parser
– Collins parser
– RASP (Carroll & Briscoe)
– CCG parser
(Clark & Curran – used in Pronto)
Output C&C parser
© Johan Bos
April 2008
ba('S[wq]',
fa('S[wq]',
fa('S[wq]/(S[q]/PP)',
fc('(S[wq]/(S[q]/PP))/N',
lf(1,'(S[wq]/(S[q]/PP))/(S[wq]/(S[q]/NP))'),
lf(2,'(S[wq]/(S[q]/NP))/N')),
lf(3,'N')),
fc('S[q]/PP',
fa('S[q]/(S[b]NP)',
lf(4,'(S[q]/(S[b]NP))/NP'),
lex('N','NP',
lf(5,'N'))),
lf(6,'(S[b]NP)/PP'))),
lf(7,'S[wq]S[wq]')).
w(1,'For',
w(2,which,
w(3,newspaper,
w(4,does,
w(5,'Krugman',
w(6,write,
w(7,?,
for,
which,
newspaper,
do,
krugman,
write,
?,
'IN', 'O', '(S[wq]/(S[q]/PP))/(S[wq]/(S[q]/NP))').
'WDT','O', '(S[wq]/(S[q]/NP))/N').
'NN', 'O', 'N').
'VBZ','O', '(S[q]/(S[b]NP))/NP').
'NNP','I-PER', 'N').
'VB', 'O', '(S[b]NP)/PP').
'.', 'O', 'S[wq]S[wq]').
Question Analysis
© Johan Bos
April 2008
• Tokenisation
• Part of speech tagging
• Lemmatisation
• Syntactic analysis (Parsing)
Semantic analysis (Boxing)
Architecture of PRONTO
parsing
question
answer
ccg
answer
reranking
boxing
drs
answer
selection
WordNet
NomLex
knowledge
© Johan Bos
April 2008
query
answer
typing
Indri
answer
extraction
Indexed Documents
Boxing (Semantic Analysis)
© Johan Bos
April 2008
• Providing a semantic analysis on the
basis of the syntactic analysis
• A semantic analysis of a question offers
an abstract representation of the
meaning of the question
• Boxer uses a particular semantic
theory:
Discourse Representation Theory
Discourse Representation Theory
© Johan Bos
April 2008
• Meaning of natural language expressions
represented in first-order logic
• No formulas but box representation (without
explicit quantification and conjunction)
• DRT covers a wide range
of linguistic phenomena
(Kamp & Reyle)
Output of Boxer
DRS (Discourse Representation Structure):
© Johan Bos
April 2008
_______________________
____________________________________
| x0
| | x1
|
|_______________________| |____________________________________|
(| named(x0,krugman,per) |+| write(x1)
|)
| named(x0,paul,per)
| | event(x1)
|
|
| | agent(x1,x0)
|
|_______________________| | _______________
____________ |
| | x2
|
|
| |
| |_______________|
|____________| |
| | newspaper(x2) | ? | event(x1) | |
| |_______________|
| for(x1,x2) | |
|
|____________| |
|____________________________________|
Paul Krugman. For which newspaper does Krugman write?
Focus and Topic
• Information expressed in a question
can be structured into two parts:
– the focus: information that is asked for
– the topic: information about focus
• Example:
© Johan Bos
April 2008
How many inhabitants does Rome have?
