SENTIMENT ANALYSIS

Mausam

(With slides from Jan Wiebe, Kavita Ganesan, Heng Ji, Dan Jurafsky, Chris Manning)
Motivation
“What people think?”
What others think has always been an important piece of information

“Which car should I buy?”

“Which schools should I

apply to?”

“Which Professor to work for?”

“Whom should I vote for?”

“So whom shall I ask?”
Pre Web
• Friends and relatives
• Acquaintances
• Consumer Reports

Post Web
“…I don’t know who..but apparently it’s a good phone. It has good battery life and…”
• Blogs (google blogs, livejournal)
• E-commerce sites (amazon, ebay)
• Review sites (CNET, PC Magazine)
• Discussion forums ([Link],
[Link])
• Friends and Relatives (occasionally)
 “Whoala! I have the reviews I need”
Now that I have “too much” information on one
topic…I could easily form my opinion and make
decisions…

Is this true?
…Not Quite
Searching for reviews may be difficult
Can you search for opinions as conveniently
as general Web search?
eg: is it easy to search for “iPhone vs Google Phone”?
“Let me look at reviews on one site only…”

Problems?
• Biased views
• all reviewers on one site may have the same opinion

• Fake reviews/Spam (sites like YellowPages, CitySearch are prone to this)

• people post good reviews about their own product OR services
• some posts are plain spams
 Coincidence or Fake?
Reviews for a moving
company from YellowPages

• # of merchants
reviewed by the each of
these reviewers  1

• Review dates close

to one another

• All rated 5 star

• Reviewers seem to know

exact names of people
working in the company and
TOO many positive mentions
Problem Names
Subjectivity Analysis

Review Mining Appraisal Extraction

Sentiment Analysis
Opinion Mining

Synonymous
&
Interchangeably Used!
So, what is Subjectivity?
• The linguistic expression of somebody’s opinions, sentiments,
emotions…..(private states)

• private state: state that is not open to objective verification (Quirk,

Greenbaum, Leech, Svartvik (1985). A Comprehensive Grammar of the English Language.)

• Subjectivity analysis - is the computational study of affect,

opinions, and sentiments expressed in text
• blogs
• editorials
• reviews (of products, movies, books, etc.)
• newspaper articles
Example: iPhone review
Review posted on a tech blog
InfoWorld
-summary is structured
-everything else is plain text
-mixture of objective and
subjective information
-no separation between
positives and negatives
CNET
-nice structure
-positives and negatives
separated

Tech BLOG
-everything is plain text
Review on InfoWorld - -no separation between
tech news site CNET review positives and negatives
Example: iPhone review
Review posted on a tech blog

Review on InfoWorld -
tech news site CNET review
Subjectivity Analysis on iPhone Reviews

Individual’s Perspective
• Highlight of what is good and bad about iPhone
• Ex. Tech blog may contain mixture of information
• Combination of good and bad from the different sites (tech
blog, InfoWorld and CNET)
• Complementing information
• Contrasting opinions
Ex.
CNET: The iPhone lacks some basic features
Tech Blog: The iPhone has a complete set of features
Subjectivity Analysis on iPhone Reviews

Business’ Perspective
• Apple: What do consumers think about iPhone?
• Do they like it?
• What do they dislike?
• What are the major complaints?
• What features should we add?

• Apple’s competitor:
• What are iPhone’s weaknesses?
• How can we compete with them?
• Do people like everything about it?
Known as Business
Intelligence
 15

Google Product Search

•a
 16

Bing Shopping

•a
Twitter sentiment versus Gallup Poll of
Consumer Confidence
Brendan O'Connor, Ramnath Balasubramanyan, Bryan R. Routledge, and Noah A. Smith. 2010. From
Tweets to Polls: Linking Text Sentiment to Public Opinion Time Series. In ICWSM-2010
 18

Twitter sentiment:
Johan Bollen, Huina Mao, Xiaojun Zeng.
2011. Twitter mood predicts the stock
market,
Journal of Computational Science 2:1, 1-
8. 10.1016/[Link].2010.12.007.
 19

Bollen et al. (2011)

