SlideShare a Scribd company logo
1 of 41
Download to read offline
NEGATIVE SENTIMENT 
another grumbly presentation from @mediaczar
TWITTER VOLUME 
15 
10 
5 
0 
1 Jun 2014 
1 Jul 2014 
Thousands 
“facebook” AND (“newsfeed” OR “news feed”) June 1 – July 31 2014 
Data: Netbase
WHAT HAPPENED HERE? 
aND HOW DOES THAT MAKE YOU 
FEEL? 
Topic Analysis 
Emotions 
Data: Brandwatch, Netbase
DELOITTE: BALANCE OF SENTIMENT AFFECTS SALES 
The balance of sentiment in Tweets is a 
more powerful driver of sales than 
reach (or volume) alone, with positive 
Tweets having generally a higher impact 
than negative Tweets. Therefore, to 
gain the most out of the online word-of- 
mouth embodied by Tweets, 
companies would be best served by 
addressing the balance of sentiment 
about their games through increasing the 
number of positive Tweets. 
6.10% 
3.30% 
1.60% 
30% more positive Tweets 
30% fewer negative Tweets 
30% more non-Twitter advertising 
Deloitte, “Tweets for Sales, Gaming” (2013)
“SENTIMENT” HAS BECOME A MARKETING OBJECTIVE 
We want to increase the 
positive buzz around Your 
brand
“SENTIMENT” HAS BECOME A MARKETING OBJECTIVE 
Real case study.  
Name obscured to protect the innocent.
THIS IS HONEST ABE 
Public sentiment is 
everything. 
With public 
sentiment, nothing 
can fail. 
Without it, nothing 
can succeed.
I’M GOING TO BE EVEN MORE HONEST 
SENTIMENT 
ANALYSIS 
IS 
SHITE! 
DTEOLNL'T U HSO WLDH ABTA CYKO.U. .REALLY THINK!
THE PROMISE OF 
SOCIAL INTELLIGENCE 
AN ALMOST INFINITE 
SOURCE OF QUAL  QUANT 
DATA! 
FINALLY THEY WILL REVEAL 
WHAT THEY REALLY THINK!
RELEVANCE 
VOLUME 
AUTHORITY 
TOPICS 
SENTIMENT
WEIGHTED SENTIMENT, MRS BROWN’S BOYS, JULY 2014 
14.8% 
12.2% 
11.4% 
7.3% 
-4.1% 
-5.2% 
-6.3% 
-8.5% 
POSITIVE 
NEGATIVE 
4 DIFFERENT TOOLS GIVE 4 DIFFERENT MEASURES OF SAME DATA
METHODOLOGICAL PROBLEMS 
LEXICAL ANALYSIS 
CLASSIFICATION 
MANUAL INPUT
LEXICAL ANALYSIS 
visit: http://www.wjh.harvard.edu/~inquirer/spreadsheet_guide.htm 
e.g. Harvard General Enquirer: 
11.8K categorised, tagged words
QUICK  DIRTY DIY SENTIMENT ANALYSIS TOOL
SOME OBVIOUS PROBLEMS…
CAN’T HANDLE IDIOMS, SYNONYMS (OR IRONY) 
WELL, THAT'S 
JUST GREAT
FACEBOOK USED LEXICAL APPROACH 
“Posts were determined to 
be positive or negative if they 
contained at least one 
positive or negative word” 
Visit: http://www.pnas.org/content/111/24/8788.full 
LIWC: http://www.liwc.net/ 
15 
10 
5 
0 
Thousands
CLASSIFIERS  SUPERVISED LEARNING… 
TRAIN MODEL 
TAGGED 
TEST MODEL 
DATA 
TRAINING 
DATA 
TEST 
DATA 
POSITIVE 
NEUTRAL 
NEGATIVE
THIS PRESENTATION IS GOOD 
visit: text-processing.com/demo/sentiment 
CLASSIFIER SAYS “POSITIVE”
THIS PRESENTATION IS BAD 
CLASSIFIER SAYS “NEGATIVE” 
visit: text-processing.com/demo/sentiment
THIS PRESENTATION IS NOT GOOD 
CLASSIFIER SAYS “NEGATIVE” 
visit: text-processing.com/demo/sentiment
THESE ARE BOTH POSITIVE 
analysed 
analysed
THIS IS NEUTRAL 
analysed
WHAT ABOUT THIS?
DOES IT UNDERSTAND WHAT IT’S READING? 
2. 
1. 
Take the original text 
Randomise the word order 
3. 
Re-test 
recipe: http://stackoverflow.com/questions/17825945/generating-a-list-of-random-words-in-excel-but-no-duplicates
WORD ORDER MAKES NO DIFFERENCE
WORD ORDER MAKES NO DIFFERENCE
WORD ORDER MAKES NO DIFFERENCE
DOMAIN SPECIFIC 
Models trained on one set of data  
may not work well on other sets
WHAT ABOUT HUMAN MARKERS?
FAIRLY EASY TO ASSESS ENTERTAINMENT CATEGORY
EXPERIMENT: IS GUINNESS GOOD FOR YOU? 
Selected 50 positive and 50 negative tweets 
as scored by classifier. 
Passed these tweets to human markers. 
Each tweet scored 3 times (5 point scale) 
Average score compared to classifier.
MECHANICAL TURKS! 
Visit: http://www.crowdflower.com/
RESULTS: IS GUINNESS GOOD FOR YOU? 
NEG 
NEUT 
POS 
CLASSIFIER 
50 
0 
50 
MANUAL 
23 
30 
47 
AGREEMENT 
40% 
0% 
72% 
(Agreement based on #tweets with matched judgments)
HUMANS DON’T ALWAYS AGREE… 
Advert used to say 
#Guinness is good for you 
but I think it is not acceptable 
to say that these days, but in 
moderation I thrive on it at 
70 
It's called 
a rotten apple 
#twobeersonecup 
#Guinness #angryorchard 
#delicious 
http://t.co/GUFxrYoF6i
CLASSIFY THIS… 
THERE ARE 
HUGE LINES AT 
THE APPLE STORE 
TODAY 
OR
BIG ENOUGH NUMBERS 
If the sample is large enough, won’t these problems get ironed out?
SAMPLE BIAS (1936 US PRESIDENTIAL ELECTION) 
SURVEY SIZE 
ROOSEVELT 
LITERARY DIGEST 
2,400,000 
43% 
GALLUP 
50,000 
54% 
ACTUAL 
61% 
See: Tim Harford, “Big Data: are we making a big mistake?” (FT Magazine, 28 March 2008) 
I THINK YOU'll find 
iT'S 48 times bigger 
ALF LANDON
WIN 30 MINS OF FREE CONSULTANCY (VALUE £750) 
I'm loving 
#breakingnewsconf, 
@mediaczar. 
IT’S FAR TOO EASY TO GAME “POSITIVE SENTIMENT” METRICS
RECOMMENDATIONS 
LIFE ISN’T A POPULARITY CONTEST 
DON’T LET SENTIMENT BECOME A KPI 
MAKE SENTIMENT A TOOL FOR MORE COMPLEX RESEARCH 
PUT GREAT TECH TO MORE MEANINGFUL USE
THANK YOU! 
PLEASE DON’T ASK ME ANY TRICKY QUESTIONS THAT WILL MAKE ME LOOK STUPID 
I’M @MEDIACZAR ON TWITTER 
FEEL FREE TO COME AND TALK TO ME AFTERWARDS

