Title: MultiPerspective Question Answering Using the OpQA Corpus
1Multi-Perspective Question Answering Using the
OpQA Corpus
Claire Cardie
Janyce Wiebe
Cornell University
University of Pittsburgh
2Multi-Perspective Question Answering
- Fact-based question answering (QA)
- When is the first day of spring?
- Do Lipton employees take coffee breaks?
- Vs Multi-perspective question answering (MPQA).
- How does the US regard the terrorist attacks in
Iraq? - Is Derek Jeter a bum?
3Talk Outline
- Properties of Opinion vs. Fact answers
- OpQA corpus
- Traditional fact-based QA systems
- Different properties of opinion questions
- Using fine-grained opinion information for MPQA
- Annotation framework and automatic classifiers
- QA experiments
4Opinion Question Answer (OpQA) Corpus
Stoyanov, Cardie, Litman, and Wiebe 2004
- 98 documents manually tagged for opinions (from
the NRRC MPQA corpus Wilson and Wiebe 2003) - 30 questions
- 15 fact
- 15 opinion
5OpQA corpus Answer Annotations
- Two annotators
- Include every text segment contributing to an
answer - Partial answers
- When was the Kyoto protocol ratified?
- before May 2003
- Are the Japanese unanimous in their support of
Koizumi? - most Japanese support their prime minister
- Minimum spans
6Traditional Fact-based QA systems
Documents (document fragments)
Linguistic filters
Syntactic filters Semantic filters
Questions
7Characteristics of Opinion vs. Fact Answers
- Answer length
- Syntactic and semantic class
- Additional processing difficulties
- Partial answers
- Answer generator
8Fine-grained Opinion Information for MPQA
- Recent interest in the area of automatic opinion
information extraction. - E.g. Bethard, Yu, Thornton, Hatzivassiloglou,
and Jurafsky 2004, Pang and Lee 2004, Riloff
and Wiebe 2003, Wiebe and Riloff 2005,
Wilson, Wiebe, and Hwa 2004, Yu and
Hatzivassiloglou 2003 - In our evaluation
- Opinion annotation framework
- Sentence-level automatic opinion classifiers
- Subjectivity filters
- Source filter
9Opinion Annotation Framework
- Described in Wiebe, Wilson, and Cardie 2002
- Accounts for both
- Explicitly stated opinions
- Joe believes that Sue dislikes the Red Sox.
- Indirectly expressed opinions
- The aim of the report is to tarnish Chinas
image. - Attributes include strength and source.
- Manual sentence-level classification
- sentence subjective if it contains one or more
opinions of strength gt medium
- Described in Wiebe, Wilson, and Cardie 2002
- Accounts for both
- Explicitly stated opinions
- Joe believes that Sue dislikes the Red Sox.
- Indirectly expressed opinions
- The aim of the report is to tarnish Chinas
image. - Attributes include strength and source.
- Manual sentence-level classification
- sentence subjective if it contains one or more
opinions of strength gt medium
10Automatic Opinion Classifiers
- Two sentence-level opinion classifiers from Wiebe
and Riloff 2005 used for evaluation - Both classifiers use unannotated data
- Rulebased Extraction patterns bootstrapped using
word lists - NaiveBayes Trained on data obtained from
Rulebased
11Subjectivity Filters
Document Sentences
Subjectivity filters
Manual Rulebased NaiveBayes
Opinion Questions
12Subjectivity Filters Contd
- Look for the rank of the first guess containing
an answer - Compute
- Mean Reciprocal Rank (MRR) across the top 5
answers - MRR meanall_questions(1/Rank_of_first_answer)
- Mean Rank of the First Answer
- MRFA meanall_questions(Rank_of_first_answer)
13Subjectivity Filters Results
14Source Filter
- Manually identify the sources in the opinion
questions - Does France approve of the war in Iraq?
- Retains only sentences that contain opinions with
sources matching sources in the question - France has voiced some concerns with the
situation.
15Source Filter Results
- Performs well on the hardest questions in the
corpus - All questions answered within the first 25
sentences with one exception.
16Summary
- Properties of opinion vs. fact answers
- Traditional architectures unlikely to be
effective - Use of fine-grained opinion information for MPQA
- MPQA can benefit from fine-grained perspective
information
17Future Work
- Create summaries of all opinions in a document
using fine-grained opinion information - Methods used will be directly applicable to MPQA
18Thank you.Questions?
19- Did something surprising happen when Chavez
regained power in Venezuela after he was removed
by a coup? - What did South Africa want Mugabe to do after the
2002 elections? - Whats Mugabes opinion about the Wests attitude
and actions towards the 2002 Zimbabwe election?
20Characteristics of Fact vs. Opinion Answers Contd
- Syntactic Constituent of the answers
21- All improvement significant using Wilcoxon
Matched-Pairs Signed-Ranks Test (plt0.05) except
for source filter (p0.81)