FOCUS
TOPIC
Focus in DRS
© Johan Bos
April 2008
_______________________
____________________________________
| x0
| | x1
|
|_______________________| |____________________________________|
(| named(x0,krugman,per) |+| write(x1)
|)
| named(x0,paul,per)
| | event(x1)
|
|
| | agent(x1,x0)
|
|_______________________| | _______________
____________ |
| | x2
|
|
| |
| |_______________|
|____________| |
| | newspaper(x2) | ? | event(x1) | |
| |_______________|
| for(x1,x2) | |
|
|____________| |
|____________________________________|
Focus
Question Answering (QA)
Lecture 2
© Johan Bos
April 2008
• Question Analysis
Background Knowledge
• Answer Typing
Architecture of PRONTO
parsing
question
answer
ccg
answer
reranking
boxing
drs
answer
selection
WordNet
NomLex
knowledge
© Johan Bos
April 2008
query
answer
typing
Indri
answer
extraction
Indexed Documents
Knowledge Construction
• The knowledge component in Pronto
constructs a local knowledge base for a
the question under consideration
– This knowledge is used in subsequent
components
© Johan Bos
April 2008
• The task of the knowledge component
is to find all relevant knowledge that
might be used
– As little as possible to ensure efficiency
Manually Constructed Knowledge
• Linguistic knowledge
– WordNet
– NomLex
– FrameNet
© Johan Bos
April 2008
• General knowledge
– CYC
– CIA Factbook
– Gazzetteers
WordNet
© Johan Bos
April 2008
• Electronic dictionary
• Not only words and definitions, but also
relations between words
• Four parts of speech
– Nouns
– Verbs
– Adjectives
– Adverbs
WordNet SynSets
•
•
© Johan Bos
April 2008
•
Words are organised in SynSets
A SynSet is a group of words with the
same meaning --- in other words, a
set of synonyms
Example:
{ Rome, Roma, Eternal City,
Italian Capital, capital of Italy }
Senses
•
•
A word can have several different
meanings
Example: plant
– A building for industrial labour
– A living organism lacking the power of
locomotion
© Johan Bos
April 2008
•
•
The different meanings of a word are
called senses
Therefore, one word can occur in
more than one SynSet in WordNet
© Johan Bos
April 2008
SynSet Example
-
{mug, mugful}
= the quantity that can be held in a mug
-
{chump, fool, gull, mark, patsy, fall guy,
sucker, soft touch, chump, mug}
= a person who is gullible and easy to take
advantage of
-
{countenance, physiognomy, phiz, visage,
kisser, smiler, mug}
= the human face
Hypernyms and Hyponyms
• Hyperonomy is a WordNet relation defined
among two SynSets
– If A is a hypernym of B, then A is more generic
then B
• The inverse of hyperonomy is hyponomy
– If A is a hyponym of B, then A is more specific
then B
© Johan Bos
April 2008
• Take transitive closure of these relations
• Examples:
– “cow” and “horse” are hyponyms of “animal”
– “publication” is a hypernym of “book”
Examples using WordNet
• Which rock singer wrote Lithium?
– WordNet:
singer is a hyponym of person
– Knowledge:
x(singer(x)  person(x))
© Johan Bos
April 2008
• What is the population of Andorra?
– WordNet:
population is a hyponym of number
– Knowledge:
x(population(x)  number(x))
NomLex
• NomLex is a database of
nominalisation paraphrases
– A nominalisation is a
“verb promoted to a noun”
– A paraphrase links the noun to
the root verb
© Johan Bos
April 2008
• Example:
– X is an invention by Y  Y invented X
– the killing of X
 X was killed
Harvesting Knowledge
• Often existing knowledge bases are
incomplete for particular applications
• There are various ways to automatically
construct knowledge bases:
© Johan Bos
April 2008
– Instances and Hyponyms
[e.g. Hearst]
– Paraphrases
[e.g. Lin & Pantel]
Hyponyms (X such-as Y)
TREC 20.2 (Concorde)
What airlines have Concorde in their fleets?
© Johan Bos
April 2008
• WordNet has no instances of airlines.
Hyponyms (X such as Y)
© Johan Bos
April 2008
TREC 20.2 (Concorde)
What airlines have Concorde in their fleets?
• Search for “Xs such as Y” patterns in large
corpora, such as the web
• Here:
X = airline, Y a hyponym of X
• Corpus:
…airlines such as Continental and United
now fly…
Hyponyms (X such as Y)
TREC 20.2 (Concorde)
What airlines have Concorde in their fleets?