More Related Content

Viewers also liked

Generic Process from Congres Facebook Marketing
Generic Process from Congres Facebook MarketingGeneric Process from Congres Facebook Marketing
Generic Process from Congres Facebook MarketingMat Morrison
 
WOM UK 'espresso briefing' Wednesday October 28, 2009
WOM UK 'espresso briefing' Wednesday October 28, 2009WOM UK 'espresso briefing' Wednesday October 28, 2009
WOM UK 'espresso briefing' Wednesday October 28, 2009Mat Morrison
 
* INGLÉS PARA ADOLESCENTES Y NIÑOS*
* INGLÉS PARA ADOLESCENTES Y NIÑOS** INGLÉS PARA ADOLESCENTES Y NIÑOS*
* INGLÉS PARA ADOLESCENTES Y NIÑOS*colts10k
 
Facebook Pages are not a Community platform
Facebook Pages are not a Community platformFacebook Pages are not a Community platform
Facebook Pages are not a Community platformMat Morrison
 
An Incomplete Guide to Open Graph metatags for Facebook
An Incomplete Guide to Open Graph metatags for FacebookAn Incomplete Guide to Open Graph metatags for Facebook
An Incomplete Guide to Open Graph metatags for FacebookMat Morrison
 
Facebook - What You'd Forgotten (and Why it Matters)
Facebook - What You'd Forgotten (and Why it Matters)Facebook - What You'd Forgotten (and Why it Matters)
Facebook - What You'd Forgotten (and Why it Matters)Mat Morrison
 
Why your fan count doesn't matter.
Why your fan count doesn't matter.Why your fan count doesn't matter.
Why your fan count doesn't matter.Mat Morrison
 
TEDx Tuttle Presentation
TEDx Tuttle PresentationTEDx Tuttle Presentation
TEDx Tuttle PresentationMat Morrison
 
WTF Do You Want Me To Measure Now?
WTF Do You Want Me To Measure Now?WTF Do You Want Me To Measure Now?
WTF Do You Want Me To Measure Now?Mat Morrison
 
The Value of a Fan: Webit 2011: Script
The Value of a Fan: Webit 2011: ScriptThe Value of a Fan: Webit 2011: Script
The Value of a Fan: Webit 2011: ScriptMat Morrison
 
Get Tweet source data
Get Tweet source dataGet Tweet source data
Get Tweet source dataMat Morrison
 
Seminar Report 20071127
Seminar Report 20071127Seminar Report 20071127
Seminar Report 20071127tomelf2007
 

Viewers also liked (13)

Generic Process from Congres Facebook Marketing
Generic Process from Congres Facebook MarketingGeneric Process from Congres Facebook Marketing
Generic Process from Congres Facebook Marketing
 
