/
Opinion Mining From Twitter Feeds Opinion Mining From Twitter Feeds

Opinion Mining From Twitter Feeds - PowerPoint Presentation

marina-yarberry
marina-yarberry . @marina-yarberry
Follow
386 views
Uploaded On 2017-10-12

Opinion Mining From Twitter Feeds - PPT Presentation

Lenin Ravindranath Sivalingam 25 million users around the world Over 10 million tweets a day People often express opinions Richest source of user opinions Extracting movie reviews 3000 tweets tagged by 40 users ID: 595269

2012 review polar movie review 2012 movie polar polarity words extraction great lexicon good phrases features slang tweet mixed

Share:

Link:

Embed:

Download Presentation from below link

Download Presentation The PPT/PDF document "Opinion Mining From Twitter Feeds" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.


Presentation Transcript

Slide1

Opinion Mining From Twitter Feeds

Lenin Ravindranath Sivalingam

25 million users around the world

Over 10 million tweets a day

People often express opinions!

Richest source

of user opinions!

Extracting movie

reviews!

3000 tweets tagged by 40 users

Sentiment Analysis on Tweets

Data Collection

Crowd Sourcing

Positive review

Just watched 2012, it was

pretty good...

Negative review

Saw 2012…… worst movie

ever….. waste of time

Mixed review

2012 – great effects, worst

acting, was ok.

Not a review

Went to see 2012 with my

girlfriend

Tweet Extraction

Twitter Keyword Search

Tweet Cleaning

Internet Slang Translation

Phrase Extraction

Spelling Correction

POS Tagging

SVM Classifier

Feature Extraction

Polar Words Extraction

(Polar Features)

Polarity Analysis

(Polarity Reversal)

Feature Encoding

(Phrase polarity, position)

Just saw 2012 – the movie was good, but not gr8… CG was awesum…

Lexicon

Tweet

Cleaned phrases

Tagged phrases

Review

Features

just saw MMM

the movie was good

but not great

CG was awesome

Just

/RB

saw

/VB

MMM

/NN

The

/DT

movie

/NN

was

/VB good/JJBut/CC not/RB great/JJ CG/NN was/VB awesome/JJ

MMMgood/JJnot/RB great/JJawesome/JJ

MMMPOS-JJNEG-JJPOS-JJ

MMMPOS-JJNEG-RB POS-JJPOS-JJ

1:0 2:0 3:1 4:0 5:0 6:1 7:1 8:0

Mixed review

Internet slang lexicon

Results

Polarized Lexicon

# Positive words

# Negative wordsAdjective7467Verb2926Nouns2930Adverbs2611

Polarized lexicon

Couples Retreat was hilarious.

2012 was hilarious.

New moon was ahhwwesum..

I saw 2012 today with my girlfriend. I love her.

I just saw 2012. I need my money back.

Ambiguous polar words

Too much slang

No polar words

Different context

Why errors?

Approaches

Correct predictionBag of words39.24%Adjectives44.27%Polar words58.34%Our approach (Negation + Polarity + Position)76.53%

ClassesCorrect predictionPositive review tweets80.84%Negative review tweets69.03%Mixed review tweets57.67%No review tweets80.78%Classifying subjective/objective tweets83.50%

Number of phrases to include as features

1:0 2:0

3:1 4:0

5:0 6:1

7:1 8:0