• CALM predicts
Dow Jones

DJIA 3 days
later
• At least one
current hedge
fund uses this
CALM

algorithm
 20

Target Sentiment on Twitter

• Twitter Sentiment App

• Alec Go, Richa Bhayani, Lei Huang.
2009. Twitter Sentiment Classification
using Distant Supervision
Application Areas Summarized
• Businesses and organizations: interested in opinions
• product and service benchmarking
• market intelligence
• survey on a topic
• Individuals: interested in other’s opinions when
• Purchasing a product
• Using a service
• Tracking political topics
• Other decision making tasks
• Ads placements: Placing ads in user-generated content
• Place an ad when one praises an product
• Place an ad from a competitor if one criticizes a product
• Opinion search: providing general search for opinions
• Text-driven forecasting: insights about other areas from text
Definition
 Scherer Typology of Affective States

• Emotion: brief organically synchronized … evaluation of a major event

• angry, sad, joyful, fearful, ashamed, proud, elated
• Mood: diffuse non-caused low-intensity long-duration change in subjective feeling
• cheerful, gloomy, irritable, listless, depressed, buoyant
• Interpersonal stances: affective stance toward another person in a specific
interaction
• friendly, flirtatious, distant, cold, warm, supportive, contemptuous
• Attitudes: enduring, affectively colored beliefs, dispositions towards objects or
persons
• liking, loving, hating, valuing, desiring
• Personality traits: stable personality dispositions and typical behavior tendencies
• nervous, anxious, reckless, morose, hostile, jealous
Scherer Typology of Affective States
 Scherer Typology of Affective States

• Emotion: brief organically synchronized … evaluation of a major event

• angry, sad, joyful, fearful, ashamed, proud, elated
• Mood: diffuse non-caused low-intensity long-duration change in subjective feeling
• cheerful, gloomy, irritable, listless, depressed, buoyant
• Interpersonal stances: affective stance toward another person in a specific
interaction
• friendly, flirtatious, distant, cold, warm, supportive, contemptuous
• Attitudes: enduring, affectively colored beliefs, dispositions towards objects
or persons
• liking, loving, hating, valuing, desiring
• Personality traits: stable personality dispositions and typical behavior tendencies
• nervous, anxious, reckless, morose, hostile, jealous
 Sentiment Analysis
• Sentiment analysis is the detection of attitudes
“enduring, affectively colored beliefs, dispositions towards objects or
persons”
1. Holder (source) of attitude
2. Target (aspect) of attitude
3. Type of attitude
• From a set of types
• Like, love, hate, value, desire, etc.
• Or (more commonly) simple weighted polarity:
• positive, negative, neutral, together with strength
4. Text containing the attitude
27 • Sentence or entire document
Sentiment Analysis
• Simplest task:
• Is the attitude of this text positive or negative?
• More complex:
• Rank the attitude of this text from 1 to 5
• Advanced:
• Detect the target, source, or complex attitude
types
Sentiment Analysis
• Simplest task:
• Is the attitude of this text positive or negative?
• More complex:
• Rank the attitude of this text from 1 to 5
• Advanced:
• Detect the target, source, or complex attitude
types
Baseline Algorithms
Sentiment Classification in Movie Reviews
Bo Pang, Lillian Lee, and Shivakumar Vaithyanathan. 2002. Thumbs up?
Sentiment Classification using Machine Learning Techniques. EMNLP-2002, 79—
86.
Bo Pang and Lillian Lee. 2004. A Sentimental Education: Sentiment Analysis
Using Subjectivity Summarization Based on Minimum Cuts. ACL, 271-278

• Polarity detection:
• Is an IMDB movie review positive or negative?
• Data: Polarity Data 2.0:
• [Link]
 IMDB data in the Pang and Lee database

✓ ✗
when _star wars_ came out some twenty years “ snake eyes ” is the most
ago , the image of traveling throughout the aggravating kind of movie : the kind
stars has become a commonplace image . […] that shows so much potential then
when han solo goes light speed , the stars becomes unbelievably disappointing .
change to bright lines , going towards the it’s not just because this is a brian
viewer in lines that converge at an invisible depalma film , and since he’s a great
point . director and one who’s films are
cool . always greeted with at least some
_october sky_ offers a much simpler image– fanfare .
that of a single white dot , traveling horizontally and it’s not even because this was a
across the night sky . [. . . ] film starring nicolas cage and since
he gives a brauvara performance ,
this film is hardly worth his talents .
Baseline Algorithm (adapted from Pang
and Lee)