WOM UK 'espresso briefing' Wednesday October 28, 2009
WOM UK 'espresso briefing' Wednesday October 28, 2009WOM UK 'espresso briefing' Wednesday October 28, 2009
WOM UK 'espresso briefing' Wednesday October 28, 2009
 
* INGLÉS PARA ADOLESCENTES Y NIÑOS*
* INGLÉS PARA ADOLESCENTES Y NIÑOS** INGLÉS PARA ADOLESCENTES Y NIÑOS*
* INGLÉS PARA ADOLESCENTES Y NIÑOS*
 
Facebook Pages are not a Community platform
Facebook Pages are not a Community platformFacebook Pages are not a Community platform
Facebook Pages are not a Community platform
 
An Incomplete Guide to Open Graph metatags for Facebook
An Incomplete Guide to Open Graph metatags for FacebookAn Incomplete Guide to Open Graph metatags for Facebook
An Incomplete Guide to Open Graph metatags for Facebook
 
Facebook - What You'd Forgotten (and Why it Matters)
Facebook - What You'd Forgotten (and Why it Matters)Facebook - What You'd Forgotten (and Why it Matters)
Facebook - What You'd Forgotten (and Why it Matters)
 
Why your fan count doesn't matter.
Why your fan count doesn't matter.Why your fan count doesn't matter.
Why your fan count doesn't matter.
 
TEDx Tuttle Presentation
TEDx Tuttle PresentationTEDx Tuttle Presentation
TEDx Tuttle Presentation
 
WTF Do You Want Me To Measure Now?
WTF Do You Want Me To Measure Now?WTF Do You Want Me To Measure Now?
WTF Do You Want Me To Measure Now?
 
The Value of a Fan: Webit 2011: Script
The Value of a Fan: Webit 2011: ScriptThe Value of a Fan: Webit 2011: Script
The Value of a Fan: Webit 2011: Script
 
Get Tweet source data
Get Tweet source dataGet Tweet source data
Get Tweet source data
 
Seminar Report 20071127
Seminar Report 20071127Seminar Report 20071127
Seminar Report 20071127
 
powerpoint
powerpointpowerpoint
powerpoint
 

Similar to Sentiment is Shite (2014)

Are You Really Doing Effective Prevention - from IADDA Conference 2014
Are You Really Doing Effective Prevention - from IADDA Conference 2014Are You Really Doing Effective Prevention - from IADDA Conference 2014
Are You Really Doing Effective Prevention - from IADDA Conference 2014LEAD
 
Upworthy & Bill and Melinda Gates Foundation - All 7 Billion - Research Summary
Upworthy & Bill and Melinda Gates Foundation - All 7 Billion - Research SummaryUpworthy & Bill and Melinda Gates Foundation - All 7 Billion - Research Summary
Upworthy & Bill and Melinda Gates Foundation - All 7 Billion - Research SummarySean P. Wojcik
 
Future of Online Organizing training
Future of Online Organizing training  Future of Online Organizing training
Future of Online Organizing training Cheryl Contee
 
0915 omma metrics jim sterne
0915 omma metrics jim sterne0915 omma metrics jim sterne
0915 omma metrics jim sterneMediaPost
 
0915 omma metrics jim sterne
0915 omma metrics jim sterne0915 omma metrics jim sterne
0915 omma metrics jim sterneMediaPost
 
Pairing Analytics With Qualitative Methods to Understand the WHY
Pairing Analytics With Qualitative Methods to Understand the WHYPairing Analytics With Qualitative Methods to Understand the WHY
Pairing Analytics With Qualitative Methods to Understand the WHYMichele Kiss
 
Influencer poll webinar 1 7slideshare
Influencer poll webinar 1 7slideshareInfluencer poll webinar 1 7slideshare
Influencer poll webinar 1 7slideshareCision
 
7 Things That Make Content Go Viral
7 Things That Make Content Go Viral7 Things That Make Content Go Viral
7 Things That Make Content Go ViralDerek Halpern
 
UpStart Bay Area - Haas Fellows (April 8, 2014)
UpStart Bay Area - Haas Fellows (April 8, 2014)UpStart Bay Area - Haas Fellows (April 8, 2014)
UpStart Bay Area - Haas Fellows (April 8, 2014)Esther Kustanowitz
 
The Narrative Project - Overview Deck July 2014
The Narrative Project - Overview Deck July 2014The Narrative Project - Overview Deck July 2014
The Narrative Project - Overview Deck July 2014James North
 
The Bright Future of Market Research Smartees Workshop
The Bright Future of Market Research Smartees WorkshopThe Bright Future of Market Research Smartees Workshop
The Bright Future of Market Research Smartees WorkshopInSites on Stage
 
Getting Her Attention - Online and Offline Marketing Strategies that Work
Getting Her Attention - Online and Offline Marketing Strategies that WorkGetting Her Attention - Online and Offline Marketing Strategies that Work
Getting Her Attention - Online and Offline Marketing Strategies that WorkElizabeth Scott
 
