SlideShare a Scribd company logo
1 of 22
Download to read offline
Faking Sandy: Characterizing and
Identifying Fake Images on Twitter
during Hurricane Sandy	

Adi$	
  Gupta	
  
Hemank	
  Lamba	
  
Ponnurangam	
  Kumaraguru	
  
Anupam	
  Joshi	
  

	
  
Agenda
 Background	
  
 Mo$va$on	
  
 Main	
  contribu$ons	
  
 Data	
  descrip$on	
  
 Methodology	
  
 Results	
  
 Future	
  work	
  

	
  precog.iiitd.edu.in	
  

2	
  
Background: Hurricane Sandy
 Dates:	
  Oct	
  22-­‐	
  31,	
  2012	
  
 Category	
  3	
  storm	
  
 Damages	
  worth	
  $75	
  billion	
  
 Coast	
  of	
  NE	
  America	
  

	
  precog.iiitd.edu.in	
  

3	
  
Motivation
FAKE	
  
IMAGES	
  

	
  precog.iiitd.edu.in	
  

4	
  
Motivation

	
  precog.iiitd.edu.in	
  

5	
  
Our Contributions
  Performed	
  in-­‐depth	
  characteriza$on	
  of	
  tweets	
  sharing	
  
fake	
  images	
  on	
  TwiVer	
  during	
  Hurricane	
  Sandy	
  

-  Tweets	
  containing	
  the	
  fake	
  images	
  URLs	
  were	
  mostly	
  
retweets	
  (86%)	
  
-  Just	
  11%	
  overlap	
  between	
  the	
  retweet	
  and	
  follower	
  graphs	
  of	
  
tweets	
  containing	
  fake	
  images.	
  	
  	
  

  Applied	
  classifica$on	
  algorithms	
  to	
  dis$nguish	
  between	
  
tweets	
  containing	
  fake	
  and	
  real	
  images.	
  	
  
-  Best	
  accuracy	
  of	
  97%	
  was	
  achieved	
  using	
  decision	
  tree	
  
classifier,	
  using	
  tweet	
  based	
  features.	
  
	
  

	
  precog.iiitd.edu.in	
  

6	
  
Methodology

Feature	
  
Genera$on	
  
Data	
  Collec$on	
  
and	
  Filtering	
  

Evalua$ng	
  
Results	
  

Data	
  
Characteriza$on	
  
Obtaining	
  
Ground	
  Truth	
  
Classifica$on	
  Module	
  

	
  precog.iiitd.edu.in	
  

7	
  
Data Description
Total	
  tweets	
  
Total	
  unique	
  users	
  

1,174,266	
  

Tweets	
  with	
  URLs	
  

	
  precog.iiitd.edu.in	
  

1,782,526	
  
622,860	
  

8	
  
Data Filtering
 Reputable	
  online	
  resource	
  to	
  filter	
  fake	
  and	
  real	
  
images	
  
- Guardian	
  collected	
  and	
  publically	
  distributed	
  a	
  list	
  
of	
  fake	
  and	
  true	
  images	
  shared	
  during	
  Hurricane	
  
Sandy	
  
	
  
	
  
Tweets	
  with	
  fake	
  images	
  
10,350	
  
	
  
Users	
  with	
  fake	
  images	
  
10,215	
  
	
  
	
  
Tweets	
  with	
  real	
  images	
  
5,767	
  
	
  
	
  
Users	
  with	
  real	
  images	
  
5,678	
  
	
  

 One	
  of	
  the	
  biggest	
  fake	
  content	
  propaga$on	
  
datasets	
  that	
  have	
  been	
  studied	
  by	
  researchers	
  
	
  precog.iiitd.edu.in	
  

9	
  
Characterization – Fake Image
Propagation
 86%	
  of	
  tweets	
  spreading	
  the	
  fake	
  images	
  were	
  
retweets	
  
 Top	
  30	
  users	
  out	
  of	
  10,215	
  users	
  (0.3%)	
  resulted	
  
in	
  90%	
  of	
  the	
  retweets	
  of	
  fake	
  images	
  

	
  precog.iiitd.edu.in	
  

10	
  
Network Analysis

Tweet	
  –	
  Retweet	
  graph	
  for	
  the	
  spread	
  of	
  
fake	
  images	
  at	
  ‘nth’	
  and	
  ‘n+1th‘	
  hour	
  

	
  precog.iiitd.edu.in	
  

11	
  
Role of Explicit Twitter Network
 Analyzing	
  role	
  of	
  follower	
  network	
  in	
  fake	
  