• Tokenization
• Feature Extraction
• Classification using different classifiers
• Naïve Bayes
• MaxEnt
• SVM
 35

Sentiment Tokenization Issues

• Deal with HTML and XML markup

• Twitter mark-up (names, hash tags)
• Capitalization (preserve for
words in all caps)
• Phone numbers, dates
• Emoticons
 36

Extracting Features for Sentiment

Classification

• How to handle negation

• I didn’t like this movie

vs
• I really like this movie

• Which words to use?

• Only adjectives
• All words
• All words turns out to work better, at least on this data
 Negation
Das, Sanjiv and Mike Chen. 2001. Yahoo! for Amazon: Extracting market sentiment from
stock message boards. In Proceedings of the Asia Pacific Finance Association Annual
Conference (APFA).
Bo Pang, Lillian Lee, and Shivakumar Vaithyanathan. 2002. Thumbs up? Sentiment Classification using
Machine Learning Techniques. EMNLP-2002, 79—86.

Add NOT_ to every word between negation and following

punctuation:

didn’t like this movie , but I

didn’t NOT_like NOT_this NOT_movie but I

38
38

Accounting for Negation

• Let us consider the following positive sentence:
• Example: Luckily, the smelly poo did not leave awfully nasty
stains on my favorite shoes!
• Rest of Sentence (RoS):
• Following: Luckily, the smelly poo did not leave awfully nasty
stains on my favorite shoes!
• Around: Luckily, the smelly poo did not leave awfully nasty
stains on my favorite shoes!
• First Sentiment-Carrying Word (FSW):
• Following: Luckily, the smelly poo did not leave awfully nasty
stains on my favorite shoes!
• Around: Luckily, the smelly poo did not leave awfully nasty
stains on my favorite shoes!

Determining Negation Scope and Strength in Sentiment Analysis, Hogenboom et al SMC 2011.
39
39

Accounting for Negation

• Let us consider the following positive sentence:
• Example: Luckily, the smelly poo did not leave awfully nasty
stains on my favorite shoes!
• Next Non-Adverb (NNA):
• Following: Luckily, the smelly poo did not leave awfully nasty
stains on my favorite shoes!
• Fixed Window Length (FWL):
• Following (3): Luckily, the smelly poo did not leave awfully
nasty stains on my favorite shoes!
• Around (3): Luckily, the smelly poo did not leave awfully nasty
stains on my favorite shoes!

SMC 2011
40

KEYWORDS SELECTION FROM TEXT

• Pang et. al. (2002)
• Binary Classification of unigrams
• Positive
• Negative

• Unigram method reached 80% accuracy.

N-GRAM BASED CLASSIFICATION

• Learn N-Grams (frequencies) from pre-annotated training
data.
40
• Use this model to classify new incoming sample.
41

PART-OF-SPEECH BASED PATTERNS

• Extract POS patterns from training data.
• Usually used for subjective vs objective classification.
• Adjectives and Adverbs contain sentiments
• Example patterns
• *-JJ-NN : trigram pattern
• JJ-NNP : bigram pattern
• *-JJ : bigram pattern

41
 42

Reminder: Naïve Bayes

cNB = argmax P(c j )

c j ÎC
Õ P(wi | c j )
iÎ positions

count(w, c) +1
P̂(w | c) =
count(c) + V
 43

Binarized (Boolean feature) Multinomial Naïve Bayes

• Intuition:
• For sentiment (and probably for other text classification domains)
• Word occurrence may matter more than word frequency
• The occurrence of the word fantastic tells us a lot
• The fact that it occurs 5 times may not tell us much more.
• Boolean Multinomial Naïve Bayes
• Clips all the word counts in each document at 1
 Boolean Multinomial Naïve Bayes:
Learning
• From training corpus, extract Vocabulary
• Calculate P(cj) terms • Calculate P(wk | cj) terms
j  single doc containing all docsj
• For each cj in C do • Text
Remove duplicates in each doc:
docsj  all docs with class =cj • For
• For eacheach word
word wktype w in docj
in Vocabulary
n• Retain
# of only a single instance
occurrences of w inofText
w
| docs j | k k j
P(c j ) ¬ nk + a
| total # documents| P(wk | c j ) ¬
n + a | Vocabulary |
 45