Social Media Nethnography: Mining and Understanding the Consumer Conversation
Social Media Nethnography: Mining and Understanding the Consumer ConversationSocial Media Nethnography: Mining and Understanding the Consumer Conversation
Social Media Nethnography: Mining and Understanding the Consumer ConversationSchillewaert Niels
 
Influencing the Influencer (Influencer Marketing Done Right)
Influencing the Influencer (Influencer Marketing Done Right)Influencing the Influencer (Influencer Marketing Done Right)
Influencing the Influencer (Influencer Marketing Done Right)CleverGirlsColl
 
Social Media for Nonprofits | National Philanthropy Day Conference
Social Media for Nonprofits | National Philanthropy Day ConferenceSocial Media for Nonprofits | National Philanthropy Day Conference
Social Media for Nonprofits | National Philanthropy Day ConferenceWahine Media
 
Buzz Awards Seminar 2009 Emanuel Rosen
Buzz Awards Seminar 2009 Emanuel RosenBuzz Awards Seminar 2009 Emanuel Rosen
Buzz Awards Seminar 2009 Emanuel RosenWillem Sodderland
 

Similar to Sentiment is Shite (2014) (20)

Words of Wisdom
Words of WisdomWords of Wisdom
Words of Wisdom
 
Are You Really Doing Effective Prevention - from IADDA Conference 2014
Are You Really Doing Effective Prevention - from IADDA Conference 2014Are You Really Doing Effective Prevention - from IADDA Conference 2014
Are You Really Doing Effective Prevention - from IADDA Conference 2014
 
Upworthy & Bill and Melinda Gates Foundation - All 7 Billion - Research Summary
Upworthy & Bill and Melinda Gates Foundation - All 7 Billion - Research SummaryUpworthy & Bill and Melinda Gates Foundation - All 7 Billion - Research Summary
Upworthy & Bill and Melinda Gates Foundation - All 7 Billion - Research Summary
 
Focus Group Testing
Focus Group TestingFocus Group Testing
Focus Group Testing
 
Future of Online Organizing training
Future of Online Organizing training  Future of Online Organizing training
Future of Online Organizing training
 
0915 omma metrics jim sterne
0915 omma metrics jim sterne0915 omma metrics jim sterne
0915 omma metrics jim sterne
 
0915 omma metrics jim sterne
0915 omma metrics jim sterne0915 omma metrics jim sterne
0915 omma metrics jim sterne
 
Pairing Analytics With Qualitative Methods to Understand the WHY
Pairing Analytics With Qualitative Methods to Understand the WHYPairing Analytics With Qualitative Methods to Understand the WHY
Pairing Analytics With Qualitative Methods to Understand the WHY
 
Influencer poll webinar 1 7slideshare
Influencer poll webinar 1 7slideshareInfluencer poll webinar 1 7slideshare
Influencer poll webinar 1 7slideshare
 
7 Things That Make Content Go Viral
7 Things That Make Content Go Viral7 Things That Make Content Go Viral
7 Things That Make Content Go Viral
 
UpStart Bay Area - Haas Fellows (April 8, 2014)
UpStart Bay Area - Haas Fellows (April 8, 2014)UpStart Bay Area - Haas Fellows (April 8, 2014)
UpStart Bay Area - Haas Fellows (April 8, 2014)
 
The Narrative Project - Overview Deck July 2014
The Narrative Project - Overview Deck July 2014The Narrative Project - Overview Deck July 2014
The Narrative Project - Overview Deck July 2014
 
The Bright Future of Market Research Smartees Workshop
The Bright Future of Market Research Smartees WorkshopThe Bright Future of Market Research Smartees Workshop
The Bright Future of Market Research Smartees Workshop
 
Getting Her Attention - Online and Offline Marketing Strategies that Work
Getting Her Attention - Online and Offline Marketing Strategies that WorkGetting Her Attention - Online and Offline Marketing Strategies that Work
Getting Her Attention - Online and Offline Marketing Strategies that Work
 
Social Media Nethnography: Mining and Understanding the Consumer Conversation
Social Media Nethnography: Mining and Understanding the Consumer ConversationSocial Media Nethnography: Mining and Understanding the Consumer Conversation
Social Media Nethnography: Mining and Understanding the Consumer Conversation
 
Real House Wives of Social Media
Real House Wives of Social MediaReal House Wives of Social Media
Real House Wives of Social Media
 
Facebook Engagement Strategies for Healthcare Organizations
Facebook Engagement Strategies for Healthcare OrganizationsFacebook Engagement Strategies for Healthcare Organizations
Facebook Engagement Strategies for Healthcare Organizations
 
Influencing the Influencer (Influencer Marketing Done Right)
Influencing the Influencer (Influencer Marketing Done Right)Influencing the Influencer (Influencer Marketing Done Right)
Influencing the Influencer (Influencer Marketing Done Right)
 
Social Media for Nonprofits | National Philanthropy Day Conference
Social Media for Nonprofits | National Philanthropy Day ConferenceSocial Media for Nonprofits | National Philanthropy Day Conference
Social Media for Nonprofits | National Philanthropy Day Conference
 