image	
  propaga$on	
  
 We	
  crawled	
  the	
  TwiVer	
  network	
  for	
  all	
  users	
  
who	
  tweeted	
  the	
  fake	
  image	
  URLs	
  

	
  precog.iiitd.edu.in	
  

12	
  
Algorithm

	
  precog.iiitd.edu.in	
  

13	
  
Results
Total	
  edges	
  in	
  the	
  retweet	
  network	
  
Total	
  edges	
  in	
  the	
  follower-­‐followee	
  network	
  
Total	
  edges	
  that	
  exist	
  in	
  both	
  retweet	
  
network	
  and	
  the	
  follower-­‐	
  followee	
  network	
  
%age	
  Overlap	
  

	
  precog.iiitd.edu.in	
  

10,508	
  
10,799,122	
  
1,215	
  
11%	
  

14	
  
Classification
 5	
  fold	
  cross	
  valida$on	
  
User	
  Features	
  [F1]	
  
Number	
  of	
  Friends	
  
Number	
  of	
  Followers	
  
Follower-­‐Friend	
  Ra$o	
  
Number	
  of	
  $mes	
  listed	
  
User	
  has	
  a	
  URL	
  
User	
  is	
  a	
  verified	
  user	
  
Age	
  of	
  user	
  account	
  

	
  precog.iiitd.edu.in	
  

Tweet	
  Features	
  [F2]	
  
Length	
  of	
  Tweet	
  
Number	
  of	
  Words	
  
Contains	
  Ques$on	
  Mark?	
  
Contains	
  Exclama$on	
  Mark?	
  
Number	
  of	
  Ques$on	
  Marks	
  
Number	
  of	
  Exclama$on	
  Marks	
  
Contains	
  Happy	
  Emo$con	
  
Contains	
  Sad	
  Emo$con	
  
Contains	
  First	
  Order	
  Pronoun	
  
Contains	
  Second	
  Order	
  Pronoun	
  
Contains	
  Third	
  Order	
  Pronoun	
  
Number	
  of	
  uppercase	
  characters	
  
Number	
  of	
  nega$ve	
  sen$ment	
  words	
  
Number	
  of	
  posi$ve	
  sen$ment	
  words	
  
Number	
  of	
  men$ons	
  
Number	
  of	
  hashtags	
  
Number	
  of	
  URLs	
  
Retweet	
  count	
  
15	
  
Classification Results
F1	
  (user)	
   F2	
  (tweet)	
  

F1+F2	
  

Naïve	
  Bayes	
  

56.32%	
  

91.97%	
  

91.52%	
  

Decision	
  Tree	
  

53.24%	
  

97.65%	
  

96.65%	
  

•  Best	
  results	
  were	
  obtained	
  from	
  Decision	
  Tree	
  classifier,	
  we	
  got	
  97%	
  
accuracy	
  in	
  predic$ng	
  fake	
  images	
  from	
  real.	
  	
  
•  Tweet	
  based	
  features	
  are	
  very	
  effec$ve	
  in	
  dis$nguishing	
  fake	
  images	
  tweets	
  
from	
  real,	
  while	
  the	
  performance	
  of	
  user	
  based	
  features	
  was	
  very	
  poor.	
  	
  
	
  
•  Our	
  results	
  provided	
  a	
  proof	
  of	
  concept	
  that,	
  automated	
  techniques	
  can	
  be	
  
used	
  in	
  iden$fying	
  real	
  images	
  from	
  fake	
  images	
  posted	
  on	
  TwiVer.	
  
	
  
	
  precog.iiitd.edu.in	
  

16	
  
Future Work
Fake	
  Charity	
  

Rumor	
  
	
  precog.iiitd.edu.in	
  

17	
  
Some Attractions

	
  precog.iiitd.edu.in	
  

18	
  
Some Attractions

	
  precog.iiitd.edu.in	
  

19	
  
Q&A

	
  precog.iiitd.edu.in	
  

20	
  
Thank You!!!