Boolean Multinomial Naïve Bayes

on a test document d
• First remove all duplicate words from d
• Then compute NB using the same equation:

cNB = argmax P(c j )

c j ÎC
Õ P(wi | c j )
iÎ positions
 46

Other issues in Classification

• MaxEnt and SVM tend to do better than Naïve Bayes
Problems:
What makes reviews hard to classify?
• Subtlety:
• Perfume review in Perfumes: the Guide:
• “If you are reading this because it is your darling
fragrance, please wear it at home exclusively, and tape
the windows shut.”
• Dorothy Parker on Katherine Hepburn
• “She runs the gamut of emotions from A to B”

48
49

CHALLENGES
• Ambiguous words
• This music cd is literal waste of time.
(negative)
• Please throw your waste material here.
(neutral)
• Sarcasm detection and handling
• “All the features you want - too bad they don’t
work. :-P”
• (Almost) No resources and tools for low/scarce resource
languages like Indian languages. 49
51

User written: grammar, spellings…

Hi,
I have Haier phone.. It was good when i was buing this phone.. But I
invented A lot of bad features by this phone those are It’s cost is low
but Software is not good and Battery is very bad..,,Ther are no signals
at out side of the city..,, People can’t understand this type of
software..,, There aren’t features in this phone, Design is better not
good..,, Sound also bad..So I’m not intrest this [Link] are giving
heare phones it is good. They are giving more talktime and validity
these are also [Link] are givingLack of punctuation
colour marks,time it is
screen at display
also good because other phones aren’t Grammatical
this type of [Link]
errors is also
low wait.

Wait.. err.. Come again

From: [Link]
52

Alternating Sentiment
I suggest that instead of fillings songs in tunes you should
fill tunes (not made of songs) only. The phone has good
popularity in old age people. Third i had tried much for its
data cable but i find it nowhere. It should be supplied with
set with some extra cost.
Good features of this phone are its cheapest price and
durability . It should have some features more than nokia
1200. it is easily available in market and repair is also
available

From: [Link]
53

Subject Centrality
• I have this personal experience of using this cell phone. I bought it one and half years back. It had
modern features that a normal cell phone has, and the look is excellent. I was very impressed by the
design. I bought it for Rs. 8000. It was a gift for someone. It worked fine for first one month, and then
started the series of multiple faults it has. First the speaker didnt work, I took it to the service centre
(which is like a govt. office with no work). It took 15 days to repair the handset, moreover they
charged me Rs. 500. Then after 15 days again the mike didnt work, then again same set of time
was consumed for the repairs and it continued. Later the camera didnt work, the speakes were
rubbish, it used to hang. It started restarting automatically. And the govt. office had staff which I
doubt have any knoledge of cell phones??
These multiple faults continued for as long as one year, when the warranty period ended. In this
period of time I spent a considerable amount on the petrol, a lot of time (as the service centre is a
govt. office). And at last the phone is still working, but now it works as a paper weight. The company
who produces such items must be sacked. I understand that it might be fault with one prticular
handset, but the company itself never bothered for replacement and I have never seen such
miserable cust service. For a comman man like me, Rs. 8000 is a big amount. And I spent almost
the same amount to get it work, if any has a good suggestion and can gude me how to sue such
companies, please guide.
For this the quality team is faulty, the cust service is really miserable and the worst condition of
any organisation I have ever seen is with the service centre for Fly and Sony Erricson, (it’s near
Sancheti hospital, Pune). I dont have any thing else to say.