Buzz Awards Seminar 2009 Emanuel Rosen
Buzz Awards Seminar 2009 Emanuel RosenBuzz Awards Seminar 2009 Emanuel Rosen
Buzz Awards Seminar 2009 Emanuel Rosen
 

More from Mat Morrison

Engagement vs Social Media in 'Brand Republic' archives
Engagement vs Social Media in 'Brand Republic' archivesEngagement vs Social Media in 'Brand Republic' archives
Engagement vs Social Media in 'Brand Republic' archivesMat Morrison
 
Redbull anecdotal example
Redbull anecdotal exampleRedbull anecdotal example
Redbull anecdotal exampleMat Morrison
 
What do we want from our Facebook fans?
What do we want from our Facebook fans?What do we want from our Facebook fans?
What do we want from our Facebook fans?Mat Morrison
 
Posts on Kohl's Facebook Page August - October 2012
Posts on Kohl's Facebook Page August - October 2012Posts on Kohl's Facebook Page August - October 2012
Posts on Kohl's Facebook Page August - October 2012Mat Morrison
 
Marks & Spencer Facebook Page
Marks & Spencer Facebook PageMarks & Spencer Facebook Page
Marks & Spencer Facebook PageMat Morrison
 
How People Use Facebook -- And Why It Matters
How People Use Facebook -- And Why It MattersHow People Use Facebook -- And Why It Matters
How People Use Facebook -- And Why It MattersMat Morrison
 
Supermarkets Xmas (for Jamie Riddell)
Supermarkets Xmas (for Jamie Riddell)Supermarkets Xmas (for Jamie Riddell)
Supermarkets Xmas (for Jamie Riddell)Mat Morrison
 
The Value of a Fan: Webit 2011
The Value of a Fan: Webit 2011The Value of a Fan: Webit 2011
The Value of a Fan: Webit 2011Mat Morrison
 
Optimising the Facebook Stack for Earned Media
Optimising the Facebook Stack for Earned MediaOptimising the Facebook Stack for Earned Media
Optimising the Facebook Stack for Earned MediaMat Morrison
 
Socialised Content
Socialised ContentSocialised Content
Socialised ContentMat Morrison
 
Negative Sentiment (or "Sentiment Analysis is Sh*te")
Negative Sentiment (or "Sentiment Analysis is Sh*te")Negative Sentiment (or "Sentiment Analysis is Sh*te")
Negative Sentiment (or "Sentiment Analysis is Sh*te")Mat Morrison
 
Social Media 2010: Harnessing the Power of Social Networks
Social Media 2010: Harnessing the Power of Social NetworksSocial Media 2010: Harnessing the Power of Social Networks
Social Media 2010: Harnessing the Power of Social NetworksMat Morrison
 
Social Media Week 2010: Social Media For Enterprise
Social Media Week 2010: Social Media For EnterpriseSocial Media Week 2010: Social Media For Enterprise
Social Media Week 2010: Social Media For EnterpriseMat Morrison
 
Association of Publishing Agencies: Ad Exchanges
Association of Publishing Agencies: Ad Exchanges Association of Publishing Agencies: Ad Exchanges
Association of Publishing Agencies: Ad Exchanges Mat Morrison
 
PRCA breakfast: Positioning Digital Properly
PRCA breakfast: Positioning Digital ProperlyPRCA breakfast: Positioning Digital Properly
PRCA breakfast: Positioning Digital ProperlyMat Morrison
 
Vienna Social Media Conf, 20 April 2009
Vienna Social Media Conf, 20 April 2009Vienna Social Media Conf, 20 April 2009
Vienna Social Media Conf, 20 April 2009Mat Morrison
 
The Integration Triangle
The Integration TriangleThe Integration Triangle
The Integration TriangleMat Morrison
 
1 A Social Media Checklist
1   A Social Media Checklist1   A Social Media Checklist
1 A Social Media ChecklistMat Morrison
 

More from Mat Morrison (20)

Engagement vs Social Media in 'Brand Republic' archives
Engagement vs Social Media in 'Brand Republic' archivesEngagement vs Social Media in 'Brand Republic' archives
Engagement vs Social Media in 'Brand Republic' archives
 
Redbull anecdotal example
Redbull anecdotal exampleRedbull anecdotal example
Redbull anecdotal example
 
Shareable Content
Shareable ContentShareable Content
Shareable Content
 
What do we want from our Facebook fans?
What do we want from our Facebook fans?What do we want from our Facebook fans?
What do we want from our Facebook fans?
 