	

pk@iiitd.ac.in	
  
adi$g@iiitd.ac.in	
  
	
  
	
  
For any further information, please write to
pk@iiitd.ac.in
precog.iiitd.edu.in

22	
  

More Related Content

What's hot

Ordinary Influencers on Twitter
Ordinary Influencers on TwitterOrdinary Influencers on Twitter
Ordinary Influencers on Twitter
Winter Mason
 

What's hot (8)

Analyzing Real Time News
Analyzing Real Time NewsAnalyzing Real Time News
Analyzing Real Time News
 
Analytics For Switch To Airtel Kenya Campaign Safaricom Teaser
Analytics For Switch To Airtel Kenya Campaign Safaricom TeaserAnalytics For Switch To Airtel Kenya Campaign Safaricom Teaser
Analytics For Switch To Airtel Kenya Campaign Safaricom Teaser
 
Week 7.3 Semantic Attacks - Spear Phishing
Week 7.3 Semantic Attacks - Spear PhishingWeek 7.3 Semantic Attacks - Spear Phishing
Week 7.3 Semantic Attacks - Spear Phishing
 
Ordinary Influencers on Twitter
Ordinary Influencers on TwitterOrdinary Influencers on Twitter
Ordinary Influencers on Twitter
 
Twitter r t under crisis
Twitter r t under crisisTwitter r t under crisis
Twitter r t under crisis
 
#ArsonEmergency and Australia's "Black Summer": Polarisation and misinformati...
#ArsonEmergency and Australia's "Black Summer": Polarisation and misinformati...#ArsonEmergency and Australia's "Black Summer": Polarisation and misinformati...
#ArsonEmergency and Australia's "Black Summer": Polarisation and misinformati...
 
Twitter: Social Network Or News Medium?
Twitter: Social Network Or News Medium?Twitter: Social Network Or News Medium?
Twitter: Social Network Or News Medium?
 
Google analytics magic tricks - annotated v2.0 - Measurefest 2017
Google analytics magic tricks - annotated v2.0 -  Measurefest 2017 Google analytics magic tricks - annotated v2.0 -  Measurefest 2017
Google analytics magic tricks - annotated v2.0 - Measurefest 2017
 

Viewers also liked

$1.00 per RT #BostonMarathon #PrayForBoston: Analyzing Fake Content on Twitter
$1.00 per RT #BostonMarathon #PrayForBoston: Analyzing Fake Content on Twitter$1.00 per RT #BostonMarathon #PrayForBoston: Analyzing Fake Content on Twitter
$1.00 per RT #BostonMarathon #PrayForBoston: Analyzing Fake Content on Twitter
IIIT Hyderabad
 
Analyzing Social and Stylometric Features to Identify Spear phishing Emails
Analyzing Social and Stylometric Features to Identify Spear phishing EmailsAnalyzing Social and Stylometric Features to Identify Spear phishing Emails
Analyzing Social and Stylometric Features to Identify Spear phishing Emails
IIIT Hyderabad
 
Credibility Ranking of Tweets during High Impact Events
Credibility Ranking of Tweets during High Impact EventsCredibility Ranking of Tweets during High Impact Events
Credibility Ranking of Tweets during High Impact Events
IIIT Hyderabad
 
Emerging Phishing Trends and Effectiveness of the Anti-Phishing Landing Page
Emerging Phishing Trends and Effectiveness of the Anti-Phishing Landing PageEmerging Phishing Trends and Effectiveness of the Anti-Phishing Landing Page
Emerging Phishing Trends and Effectiveness of the Anti-Phishing Landing Page
IIIT Hyderabad
 
Keynote at 4th International Symposium on Secuirty in Computing at Communicat...
Keynote at 4th International Symposium on Secuirty in Computing at Communicat...Keynote at 4th International Symposium on Secuirty in Computing at Communicat...
Keynote at 4th International Symposium on Secuirty in Computing at Communicat...
IIIT Hyderabad
 
Studying user footprints in different online social networks
Studying user footprints in different online social networksStudying user footprints in different online social networks
Studying user footprints in different online social networks
IIIT Hyderabad
 

Viewers also liked (18)

Stop-Think-Connect: Past, present, and future. APWG Bern Symposium
Stop-Think-Connect: Past, present, and future. APWG Bern SymposiumStop-Think-Connect: Past, present, and future. APWG Bern Symposium
Stop-Think-Connect: Past, present, and future. APWG Bern Symposium
 
$1.00 per RT #BostonMarathon #PrayForBoston: Analyzing Fake Content on Twitter
$1.00 per RT #BostonMarathon #PrayForBoston: Analyzing Fake Content on Twitter$1.00 per RT #BostonMarathon #PrayForBoston: Analyzing Fake Content on Twitter
$1.00 per RT #BostonMarathon #PrayForBoston: Analyzing Fake Content on Twitter
 