From: [Link]
 54

Thwarted Expectations
and Ordering Effects

• “This film should be brilliant. It sounds like a great

plot, the actors are first grade, and the supporting
cast is good as well, and Stallone is attempting to
deliver a good performance. However, it can’t hold
up.”
• Well as usual Keanu Reeves is nothing special, but
surprisingly, the very talented Laurence Fishbourne
is not so good either, I was surprised.
 55

Thwarted Expectations
and Ordering Effects

• “This film should be brilliant. It sounds like a great

plot, the actors are first grade, and the supporting
cast is good as well, and Stallone is attempting to
deliver a good performance. However, it can’t hold
up.”
• Well as usual Keanu Reeves is nothing special, but
surprisingly, the very talented Laurence Fishbourne
is not so good either, I was surprised.
Sentiment Lexicons
 The General Inquirer
Philip J. Stone, Dexter C Dunphy, Marshall S. Smith, Daniel M. Ogilvie. 1966. The
General Inquirer: A Computer Approach to Content Analysis. MIT Press

• Home page: [Link]

• List of Categories:
[Link]
• Spreadsheet: [Link]
• Categories:
• Positiv (1915 words) and Negativ (2291 words)
• Strong vs Weak, Active vs Passive, Overstated versus Understated
• Pleasure, Pain, Virtue, Vice, Motivation, Cognitive Orientation, etc
• Free for Research Use
 LIWC (Linguistic Inquiry and Word Count)
Pennebaker, J.W., Booth, R.J., & Francis, M.E. (2007). Linguistic Inquiry and Word
Count: LIWC 2007. Austin, TX

• Home page: [Link]

• 2300 words, >70 classes
• Affective Processes
• negative emotion (bad, weird, hate, problem, tough)
• positive emotion (love, nice, sweet)
• Cognitive Processes
• Tentative (maybe, perhaps, guess), Inhibition (block, constraint)
• Pronouns, Negation (no, never), Quantifiers (few, many)
• $30 or $90 fee
 59

MPQA Subjectivity Cues Lexicon

Theresa Wilson, Janyce Wiebe, and Paul Hoffmann (2005). Recognizing Contextual Polarity in
Phrase-Level Sentiment Analysis. Proc. of HLT-EMNLP-2005.

Riloff and Wiebe (2003). Learning extraction patterns for subjective expressions. EMNLP-2003.

• Home page: [Link]

• 6885 words
• 2718 positive
• 4912 negative
• Each word annotated for intensity (strong, weak)
• GNU GPL
 60

Bing Liu Opinion Lexicon

Minqing Hu and Bing Liu. Mining and Summarizing Customer Reviews. ACM
SIGKDD-2004.

• Bing Liu's Page on Opinion Mining

• [Link]

• 6786 words
• 2006 positive
• 4783 negative
 SentiWordNet
Stefano Baccianella, Andrea Esuli, and Fabrizio Sebastiani. 2010
SENTIWORDNET 3.0: An Enhanced Lexical Resource for Sentiment Analysis
and Opinion Mining. LREC-2010

• Home page: [Link]

• All WordNet synsets automatically annotated for degrees of
positivity, negativity, and neutrality/objectiveness
• [estimable(J,3)] “may be computed or estimated”
Pos 0 Neg 0 Obj 1
• [estimable(J,1)] “deserving of respect or high regard”
Pos .75 Neg 0 Obj .25
62

ADVANTAGES AND DISADVANTAGES

• Advantages
• Fast
• No Training data necessary
• Good initial accuracy
• Disadvantages
• Does not deal with multiple word senses
• Does not work for multiple word phrases

62
 63

Disagreements between polarity lexicons

Christopher Potts, Sentiment Tutorial, 2011

Opinion General SentiWordNet LIWC

Lexicon Inquirer
MPQA 33/5402 49/2867 (2%) 1127/4214 (27%) 12/363 (3%)
(0.6%)
Opinion 32/2411 (1%) 1004/3994 (25%) 9/403 (2%)
Lexicon
General 520/2306 (23%) 1/204 (0.5%)
Inquirer
SentiWordNet 174/694
(25%)
LIWC
Analyzing the polarity of each word in IMDB
Potts, Christopher. 2011. On the negativity of negation. SALT 20, 636-659.

• How likely is each word to appear in each sentiment class?

• Count(“bad”) in 1-star, 2-star, 3-star, etc.
• But can’t use raw counts:
• Instead, likelihood:
f (w, c)
P(w | c) =
åwÎc f (w, c)
• Make them comparable between words
• Scaled likelihood:
P(w | c)
P(w)
 Analyzing the polarity of each word in IMDB
Potts, Christopher. 2011. On the negativity of negation. SALT 20, 636-659.

