Download - Subjectivity and Sentiment Analysis
![Page 1: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/1.jpg)
Slides by Carmen Banea based on presentations by Jan Wiebe (University of
Pittsburg) and Bing Liu (University of Illinois)
Subjectivity and Sentiment Analysis
![Page 2: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/2.jpg)
OverviewSubjectivity Analysis
DefinitionApplications
Sentiment AnalysisDefinitionApplications
Resources and Tools for Subjectivity and Sentiment ResearchLexiconsCorporaTools
Subjectivity Analysis at UNT
![Page 3: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/3.jpg)
I. Subjectivity Analysis
Definition & Applications
![Page 4: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/4.jpg)
What is subjectivity?The linguistic expression of somebody’s
opinions, sentiments, emotions, evaluations, beliefs, speculations (private states)
Private state: state that is not open to objective observation or verificationQuirk, Greenbaum, Leech, Svartvik (1985). A Comprehensive Grammar of the English Language.
Subjectivity analysis classifies content in objective or subjective
![Page 5: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/5.jpg)
ExamplesThe desire to give Broglio as many starts as
possible.The Pirates have a 9-6 record this year and
the Redbirds are 7-9.Suppose he did lie beside Lenin, would it be
permanent ?One of the obstacles to the easy control of
a 2-year old child is a lack of verbal communication.
![Page 6: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/6.jpg)
Application: Opinion Question Answering
ICWSM 20086
Q: What is the international reaction to the reelection of Robert Mugabe as President of Zimbabwe?
A: African observers generally approved of his victory while Western Governments strongly denounced it.
Opinion QA is more complex Automatic subjectivity analysis can be helpfulStoyanov, Cardie, Wiebe EMNLP05 Somasundaran, Wilson, Wiebe, Stoyanov ICWSM07
![Page 7: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/7.jpg)
Application: Information Extraction
ICWSM 20087
“The Parliament exploded into fury against the
government when word leaked out…”
Observation: subjectivity often causes false hits for IE
Goal: augment the results of IE
Subjectivity filtering strategies to improve IE Riloff, Wiebe, Phillips AAAI05
![Page 8: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/8.jpg)
More Applications
ICWSM 20088
Product review mining: What features of the ThinkPad T43 do customers like and which do they dislike?
Review classification: Is a review positive or negative toward the movie?
Tracking sentiments toward topics over time: Is anger ratcheting up or cooling down?
Prediction (election outcomes, market trends): Will Clinton or Obama win?
Expressive text-to-speech synthesis Text semantic analysis (Wiebe and Mihalcea, 2006)
(Esuli and Sebastiani, 2006)
Text summarization (Carenini et al., 2008)
![Page 9: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/9.jpg)
II. Sentiment Analysis
Definition & Applications
![Page 10: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/10.jpg)
What is sentiment analysis?Also known as opinion miningAttempts to identify the opinion/sentiment
that a person may hold towards an objectIt is a finer grain analysis compared to
subjectivity analysis
Sentiment Analysis Subjectivity analysis
PositiveSubjective
Negative
Neutral Objective
![Page 11: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/11.jpg)
Components of an opinionBasic components of an opinion:
Opinion holder: The person or organization that holds a specific opinion on a particular object.
Object: on which an opinion is expressedOpinion: a view, attitude, or appraisal on an
object from an opinion holder.
![Page 12: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/12.jpg)
Opinion mining tasksAt the document (or review) level:
Task: sentiment classification of reviewsClasses: positive, negative, and neutralAssumption: each document (or review) focuses on
a single object (not true in many discussion posts) and contains opinion from a single opinion holder.
At the sentence level:Task 1: identifying subjective/opinionated sentences
Classes: objective and subjective (opinionated)Task 2: sentiment classification of sentences
Classes: positive, negative and neutral.Assumption: a sentence contains only one opinion; not
true in many cases.Then we can also consider clauses or phrases.
![Page 13: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/13.jpg)
Opinion Mining Tasks (cont.)At the feature level:
Task 1: Identify and extract object features that have been commented on by an opinion holder (e.g., a reviewer).
Task 2: Determine whether the opinions on the features are positive, negative or neutral.
Task 3: Group feature synonyms.Produce a feature-based opinion summary of multiple
reviews.
Opinion holders: identify holders is also useful, e.g., in news articles, etc, but they are usually known in the user generated content, i.e., authors of the posts.
![Page 14: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/14.jpg)
Facts and Opinions
Two main types of textual information on the Web.Facts and Opinions
Current search engines search for facts (assume they are true)Facts can be expressed with topic keywords.