IGDTUW workshop
IGDTUW workshopIGDTUW workshop
IGDTUW workshop
 
Analyzing Social and Stylometric Features to Identify Spear phishing Emails
Analyzing Social and Stylometric Features to Identify Spear phishing EmailsAnalyzing Social and Stylometric Features to Identify Spear phishing Emails
Analyzing Social and Stylometric Features to Identify Spear phishing Emails
 
Credibility Ranking of Tweets during High Impact Events
Credibility Ranking of Tweets during High Impact EventsCredibility Ranking of Tweets during High Impact Events
Credibility Ranking of Tweets during High Impact Events
 
Emerging Phishing Trends and Effectiveness of the Anti-Phishing Landing Page
Emerging Phishing Trends and Effectiveness of the Anti-Phishing Landing PageEmerging Phishing Trends and Effectiveness of the Anti-Phishing Landing Page
Emerging Phishing Trends and Effectiveness of the Anti-Phishing Landing Page
 
Mitigating Misinformation Spread on Micro-blogging Web Services using TweetCr...
Mitigating Misinformation Spread on Micro-blogging Web Services using TweetCr...Mitigating Misinformation Spread on Micro-blogging Web Services using TweetCr...
Mitigating Misinformation Spread on Micro-blogging Web Services using TweetCr...
 
Keynote at 4th International Symposium on Secuirty in Computing at Communicat...
Keynote at 4th International Symposium on Secuirty in Computing at Communicat...Keynote at 4th International Symposium on Secuirty in Computing at Communicat...
Keynote at 4th International Symposium on Secuirty in Computing at Communicat...
 
Digital Forces - Social: Future Trends, Student Projects Highlight, Software ...
Digital Forces - Social: Future Trends, Student Projects Highlight, Software ...Digital Forces - Social: Future Trends, Student Projects Highlight, Software ...
Digital Forces - Social: Future Trends, Student Projects Highlight, Software ...
 
Designing and Evaluating Techniques to
 Mitigate Misinformation Spread on 
Mi...
Designing and Evaluating Techniques to
 Mitigate Misinformation Spread on 
Mi...Designing and Evaluating Techniques to
 Mitigate Misinformation Spread on 
Mi...
Designing and Evaluating Techniques to
 Mitigate Misinformation Spread on 
Mi...
 
Studying user footprints in different online social networks
Studying user footprints in different online social networksStudying user footprints in different online social networks
Studying user footprints in different online social networks
 
Week 7.1 Link Farming
Week 7.1 Link FarmingWeek 7.1 Link Farming
Week 7.1 Link Farming
 
Week 6.2: eCrime
Week 6.2: eCrimeWeek 6.2: eCrime
Week 6.2: eCrime
 
Week 6.1: eCrime
Week 6.1: eCrimeWeek 6.1: eCrime
Week 6.1: eCrime
 
Week 8.2 Anonymous Networks
Week 8.2 Anonymous NetworksWeek 8.2 Anonymous Networks
Week 8.2 Anonymous Networks
 
Week 8.1 Profile Linking on Online Social Media
Week 8.1 Profile Linking on Online Social MediaWeek 8.1 Profile Linking on Online Social Media
Week 8.1 Profile Linking on Online Social Media
 
ICWSM 2016 paper presentation, Megha Arora
ICWSM 2016 paper presentation, Megha AroraICWSM 2016 paper presentation, Megha Arora
ICWSM 2016 paper presentation, Megha Arora
 
Credibility, Identity Resolution, Privacy, and Policing in Online Social Media
Credibility, Identity Resolution, Privacy, and Policing in Online Social MediaCredibility, Identity Resolution, Privacy, and Policing in Online Social Media
Credibility, Identity Resolution, Privacy, and Policing in Online Social Media
 

Similar to Faking Sandy: Characterizing and Identifying Fake Images on Twitter during Hurricane Sandy

TweetCred: Real-Time Credibility Assessment of 
 Content on Twitter @ Socinfo...
TweetCred: Real-Time Credibility Assessment of 
 Content on Twitter @ Socinfo...TweetCred: Real-Time Credibility Assessment of 
 Content on Twitter @ Socinfo...
TweetCred: Real-Time Credibility Assessment of 
 Content on Twitter @ Socinfo...
IIIT Hyderabad
 