POS good (883,417 tokens) amazing (103,509 tokens) great (648,110 tokens) awesome (47,142 tokens)

0.28 ●
0.27 ●
Scaled likelihood

Pr(c|w)

●
●
0.17 0.17 ●
0.16
●

●
P(w|c)/P(w)

● ●
0.12 ●
●
●
●
0.11 ●
0.1 ●
●
●
● ● ●
● ●
0.08 ●
● ●
●
●
●
●
● ● ●
●
0.05 ● ● ● ●
0.05 0.05 ● ● ●

1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

Rating
NEG good (20,447 tokens) depress(ed/ing) (18,498 tokens) bad (368,273 tokens) terrible (55,492 tokens)
●
0.28
Scaled likelihood

0.21 ● ●
Pr(c|w)

●
0.16 ● ● ● ● 0.16
P(w|c)/P(w)

●
●
● 0.13 ● ●
0.12
0.1 0.11 ●
●
●
● ●
● ● ●
● ● ● ●
0.08 ●

●
●

● ●
●

0.03
●
● ● 0.04 ● ●
0.03
●
● ● ●

1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10 1 2 3 4 5 6 7 8 9 10

Rating
 Other sentiment feature: Logical negation
Potts, Christopher. 2011. On the negativity of negation. SALT 20, 636-659.

• Is logical negation (no, not) associated with

negative sentiment?
• Potts experiment:
• Count negation (not, n’t, no, never) in online reviews
• Regress against the review rating
 Potts 2011 Results:
More negation in negative sentiment

a
Scaled likelihood
P(w|c)/P(w)
 68

Semi-supervised learning of lexicons

• Use a small amount of information
• A few labeled examples
• A few hand-built patterns
• To bootstrap a lexicon
 69

Using WordNet to learn polarity

S.M. Kim and E. Hovy. 2004. Determining the sentiment of opinions. COLING 2004
M. Hu and B. Liu. Mining and summarizing customer reviews. In Proceedings of KDD, 2004

• WordNet: online thesaurus (covered in later lecture).

• Create positive (“good”) and negative seed-words (“terrible”)
• Find Synonyms and Antonyms
• Positive Set: Add synonyms of positive words (“well”) and antonyms
of negative words
• Negative Set: Add synonyms of negative words (“awful”) and
antonyms of positive words (”evil”)
• Repeat, following chains of synonyms
• Filter
70

WordNet
71

WordNet
72

WordNet relations
73

WordNet relations
74

WordNet relations
75

WordNet glosses
 93

Semi-supervised learning of lexicons

• Use a small amount of information
• A few labeled examples
• A few hand-built patterns
• To bootstrap a lexicon
 Hatzivassiloglou and McKeown intuition
for identifying word polarity
Vasileios Hatzivassiloglou and Kathleen R. McKeown. 1997. Predicting
the Semantic Orientation of Adjectives. ACL, 174–181

• Adjectives conjoined by “and” have same polarity

• Fair and legitimate, corrupt and brutal
• *fair and brutal, *corrupt and legitimate
• Adjectives conjoined by “but” do not
• fair but brutal

94
 95

Hatzivassiloglou & McKeown 1997

Step 1
• Label seed set of 1336 adjectives (all >20 in 21 million
word WSJ corpus)
• 657 positive
• adequate central clever famous intelligent remarkable
reputed sensitive slender thriving…
• 679 negative
• contagious drunken ignorant lanky listless primitive
strident troublesome unresolved unsuspecting…
 96

Hatzivassiloglou & McKeown 1997

Step 2

• Expand seed set to conjoined adjectives

nice, helpful

nice, classy
 97

Hatzivassiloglou & McKeown 1997 Step 3

3. A supervised learning algorithm builds a graph of
adjectives linked by the same or different semantic
orientation

scenic nice

terrible
painful
handsome

fun
expensive

comfortable
 98

Hatzivassiloglou & McKeown 1997 Step 4

4. A clustering algorithm partitions the adjectives into two
subsets

+
slow
scenic
nice
terrible
handsome painful

fun
expensive

comfortable
 101

Output polarity lexicon

• Positive
• bold decisive disturbing generous good honest important large
mature patient peaceful positive proud sound stimulating
straightforward strange talented vigorous witty…
• Negative
• ambiguous cautious cynical evasive harmful hypocritical inefficient
insecure irrational irresponsible minor outspoken pleasant reckless
risky selfish tedious unsupported vulnerable wasteful…
 102