Posts on Kohl's Facebook Page August - October 2012
Posts on Kohl's Facebook Page August - October 2012Posts on Kohl's Facebook Page August - October 2012
Posts on Kohl's Facebook Page August - October 2012
 
Marks & Spencer Facebook Page
Marks & Spencer Facebook PageMarks & Spencer Facebook Page
Marks & Spencer Facebook Page
 
How People Use Facebook -- And Why It Matters
How People Use Facebook -- And Why It MattersHow People Use Facebook -- And Why It Matters
How People Use Facebook -- And Why It Matters
 
Measurement.ie
Measurement.ieMeasurement.ie
Measurement.ie
 
Supermarkets Xmas (for Jamie Riddell)
Supermarkets Xmas (for Jamie Riddell)Supermarkets Xmas (for Jamie Riddell)
Supermarkets Xmas (for Jamie Riddell)
 
The Value of a Fan: Webit 2011
The Value of a Fan: Webit 2011The Value of a Fan: Webit 2011
The Value of a Fan: Webit 2011
 
Optimising the Facebook Stack for Earned Media
Optimising the Facebook Stack for Earned MediaOptimising the Facebook Stack for Earned Media
Optimising the Facebook Stack for Earned Media
 
Socialised Content
Socialised ContentSocialised Content
Socialised Content
 
Negative Sentiment (or "Sentiment Analysis is Sh*te")
Negative Sentiment (or "Sentiment Analysis is Sh*te")Negative Sentiment (or "Sentiment Analysis is Sh*te")
Negative Sentiment (or "Sentiment Analysis is Sh*te")
 
Social Media 2010: Harnessing the Power of Social Networks
Social Media 2010: Harnessing the Power of Social NetworksSocial Media 2010: Harnessing the Power of Social Networks
Social Media 2010: Harnessing the Power of Social Networks
 
Social Media Week 2010: Social Media For Enterprise
Social Media Week 2010: Social Media For EnterpriseSocial Media Week 2010: Social Media For Enterprise
Social Media Week 2010: Social Media For Enterprise
 
Association of Publishing Agencies: Ad Exchanges
Association of Publishing Agencies: Ad Exchanges Association of Publishing Agencies: Ad Exchanges
Association of Publishing Agencies: Ad Exchanges
 
PRCA breakfast: Positioning Digital Properly
PRCA breakfast: Positioning Digital ProperlyPRCA breakfast: Positioning Digital Properly
PRCA breakfast: Positioning Digital Properly
 
Vienna Social Media Conf, 20 April 2009
Vienna Social Media Conf, 20 April 2009Vienna Social Media Conf, 20 April 2009
Vienna Social Media Conf, 20 April 2009
 
The Integration Triangle
The Integration TriangleThe Integration Triangle
The Integration Triangle
 
1 A Social Media Checklist
1   A Social Media Checklist1   A Social Media Checklist
1 A Social Media Checklist
 

Recently uploaded

modul pembelajaran robotic Workshop _ by Slidesgo.pptx
modul pembelajaran robotic Workshop _ by Slidesgo.pptxmodul pembelajaran robotic Workshop _ by Slidesgo.pptx
modul pembelajaran robotic Workshop _ by Slidesgo.pptxaleedritatuxx
 
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)jennyeacort
 
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...Boston Institute of Analytics
 
RadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfRadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfgstagge
 
ASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel CanterASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel Cantervoginip
 
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...Amil Baba Dawood bangali
 
LLMs, LMMs, their Improvement Suggestions and the Path towards AGI
LLMs, LMMs, their Improvement Suggestions and the Path towards AGILLMs, LMMs, their Improvement Suggestions and the Path towards AGI
LLMs, LMMs, their Improvement Suggestions and the Path towards AGIThomas Poetter
 
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...ssuserf63bd7
 
Advanced Machine Learning for Business Professionals
Advanced Machine Learning for Business ProfessionalsAdvanced Machine Learning for Business Professionals
Advanced Machine Learning for Business ProfessionalsVICTOR MAESTRE RAMIREZ
 
Real-Time AI Streaming - AI Max Princeton
Real-Time AI  Streaming - AI Max PrincetonReal-Time AI  Streaming - AI Max Princeton
Real-Time AI Streaming - AI Max PrincetonTimothy Spann
 
detection and classification of knee osteoarthritis.pptx
detection and classification of knee osteoarthritis.pptxdetection and classification of knee osteoarthritis.pptx
detection and classification of knee osteoarthritis.pptxAleenaJamil4
 
Data Factory in Microsoft Fabric (MsBIP #82)
Data Factory in Microsoft Fabric (MsBIP #82)Data Factory in Microsoft Fabric (MsBIP #82)
Data Factory in Microsoft Fabric (MsBIP #82)Cathrine Wilhelmsen
 
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024Timothy Spann
 
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfPredicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfBoston Institute of Analytics
 
Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...
Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...
Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...limedy534
 
Defining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data StoryDefining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data StoryJeremy Anderson
 
RS 9000 Call In girls Dwarka Mor (DELHI)⇛9711147426🔝Delhi
RS 9000 Call In girls Dwarka Mor (DELHI)⇛9711147426🔝DelhiRS 9000 Call In girls Dwarka Mor (DELHI)⇛9711147426🔝Delhi
RS 9000 Call In girls Dwarka Mor (DELHI)⇛9711147426🔝Delhijennyeacort
 
20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdf20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdfHuman37
 
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degreeyuu sss
 

Recently uploaded (20)

modul pembelajaran robotic Workshop _ by Slidesgo.pptx
modul pembelajaran robotic Workshop _ by Slidesgo.pptxmodul pembelajaran robotic Workshop _ by Slidesgo.pptx
modul pembelajaran robotic Workshop _ by Slidesgo.pptx
 
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
Call Us ➥97111√47426🤳Call Girls in Aerocity (Delhi NCR)
 
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
NLP Data Science Project Presentation:Predicting Heart Disease with NLP Data ...
 
RadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfRadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdf
 
ASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel CanterASML's Taxonomy Adventure by Daniel Canter
ASML's Taxonomy Adventure by Daniel Canter
 
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
NO1 Certified Black Magic Specialist Expert Amil baba in Lahore Islamabad Raw...
 
LLMs, LMMs, their Improvement Suggestions and the Path towards AGI
LLMs, LMMs, their Improvement Suggestions and the Path towards AGILLMs, LMMs, their Improvement Suggestions and the Path towards AGI
LLMs, LMMs, their Improvement Suggestions and the Path towards AGI
 
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
Statistics, Data Analysis, and Decision Modeling, 5th edition by James R. Eva...
 
Advanced Machine Learning for Business Professionals
Advanced Machine Learning for Business ProfessionalsAdvanced Machine Learning for Business Professionals
Advanced Machine Learning for Business Professionals
 
Real-Time AI Streaming - AI Max Princeton
Real-Time AI  Streaming - AI Max PrincetonReal-Time AI  Streaming - AI Max Princeton
Real-Time AI Streaming - AI Max Princeton
 
detection and classification of knee osteoarthritis.pptx
detection and classification of knee osteoarthritis.pptxdetection and classification of knee osteoarthritis.pptx
detection and classification of knee osteoarthritis.pptx
 
Data Factory in Microsoft Fabric (MsBIP #82)
Data Factory in Microsoft Fabric (MsBIP #82)Data Factory in Microsoft Fabric (MsBIP #82)
Data Factory in Microsoft Fabric (MsBIP #82)
 
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
April 2024 - NLIT Cloudera Real-Time LLM Streaming 2024
 
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfPredicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
 
Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...
Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...
Effects of Smartphone Addiction on the Academic Performances of Grades 9 to 1...
 
Defining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data StoryDefining Constituents, Data Vizzes and Telling a Data Story
Defining Constituents, Data Vizzes and Telling a Data Story
 
RS 9000 Call In girls Dwarka Mor (DELHI)⇛9711147426🔝Delhi
RS 9000 Call In girls Dwarka Mor (DELHI)⇛9711147426🔝DelhiRS 9000 Call In girls Dwarka Mor (DELHI)⇛9711147426🔝Delhi
RS 9000 Call In girls Dwarka Mor (DELHI)⇛9711147426🔝Delhi
 
20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdf20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdf
 
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
 
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
办美国阿肯色大学小石城分校毕业证成绩单pdf电子版制作修改#真实留信入库#永久存档#真实可查#diploma#degree
 

Sentiment is Shite (2014)