© Johan Bos
April 2008
• Knowledge (Acquaint corpus):
Air Asia, Air Canada, Air France, Air Mandalay, Air Zimbabwe,
Alaska, Aloha, American Airlines, Angel Airlines, Ansett, Asiana,
Bangkok Airways, Belgian Carrier Sabena, British Airways,
Canadian, Cathay Pacific, China Eastern Airlines, China Xinhua
Airlines, Continental, Garuda, Japan Airlines, Korean Air, Lai, Lao
Aviation, Lufthansa, Malaysia Airlines, Maylasian Airlines, Midway,
Northwest, Orient Thai Airlines, Qantas, Seage Air, Shanghai
Airlines, Singapore Airlines, Skymark Airlines Co., South Africa,
Swiss Air, US Airways, United, Virgin, Yangon Airways
© Johan Bos
April 2008
Paraphrases
• Several methods have been developed
for automatically finding paraphrases in
large corpora
• This usually proceeds by starting with
seed patterns of known positive
instances
• Using bootstrapping new patterns are
found, and new seeds can be used
Seed example
• Start: Oswald killed JFK
• Search for "Oswald * JFK"
• Results:
– Oswald assassinated JFK
– Oswald shot JFK
© Johan Bos
April 2008
• Use these new patters to find other
pairs and start again
Paraphrase Example
TREC 4.2 (James Dean)
When did James Dean die?
Knowledge:
© Johan Bos
April 2008
xt(e(kill(e)&theme(e,x)&in(e,t))
 e(die(e)&agent(e,x)&in(e,t)))
Paraphrase Example
TREC 4.2 (James Dean)
When did James Dean die?
Knowledge:
© Johan Bos
April 2008
xt(e(kill(e)&theme(e,x)&in(e,t))
 e(die(e)&agent(e,x)&in(e,t)))
APW19990929.0165: In 1955, actor James Dean was
killed in a two-car collision near Cholame, Calif.
Question Answering (QA)
Lecture 2
© Johan Bos
April 2008
• Question Analysis
• Background Knowledge
Answer Typing
Architecture of PRONTO
parsing
question
answer
ccg
answer
reranking
boxing
drs
answer
selection
WordNet
NomLex
knowledge
© Johan Bos
April 2008
query
answer
typing
Indri
answer
extraction
Indexed Documents
Answer Typing
• Providing information on the expected
answer type
– Type of question
– Type (sortal ontology or taxonomy)
– Answer cardinality
© Johan Bos
April 2008
• Issues
– Ambiguities
– Vagueness
– Classification problems
Question Types
© Johan Bos
April 2008
• Wh-questions:
– Where was Franz Kafka born?
– How many countries are member of
OPEC?
– Who is Thom Yorke?
– Why did David Koresh ask the FBI for a
word processor?
– How did Frank Zappa die?
– Which boxer beat Muhammed Ali?
Question Types
• Yes-no questions:
– Does light have weight?
– Scotland is part of England – true or false?
© Johan Bos
April 2008
• Choice-questions:
– Did Italy or Germany win the world cup in
1982?
– Who is Harry Potter’s best friend – Ron,
Hermione or Sirius?
Indirect Questions
• Imperative mood:
– Name four European countries that
produce wine.
– Give the date of birth of Franz Kafka.
• Declarative mood:
© Johan Bos
April 2008
– I would like to know when Jim Morrison
was born.
Answer Type Taxonomies
© Johan Bos
April 2008
• Simple Answer-Type Taxonomy:
PERSON
NUMERAL
DATE
MEASURE
LOCATION
ORGANISATION
Expected Answer Types
• PERSON:
© Johan Bos
April 2008
– Who won the Nobel prize for Peace?
– Which rock singer wrote Lithium?
Expected Answer Types
• NUMERAL:
© Johan Bos
April 2008
– How many inhabitants does Rome have?
– What’s the population of Scotland?
Expected Answer Types
• DATE:
© Johan Bos
April 2008
– When was JFK killed?
– In what year did Rome become the capital
of Italy?
Expected Answer Types
• MEASURE:
© Johan Bos
April 2008
– How much does a 125 gallon fish tank
cost?
– How tall is an African elephant?
– How heavy is a Boeing 777?
Expected Answer Types
• LOCATION:
© Johan Bos
April 2008
– Where does Angus Young of AC/DC live?