Search engines do not search for opinionsOpinions are hard to express with a few
keywordsHow do people think of Motorola Cell phones?
Current search ranking strategy is not appropriate for opinion retrieval/search.
![Page 15: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/15.jpg)
ApplicationsBusinesses and organizations:
product and service benchmarking.market intelligence.Business spends a huge amount of money to find
consumer sentiments and opinions.Consultants, surveys and focused groups, etc
Individuals: interested in other’s opinions when purchasing a product or using a service, finding opinions on political topics
Ads placements: Placing ads in the user-generated contentPlace an ad when one praises a product.Place an ad from a competitor if one criticizes a
product.Opinion retrieval/search: providing general search
for opinions.
![Page 16: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/16.jpg)
Two types of evaluationsDirect Opinions: sentiment expressions on
some objects, e.g., products, events, topics, persons.E.g., “the picture quality of this camera is
great”Subjective
Comparisons: relations expressing similarities or differences of more than one object. Usually expressing an ordering.E.g., “car x is cheaper than car y.”Objective or subjective.
![Page 17: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/17.jpg)
Opinion search (Liu, Web Data Mining book, 2007)
Can you search for opinions as conveniently as general Web search?
Whenever you need to make a decision, you may want some opinions from others,Wouldn’t it be nice? you can find them on a
search system instantly, by issuing queries such as
Opinions: “Motorola cell phones”Comparisons: “Motorola vs. Nokia”
Cannot be done yet! (but could be soon …)
![Page 18: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/18.jpg)
III. Sentiment and Subjectivity Analysis
Overview
![Page 19: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/19.jpg)
Main resources• Lexicons• General Inquirer (Stone et al., 1966)• OpinionFinder lexicon (Wiebe & Riloff,
2005)• SentiWordNet (Esuli & Sebastiani, 2006)
• Annotated corpora• Used in statistical approaches (Hu
& Liu 2004, Pang & Lee 2004)• MPQA corpus (Wiebe et. al, 2005)
• Tools • Algorithm based on minimum
cuts (Pang & Lee, 2004) • OpinionFinder (Wiebe et. al,
2005)
![Page 20: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/20.jpg)
III.1. Lexicons for Sentiment and Subjectivity Analysis
Overview
![Page 21: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/21.jpg)
Who does lexicon development ?
ICWSM 200821
Humans
Semi-automatic
Fully automatic
![Page 22: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/22.jpg)
What?
ICWSM 200822
Find relevant words, phrases, patterns that can be used to express subjectivity
Determine the polarity of subjective expressions
![Page 23: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/23.jpg)
Words
ICWSM 200823
Adjectives Hatzivassiloglou & McKeown 1997, Wiebe 2000, Kamps & Marx 2002, Andreevskaia & Bergler 2006
positive: honest important mature large patient
Ron Paul is the only honest man in Washington. Kitchell’s writing is unbelievably mature and is only
likely to get better. To humour me my patient father agrees yet again to
my choice of film
![Page 24: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/24.jpg)
Words
ICWSM 200824
Adjectivesnegative: harmful hypocritical inefficient
insecureIt was a macabre and hypocritical circus. Why are they being so inefficient ? bjective: curious,
peculiar, odd, likely, probably
![Page 25: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/25.jpg)
Words
ICWSM 200825
Adjectives Subjective (but not positive or negative
sentiment): curious, peculiar, odd, likely, probableHe spoke of Sue as his probable successor.The two species are likely to flower at different
times.
![Page 26: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/26.jpg)
Words
ICWSM 200826
Other parts of speech Turney & Littman 2003, Riloff, Wiebe & Wilson 2003, Esuli & Sebastiani 2006
Verbspositive: praise, lovenegative: blame, criticizesubjective: predict
Nounspositive: pleasure, enjoymentnegative: pain, criticismsubjective: prediction, feeling
![Page 27: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/27.jpg)
Phrases
ICWSM 200827
Phrases containing adjectives and adverbs Turney 2002, Takamura, Inui & Okumura 2007
positive: high intelligence, low costnegative: little variation, many troubles
![Page 28: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/28.jpg)
How? Patterns
ICWSM 200828
Lexico-syntactic patterns Riloff & Wiebe 2003
way with <np>: … to ever let China use force to have its way with …
expense of <np>: at the expense of the world’s security and stability
underlined <dobj>: Jiang’s subdued tone … underlined his desire to avoid disputes …
![Page 29: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/29.jpg)
How?
ICWSM 200829
How do we identify subjective items?