Integrating New & Traditional Media Relations to Strengthen Corporate Reputat...
Integrating New & Traditional Media Relations to Strengthen Corporate Reputat...Integrating New & Traditional Media Relations to Strengthen Corporate Reputat...
Integrating New & Traditional Media Relations to Strengthen Corporate Reputat...
The Hoffman Agency Asia Pacific
 

Similar to Faking Sandy: Characterizing and Identifying Fake Images on Twitter during Hurricane Sandy (20)

Privacy and Security on Online Social Media: Workshop on Data Analytics & Its...
Privacy and Security on Online Social Media: Workshop on Data Analytics & Its...Privacy and Security on Online Social Media: Workshop on Data Analytics & Its...
Privacy and Security on Online Social Media: Workshop on Data Analytics & Its...
 
Privacy and Security in Online Social Media : Trust and Credebillity on OSM
Privacy and Security in Online Social Media : Trust and Credebillity on OSMPrivacy and Security in Online Social Media : Trust and Credebillity on OSM
Privacy and Security in Online Social Media : Trust and Credebillity on OSM
 
Word embedding for detecting cyberbullying based on recurrent neural networks
Word embedding for detecting cyberbullying based on recurrent neural networksWord embedding for detecting cyberbullying based on recurrent neural networks
Word embedding for detecting cyberbullying based on recurrent neural networks
 
TweetCred: Real-Time Credibility Assessment of 
 Content on Twitter @ Socinfo...
TweetCred: Real-Time Credibility Assessment of 
 Content on Twitter @ Socinfo...TweetCred: Real-Time Credibility Assessment of 
 Content on Twitter @ Socinfo...
TweetCred: Real-Time Credibility Assessment of 
 Content on Twitter @ Socinfo...
 
Computational Verification Challenges in Social Media
Computational Verification Challenges in Social MediaComputational Verification Challenges in Social Media
Computational Verification Challenges in Social Media
 
Bullet PR; Social Media, You Tube & Search Engine Visibility, Presentation At...
Bullet PR; Social Media, You Tube & Search Engine Visibility, Presentation At...Bullet PR; Social Media, You Tube & Search Engine Visibility, Presentation At...
Bullet PR; Social Media, You Tube & Search Engine Visibility, Presentation At...
 
IRJET- An Experimental Evaluation of Mechanical Properties of Bamboo Fiber Re...
IRJET- An Experimental Evaluation of Mechanical Properties of Bamboo Fiber Re...IRJET- An Experimental Evaluation of Mechanical Properties of Bamboo Fiber Re...
IRJET- An Experimental Evaluation of Mechanical Properties of Bamboo Fiber Re...
 
IRJET- Tweet Segmentation and its Application to Named Entity Recognition
IRJET- Tweet Segmentation and its Application to Named Entity RecognitionIRJET- Tweet Segmentation and its Application to Named Entity Recognition
IRJET- Tweet Segmentation and its Application to Named Entity Recognition
 
Will It Blend – Marketing & Public Relations – Integrated Marketing Summit Da...
Will It Blend – Marketing & Public Relations – Integrated Marketing Summit Da...Will It Blend – Marketing & Public Relations – Integrated Marketing Summit Da...
Will It Blend – Marketing & Public Relations – Integrated Marketing Summit Da...
 
Twitter analytics: some thoughts on sampling, tools, data, ethics and user re...
Twitter analytics: some thoughts on sampling, tools, data, ethics and user re...Twitter analytics: some thoughts on sampling, tools, data, ethics and user re...
Twitter analytics: some thoughts on sampling, tools, data, ethics and user re...
 
Leveraging Your Brand In A Digital World – PRSA St. Louis Tech Day Nov. 5, 2010
Leveraging Your Brand In A Digital World – PRSA St. Louis Tech Day Nov. 5, 2010Leveraging Your Brand In A Digital World – PRSA St. Louis Tech Day Nov. 5, 2010
Leveraging Your Brand In A Digital World – PRSA St. Louis Tech Day Nov. 5, 2010
 
Art of Storytelling Marcom Beijing 2010
Art of Storytelling Marcom Beijing 2010Art of Storytelling Marcom Beijing 2010
Art of Storytelling Marcom Beijing 2010
 
INFERENCE ATTACK ON BROWSING HISTORY OF TWITTER USERS USING PUBLIC CLICK ANAL...
INFERENCE ATTACK ON BROWSING HISTORY OF TWITTER USERS USING PUBLIC CLICK ANAL...INFERENCE ATTACK ON BROWSING HISTORY OF TWITTER USERS USING PUBLIC CLICK ANAL...
INFERENCE ATTACK ON BROWSING HISTORY OF TWITTER USERS USING PUBLIC CLICK ANAL...
 