– What city gives a Christmas tree to
Westminster every year as a gift?
Expected Answer Types
• ORGANISATION:
© Johan Bos
April 2008
– Which company invented the
compact disk?
– Who purchased Gilman Paper
Company?
Using background knowledge
• Which rock singer …
– singer is a hyponym of person, therefore
expected answer type is PERSON
• What is the population of …
© Johan Bos
April 2008
– population is a hyponym of number,
hence answer type NUMERAL
Answer type tagging
Simple rule-based systems:
Who …
 PERSON
Where …
 LOCATION
When …
 DATE
How many …  NUMERAL
© Johan Bos
April 2008
…often fail…
– Who launched the iPod?
– Where in the human body is the liver?
– When is it time to go to bed?
© Johan Bos
April 2008
Complex taxonomies
• Simple ontologies cannot account for
the large variety of questions
• An example of a more complex
ontology is proposed by Li & Roth
• Pronto uses its own complex ontology
• Machine learning approaches are often
used to automatically tag questions
with answer types
Taxonomy of Li & Roth (1/3)
© Johan Bos
April 2008
•
ENTITY
–
–
–
–
–
–
–
–
–
–
–
–
–
animal animals
body organs of body
color colors
creative inventions, books and other creative pieces
currency currency names
– product products
dis.med. diseases and medicine
– religion religions
event events
– sport sports
food food
– substance elements and substances
instrument musical instrument
– symbol symbols and signs
lang languages
– technique techniques and methods
letter letters like a-z
– term equivalent terms
– vehicle vehicles
other other entities
– word words with a special property
plant plants
Taxonomy of Li & Roth (2/3)
•
DESCRIPTION description and abstract concepts
–
–
–
–
•
HUMAN human beings
–
–
–
–
© Johan Bos
April 2008
•
definition definition of sth.
description description of sth.
manner manner of an action
reason reasons
group a group or organization of persons
ind an individual
title title of a person
description description of a person
LOCATION locations
–
–
–
–
–
city cities
country countries
mountain mountains
other other locations
state states
Taxonomy of Li & Roth (3/3)
•
NUMERIC numeric values
© Johan Bos
April 2008
–
–
–
–
–
–
–
–
–
–
–
–
–
•
code postcodes or other codes
count number of sth.
date dates
distance linear measures
money prices
order ranks
other other numbers
period the lasting time of sth.
percent fractions
speed speed
temp temperature
size size, area and volume
weight weight
ABBREVIATION
– abb abbreviation
– exp expansion
© Johan Bos
April 2008
Pronto Answer Type Taxonomy
© Johan Bos
April 2008
Pronto Answer Type Taxonomy
Answer typing: problems
• Ambiguities
How long  distance or duration
© Johan Bos
April 2008
• Vague Wh-words
What do pinguins eat?
What is the length of a football pitch?
• Taxonomy gaps
Which alien race featured in Star Trek?
What is the cultural capital of Italy?
Answer Cardinality
© Johan Bos
April 2008
• How many distinct answers does a
question have?
• Examples:
– When did Louis Braille die?
 1 answer
– Who won a nobel prize in chemistry?
 1 or more answers
– What are the seven wonders of the world?
 exactly 7 answers
© Johan Bos
April 2008
Class activity: answer typing
1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
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”?
Lecture 3
parsing
question
answer
ccg
answer
reranking
boxing
drs
answer
selection
WordNet
NomLex
knowledge
© Johan Bos
April 2008
query
answer
typing
Indri
answer
extraction
Indexed Documents
Question Answering (QA)
Lecture 2
© Johan Bos
April 2008
Lecture 1
•
•
•
•
•
•
•
•
What is QA?
Query Log Analysis
Challenges in QA
History of QA
System Architecture
Methods
System Evaluation
State-of-the-art
• Question Analysis
• Background Knowledge
• Answer Typing
Lecture 3
•
•
•
•
•
Query Generation
Document Analysis
Semantic Indexing
Answer Extraction
Selection and Ranking
Descargar

ICoS-4