Output polarity lexicon

• Positive
• bold decisive disturbing generous good honest important large
mature patient peaceful positive proud sound stimulating
straightforward strange talented vigorous witty…
• Negative
• ambiguous cautious cynical evasive harmful hypocritical inefficient
insecure irrational irresponsible minor outspoken pleasant
reckless risky selfish tedious unsupported vulnerable wasteful…
 103

Turney Algorithm
Turney (2002): Thumbs Up or Thumbs Down? Semantic Orientation Applied to Unsupervised
Classification of Reviews

1. Extract a phrasal lexicon from reviews

2. Learn polarity of each phrase
3. Rate a review by the average polarity of its phrases
 104

Extract two-word phrases with adjectives

First Word Second Word Third Word (not

extracted)
JJ NN or NNS anything
RB, RBR, RBS JJ Not NN nor NNS
JJ JJ Not NN or NNS
NN or NNS JJ Nor NN nor NNS
RB, RBR, or RBS VB, VBD, VBN, anything
VBG
 105

How to measure polarity of a phrase?

• Positive phrases co-occur more with “excellent”
• Negative phrases co-occur more with “poor”
• But how to measure co-occurrence?
 Pointwise Mutual Information
• Mutual information between 2 random variables X
P(x,y)
I(X,Y ) = åå P(x, y) log2
and Y
P(x)P(y)
x y

• Pointwise mutual information:

• How much more do events x and y co-occur than if they were independent?
P ( x, y )
PMI( x, y )  log 2
P( x) P( y )
 Pointwise Mutual Information
• Pointwise mutual information:
• How much more do events x and y co-occur than if they were independent?

P ( x, y )
PMI( x, y )  log 2
P( x) P( y )
• PMI between two words:
• How much more do two words co-occur than if they were independent?

P(word1,word2 )
PMI(word1, word2 ) = log2
P(word1)P(word2 )
 How to Estimate Pointwise Mutual Information
• Query search engine (Altavista)
• P(word) estimated by hits(word)/N
• P(word1,word2) by hits(word1 NEAR word2)/N
• (More correctly the bigram denominator should be kN, because there are a
total of N consecutive bigrams (word1,word2), but kN bigrams that are k
words apart, but we just use N on the rest of this slide and the next.)
1
hits(word1 NEAR word2 )
PMI(word1, word2 ) = log N
2 1
N
hits(word1) N1 hits(word2 )
 109

Does phrase appear more with “poor” or “excellent”?

Polarity(phrase) = PMI(phrase,"excellent")- PMI(phrase,"poor")

1
hits(phrase NEAR "excellent") 1
hits(phrase NEAR "poor")
= log N
2 1 1
- log N
2 1
N
hits(phrase) hits("excellent")
N N
hits(phrase) N1 hits("poor")

hits(phrase NEAR "excellent") hits(phrase)hits("poor")

= log2
hits(phrase)hits("excellent") hits(phrase NEAR "poor")
æ hits(phrase NEAR "excellent")hits("poor") ö
= log2 ç ÷
è hits(phrase NEAR "poor")hits("excellent") ø
 110

Phrases from a thumbs-up review

Phrase POS Polarity
tags
online service JJ NN 2.8
online experience JJ NN 2.3
direct deposit JJ NN 1.3
local branch JJ NN 0.42
…

low fees JJ NNS 0.33

true service JJ NN -0.73
other bank JJ NN -0.85
inconveniently located JJ NN -1.5
Average 0.32
 111

Phrases from a thumbs-down review

Phrase POS Polarity
tags
direct deposits JJ NNS 5.8
online web JJ NN 1.9
very handy RB JJ 1.4
…

virtual monopoly JJ NN -2.0

lesser evil RBR JJ -2.3
other problems JJ NNS -2.8
low funds JJ NNS -6.8
unethical practices JJ NNS -8.5
Average -1.2
 112