Integrating New & Traditional Media Relations to Strengthen Corporate Reputat...
Integrating New & Traditional Media Relations to Strengthen Corporate Reputat...Integrating New & Traditional Media Relations to Strengthen Corporate Reputat...
Integrating New & Traditional Media Relations to Strengthen Corporate Reputat...
 
The War on Attention Poverty: Measuring Twitter Authority
The War on Attention Poverty: Measuring Twitter AuthorityThe War on Attention Poverty: Measuring Twitter Authority
The War on Attention Poverty: Measuring Twitter Authority
 
Forget What You Think You Know About Public Relations – It’s a Whole New Worl...
Forget What You Think You Know About Public Relations – It’s a Whole New Worl...Forget What You Think You Know About Public Relations – It’s a Whole New Worl...
Forget What You Think You Know About Public Relations – It’s a Whole New Worl...
 
Social Media @Home and @Work: Understanding Who Is Using and Why
Social Media @Home and @Work:Understanding Who Is Using and WhySocial Media @Home and @Work:Understanding Who Is Using and Why
Social Media @Home and @Work: Understanding Who Is Using and Why
 
America Saves Week 2014 eXtension Mini Grant Project-Final Report-03-14
America Saves Week 2014 eXtension Mini Grant Project-Final Report-03-14America Saves Week 2014 eXtension Mini Grant Project-Final Report-03-14
America Saves Week 2014 eXtension Mini Grant Project-Final Report-03-14
 
Detection of Cyberbullying on Social Media using Machine Learning
Detection of Cyberbullying on Social Media using Machine LearningDetection of Cyberbullying on Social Media using Machine Learning
Detection of Cyberbullying on Social Media using Machine Learning
 
Master defence 2020 - Andrew Kurochkin - Meme Generation for Social Media Aud...
Master defence 2020 - Andrew Kurochkin - Meme Generation for Social Media Aud...Master defence 2020 - Andrew Kurochkin - Meme Generation for Social Media Aud...
Master defence 2020 - Andrew Kurochkin - Meme Generation for Social Media Aud...
 

More from IIIT Hyderabad

Identify, Inspect and Intervene Multimodal Fake News
Identify, Inspect and Intervene Multimodal Fake NewsIdentify, Inspect and Intervene Multimodal Fake News
Identify, Inspect and Intervene Multimodal Fake News
IIIT Hyderabad
 
Beyond the Surface: A Computational Exploration of Linguistic Ambiguity
Beyond the Surface: A Computational Exploration of Linguistic AmbiguityBeyond the Surface: A Computational Exploration of Linguistic Ambiguity
Beyond the Surface: A Computational Exploration of Linguistic Ambiguity
IIIT Hyderabad
 
Modeling Online User Interactions and their Offline effects on Socio-Technica...
Modeling Online User Interactions and their Offline effects on Socio-Technica...Modeling Online User Interactions and their Offline effects on Socio-Technica...
Modeling Online User Interactions and their Offline effects on Socio-Technica...
IIIT Hyderabad
 
Development of Stress Induction and Detection System to Study its Effect on B...
Development of Stress Induction and Detection System to Study its Effect on B...Development of Stress Induction and Detection System to Study its Effect on B...
Development of Stress Induction and Detection System to Study its Effect on B...
IIIT Hyderabad
 
A Framework for Automatic Question Answering in Indian Languages
A Framework for Automatic Question Answering in Indian LanguagesA Framework for Automatic Question Answering in Indian Languages
A Framework for Automatic Question Answering in Indian Languages
IIIT Hyderabad
 

More from IIIT Hyderabad (20)

Responsible & Safe AI Systems at ACM India ROCS at IIT Bombay
Responsible & Safe AI Systems at ACM India ROCS at IIT BombayResponsible & Safe AI Systems at ACM India ROCS at IIT Bombay
Responsible & Safe AI Systems at ACM India ROCS at IIT Bombay
 