  • 1. NEGATIVE SENTIMENT another grumbly presentation from @mediaczar
  • 2. TWITTER VOLUME 15 10 5 0 1 Jun 2014 1 Jul 2014 Thousands “facebook” AND (“newsfeed” OR “news feed”) June 1 – July 31 2014 Data: Netbase
  • 3. WHAT HAPPENED HERE? aND HOW DOES THAT MAKE YOU FEEL? Topic Analysis Emotions Data: Brandwatch, Netbase
  • 4. DELOITTE: BALANCE OF SENTIMENT AFFECTS SALES The balance of sentiment in Tweets is a more powerful driver of sales than reach (or volume) alone, with positive Tweets having generally a higher impact than negative Tweets. Therefore, to gain the most out of the online word-of- mouth embodied by Tweets, companies would be best served by addressing the balance of sentiment about their games through increasing the number of positive Tweets. 6.10% 3.30% 1.60% 30% more positive Tweets 30% fewer negative Tweets 30% more non-Twitter advertising Deloitte, “Tweets for Sales, Gaming” (2013)
  • 5. “SENTIMENT” HAS BECOME A MARKETING OBJECTIVE We want to increase the positive buzz around Your brand
  • 6. “SENTIMENT” HAS BECOME A MARKETING OBJECTIVE Real case study. Name obscured to protect the innocent.
  • 7. THIS IS HONEST ABE Public sentiment is everything. With public sentiment, nothing can fail. Without it, nothing can succeed.
  • 8. I’M GOING TO BE EVEN MORE HONEST SENTIMENT ANALYSIS IS SHITE! DTEOLNL'T U HSO WLDH ABTA CYKO.U. .REALLY THINK!
  • 9. THE PROMISE OF SOCIAL INTELLIGENCE AN ALMOST INFINITE SOURCE OF QUAL QUANT DATA! FINALLY THEY WILL REVEAL WHAT THEY REALLY THINK!
  • 10. RELEVANCE VOLUME AUTHORITY TOPICS SENTIMENT
  • 11. WEIGHTED SENTIMENT, MRS BROWN’S BOYS, JULY 2014 14.8% 12.2% 11.4% 7.3% -4.1% -5.2% -6.3% -8.5% POSITIVE NEGATIVE 4 DIFFERENT TOOLS GIVE 4 DIFFERENT MEASURES OF SAME DATA
  • 12. METHODOLOGICAL PROBLEMS LEXICAL ANALYSIS CLASSIFICATION MANUAL INPUT
  • 13. LEXICAL ANALYSIS visit: http://www.wjh.harvard.edu/~inquirer/spreadsheet_guide.htm e.g. Harvard General Enquirer: 11.8K categorised, tagged words
  • 14. QUICK DIRTY DIY SENTIMENT ANALYSIS TOOL
  • 16. CAN’T HANDLE IDIOMS, SYNONYMS (OR IRONY) WELL, THAT'S JUST GREAT
  • 17. FACEBOOK USED LEXICAL APPROACH “Posts were determined to be positive or negative if they contained at least one positive or negative word” Visit: http://www.pnas.org/content/111/24/8788.full LIWC: http://www.liwc.net/ 15 10 5 0 Thousands
  • 18. CLASSIFIERS SUPERVISED LEARNING… TRAIN MODEL TAGGED TEST MODEL DATA TRAINING DATA TEST DATA POSITIVE NEUTRAL NEGATIVE
  • 19. THIS PRESENTATION IS GOOD visit: text-processing.com/demo/sentiment CLASSIFIER SAYS “POSITIVE”
  • 20. THIS PRESENTATION IS BAD CLASSIFIER SAYS “NEGATIVE” visit: text-processing.com/demo/sentiment
  • 21. THIS PRESENTATION IS NOT GOOD CLASSIFIER SAYS “NEGATIVE” visit: text-processing.com/demo/sentiment
  • 22. THESE ARE BOTH POSITIVE analysed analysed
  • 23. THIS IS NEUTRAL analysed
  • 25. DOES IT UNDERSTAND WHAT IT’S READING? 2. 1. Take the original text Randomise the word order 3. Re-test recipe: http://stackoverflow.com/questions/17825945/generating-a-list-of-random-words-in-excel-but-no-duplicates
  • 26. WORD ORDER MAKES NO DIFFERENCE
  • 27. WORD ORDER MAKES NO DIFFERENCE
  • 28. WORD ORDER MAKES NO DIFFERENCE
  • 29. DOMAIN SPECIFIC Models trained on one set of data may not work well on other sets
  • 30. WHAT ABOUT HUMAN MARKERS?
  • 31. FAIRLY EASY TO ASSESS ENTERTAINMENT CATEGORY
  • 32. EXPERIMENT: IS GUINNESS GOOD FOR YOU? Selected 50 positive and 50 negative tweets as scored by classifier. Passed these tweets to human markers. Each tweet scored 3 times (5 point scale) Average score compared to classifier.
  • 33. MECHANICAL TURKS! Visit: http://www.crowdflower.com/
  • 34. RESULTS: IS GUINNESS GOOD FOR YOU? NEG NEUT POS CLASSIFIER 50 0 50 MANUAL 23 30 47 AGREEMENT 40% 0% 72% (Agreement based on #tweets with matched judgments)
  • 35. HUMANS DON’T ALWAYS AGREE… Advert used to say #Guinness is good for you but I think it is not acceptable to say that these days, but in moderation I thrive on it at 70 It's called a rotten apple #twobeersonecup #Guinness #angryorchard #delicious http://t.co/GUFxrYoF6i
  • 36. CLASSIFY THIS… THERE ARE HUGE LINES AT THE APPLE STORE TODAY OR
  • 37. BIG ENOUGH NUMBERS If the sample is large enough, won’t these problems get ironed out?
  • 38. SAMPLE BIAS (1936 US PRESIDENTIAL ELECTION) SURVEY SIZE ROOSEVELT LITERARY DIGEST 2,400,000 43% GALLUP 50,000 54% ACTUAL 61% See: Tim Harford, “Big Data: are we making a big mistake?” (FT Magazine, 28 March 2008) I THINK YOU'll find iT'S 48 times bigger ALF LANDON
  • 39. WIN 30 MINS OF FREE CONSULTANCY (VALUE £750) I'm loving #breakingnewsconf, @mediaczar. IT’S FAR TOO EASY TO GAME “POSITIVE SENTIMENT” METRICS
  • 40. RECOMMENDATIONS LIFE ISN’T A POPULARITY CONTEST DON’T LET SENTIMENT BECOME A KPI MAKE SENTIMENT A TOOL FOR MORE COMPLEX RESEARCH PUT GREAT TECH TO MORE MEANINGFUL USE
  • 41. THANK YOU! PLEASE DON’T ASK ME ANY TRICKY QUESTIONS THAT WILL MAKE ME LOOK STUPID I’M @MEDIACZAR ON TWITTER FEEL FREE TO COME AND TALK TO ME AFTERWARDS

Editor's Notes

  1. "respondents who returned their questionnaires represented only that subset of the population with a relatively intense interest in the subject at hand, and as such constitute in no sense a random sample... it seems clear that the minority of anti-Roosevelt voters felt more strongly about the election than did the pro-Roosevelt majority.” 50x bigger sample. Wrong answer.