Assume that contexts are coherent
![Page 30: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/30.jpg)
Conjunction
ICWSM 200830
![Page 31: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/31.jpg)
Statistical association
ICWSM 200831
If words of the same orientation likely to co-occur together, then the presence of one makes the other more probable (co-occur within a window, in a particular context, etc.)
Use statistical measures of association to capture this interdependence E.g., Mutual Information (Church & Hanks 1989)
![Page 32: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/32.jpg)
How?
ICWSM 200832
How do we identify subjective items?
Assume that contexts are coherentAssume that alternatives are similarly
subjective (“plug into” subjective contexts)
![Page 33: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/33.jpg)
How? Summary
ICWSM 200833
How do we identify subjective items?
Assume that contexts are coherentAssume that alternatives are similarly
subjectiveTake advantage of specific words
![Page 34: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/34.jpg)
*We cause great leaders
ICWSM 200834
![Page 35: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/35.jpg)
III.2. Corpora for Sentiment and Subjectivity Analysis
Overview
![Page 36: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/36.jpg)
Definitions and Annotation Scheme
ICWSM 200836
Manual annotation: human markup of corpora (bodies of text)
Why? Understand the problemCreate gold standards (and training data)
Wiebe, Wilson, Cardie LRE 2005Wilson & Wiebe ACL-2005 workshopSomasundaran, Wiebe, Hoffmann, Litman ACL-2006 workshopSomasundaran, Ruppenhofer, Wiebe SIGdial 2007Wilson 2008 PhD dissertation
![Page 37: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/37.jpg)
Overview
ICWSM 200837
Fine-grained: expression-level rather than sentence or document level
Annotate Subjective expressionsmaterial attributed to a source, but
presented objectively
![Page 38: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/38.jpg)
Corpus
ICWSM 200838
MPQA: www.cs.pitt.edu/mqpa/databaserelease (version 2)
English language versions of articles from the world press (187 news sources)
Also includes contextual polarity annotations (later)
Themes of the instructions:No rules about how particular words should be annotated.
Don’t take expressions out of context and think about what they could mean, but judge them as they are used in that sentence.
![Page 39: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/39.jpg)
Gold Standards
ICWSM 200839
Derived from manually annotated dataDerived from “found” data (examples):
Blog tags Balog, Mishne, de Rijke EACL 2006
Websites for reviews, complaints, political arguments amazon.com Pang and Lee ACL 2004complaints.com Kim and Hovy ACL 2006bitterlemons.com Lin and Hauptmann ACL 2006
Word lists (example):General Inquirer Stone et al. 1996
![Page 40: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/40.jpg)
III.3. Tools for Sentiment and Subjectivity Analysis
Overview
![Page 41: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/41.jpg)
Lexicon-based toolsUse sentiment and subjectivity lexiconsRule-based classifier
A sentence is subjective if it has at least two words in the lexicon
A sentence is objective otherwise
![Page 42: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/42.jpg)
Corpus-based toolsUse corpora annotated for subjectivity
and/or sentimentTrain machine learning algorithms:
Naïve bayesDecision treesSVM …
Learn to automatically annotate new text
![Page 43: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/43.jpg)
IV. Multilingual Subjectivity Analysis
Research @ UNT
![Page 44: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/44.jpg)
Focus on Multilingual Subjectivity Research! Why?
internetworldstats.com, June 30, 2008
![Page 45: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/45.jpg)
Subjectivity Analysis on a New Language Using Parallel Texts
Bilingual Dictionary
Parallel Texts
Subjectivity analysis tool on target languageSubjectivity analysis tool on target languageTarget language
= Romanian
![Page 46: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/46.jpg)
Extract a Subjectivity Lexicon using Bootstrapping
seedsseeds query Candidate synonymsCandidate synonyms
Max. no. of iterations?
no
yes
Candidate synonymsCandidate synonyms
Selected synonymsSelected synonyms
Variable filtering
Online dictionary
Fixed filtering
![Page 47: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/47.jpg)
Subjectivity Analysis on a New Language Using Machine Translation
annotations
annotations
![Page 48: Subjectivity and Sentiment Analysis](https://reader036.vdocuments.site/reader036/viewer/2022081504/56815065550346895dbe6326/html5/thumbnails/48.jpg)
ConclusionsSubjectivity and sentiment analysis is an emerging
field in NLP with very interesting applicationsA lot can be learned from the amount of
unstructured/structured information on the web which can aid in subjectivity and sentiment analysis
Trends:Develop robust automatic systems that would
perform subjectivity/polarity annotationCarry out research in other languages and leverage
on the tools and resources already developed for English
Use subjectivity/polarity filtering in pre-processing of NLP tasks