Results of Turney algorithm

• 410 reviews from Epinions
• 170 (41%) negative
• 240 (59%) positive
• Majority class baseline: 59%
• Turney algorithm: 74%

• Phrases rather than words

• Learns domain-specific information
 Summary on Learning Lexicons

• Advantages:
• Can be domain-specific
• Can be more robust (more words)

• Intuition
• Start with a seed set of words (‘good’, ‘poor’)
• Find other words that have similar polarity:
• Using “and” and “but”
• Using words that occur nearby in the same document
• Using WordNet synonyms and antonyms

• Use seeds and semi-supervised learning to induce lexicons

 114

PMI based Sentiment Mining Algorithm

•Synonymous words have high Web-PMI with
each other
great +
intuitive poor - camera
excellent +
unknown terrible - context
adjective ...
Web-PMI known-polarity
adjectives
WebPMI(adj, great) = HITS(“camera” near adj, great)
HITS(“camera” NEAR adj) x HITS(“camera” NEAR great)

+/-
training

(:) words

WebPMI feature vector

classifier +/-
F1 Scores: 0.78(+) 0.76(-
 167

Finding sentiment of a sentence

• Important for finding aspects or attributes

• Target of sentiment

• The food was great but the service was awful

 Finding aspect/attribute/target of sentiment
M. Hu and B. Liu. 2004. Mining and summarizing customer reviews. In Proceedings of KDD.
S. Blair-Goldensohn, K. Hannan, R. McDonald, T. Neylon, G. Reis, and J. Reynar. 2008. Building a
Sentiment Summarizer for Local Service Reviews. WWW Workshop.

• Frequent phrases + rules

• Find all highly frequent phrases across reviews (“fish tacos”)
• Filter by rules like “occurs right after sentiment word”
• “…great fish tacos” means fish tacos a likely aspect

Casino casino, buffet, pool, resort, beds

Children’s Barber haircut, job, experience, kids
Greek Restaurant food, wine, service, appetizer, lamb
Department Store selection, department, sales, shop, clothing
 169

Finding aspect/attribute/target of sentiment

• The aspect name may not be in the sentence
• For restaurants/hotels, aspects are well-understood
• Supervised classification
• Hand-label a small corpus of restaurant review sentences with
aspect
• food, décor, service, value, NONE
• Train a classifier to assign an aspect to a sentence
• “Given this sentence, is the aspect food, décor, service, value, or
NONE”
 170

Putting it all together:

Finding sentiment for aspects
S. Blair-Goldensohn, K. Hannan, R. McDonald, T. Neylon, G. Reis, and J. Reynar. 2008. Building a
Sentiment Summarizer for Local Service Reviews. WWW Workshop

Sentences Sentences Sentences

& Phrases & Phrases & Phrases
Final
Summary
Reviews

Text Sentiment Aspect

Aggregator
Extractor Classifier Extractor
 Results of Blair-Goldensohn et al. method

Rooms (3/5 stars, 41 comments)

(+) The room was clean and everything worked fine – even the water pressure ...
(+) We went because of the free room and was pleasantly pleased ...
(-) …the worst hotel I had ever stayed at ...
Service (3/5 stars, 31 comments)
(+) Upon checking out another couple was checking early due to a problem ...
(+) Every single hotel staff member treated us great and answered every ...
(-) The food is cold and the service gives new meaning to SLOW.
Dining (3/5 stars, 18 comments)
(+) our favorite place to stay in [Link] food is great also the service ...
(+) Offer of free buffet for joining the Play
 173

How to deal with 7 stars?

Bo Pang and Lillian Lee. 2005. Seeing stars: Exploiting class relationships for sentiment
categorization with respect to rating scales. ACL, 115–124

1. Map to binary
2. Use linear or ordinal regression
• Or specialized models like metric labeling
 Summary on Sentiment

• Generally modeled as classification or regression task

• predict a binary or ordinal label
• Features:
• Negation is important
• Using all words (in naïve bayes) works well for some tasks
• Finding subsets of words may help in other tasks
• Hand-built polarity lexicons
• Use seeds and semi-supervised learning to induce lexicons