International Collaboration: Experiences, Challenges, Success stories
International Collaboration: Experiences, Challenges, Success storiesInternational Collaboration: Experiences, Challenges, Success stories
International Collaboration: Experiences, Challenges, Success stories
 
Responsible & Safe AI: #LegalBias #Inconsistency #BiasinLLMs #MultiModalBias
Responsible & Safe AI: #LegalBias #Inconsistency #BiasinLLMs #MultiModalBiasResponsible & Safe AI: #LegalBias #Inconsistency #BiasinLLMs #MultiModalBias
Responsible & Safe AI: #LegalBias #Inconsistency #BiasinLLMs #MultiModalBias
 
Identify, Inspect and Intervene Multimodal Fake News
Identify, Inspect and Intervene Multimodal Fake NewsIdentify, Inspect and Intervene Multimodal Fake News
Identify, Inspect and Intervene Multimodal Fake News
 
#ChatGPT #ResponsibleAI
#ChatGPT #ResponsibleAI#ChatGPT #ResponsibleAI
#ChatGPT #ResponsibleAI
 
Data Science for Social Good: #MentalHealth #CodeMix #LegalNLP #AISafety
Data Science for Social Good: #MentalHealth #CodeMix #LegalNLP #AISafetyData Science for Social Good: #MentalHealth #CodeMix #LegalNLP #AISafety
Data Science for Social Good: #MentalHealth #CodeMix #LegalNLP #AISafety
 
It is our choices, Harry, that show what we truly are, far more than our abil...
It is our choices, Harry, that show what we truly are, far more than our abil...It is our choices, Harry, that show what we truly are, far more than our abil...
It is our choices, Harry, that show what we truly are, far more than our abil...
 
Beyond the Surface: A Computational Exploration of Linguistic Ambiguity
Beyond the Surface: A Computational Exploration of Linguistic AmbiguityBeyond the Surface: A Computational Exploration of Linguistic Ambiguity
Beyond the Surface: A Computational Exploration of Linguistic Ambiguity
 
Data Science for Social Good: #LegalNLP #AlgorithmicBias...
Data Science for Social Good:                      #LegalNLP #AlgorithmicBias...Data Science for Social Good:                      #LegalNLP #AlgorithmicBias...
Data Science for Social Good: #LegalNLP #AlgorithmicBias...
 
How to Write a (Good) Research Paper
How to Write a (Good) Research Paper How to Write a (Good) Research Paper
How to Write a (Good) Research Paper
 
Data Science for Social Good: #LegalNLP #AlgorithmicBias
Data Science for Social Good: #LegalNLP #AlgorithmicBiasData Science for Social Good: #LegalNLP #AlgorithmicBias
Data Science for Social Good: #LegalNLP #AlgorithmicBias
 
Social Computing Research in India
Social Computing Research in IndiaSocial Computing Research in India
Social Computing Research in India
 
Social Computing Research in India
Social Computing Research in IndiaSocial Computing Research in India
Social Computing Research in India
 
Modeling Online User Interactions and their Offline effects on Socio-Technica...
Modeling Online User Interactions and their Offline effects on Socio-Technica...Modeling Online User Interactions and their Offline effects on Socio-Technica...
Modeling Online User Interactions and their Offline effects on Socio-Technica...
 
Privacy. Winter School on “Topics in Digital Trust”. IIT Bombay
Privacy. Winter School on “Topics in Digital Trust”. IIT BombayPrivacy. Winter School on “Topics in Digital Trust”. IIT Bombay
Privacy. Winter School on “Topics in Digital Trust”. IIT Bombay
 
It is our choices, Harry, that show what we truly are, far more than our abil...
It is our choices, Harry, that show what we truly are, far more than our abil...It is our choices, Harry, that show what we truly are, far more than our abil...
It is our choices, Harry, that show what we truly are, far more than our abil...
 
It is our choices, Harry, that show what we truly are, far more than our abil...
It is our choices, Harry, that show what we truly are, far more than our abil...It is our choices, Harry, that show what we truly are, far more than our abil...
It is our choices, Harry, that show what we truly are, far more than our abil...
 
Leveraging Social Media for Financial Advice
Leveraging Social Media for Financial AdviceLeveraging Social Media for Financial Advice
Leveraging Social Media for Financial Advice
 
Development of Stress Induction and Detection System to Study its Effect on B...
Development of Stress Induction and Detection System to Study its Effect on B...Development of Stress Induction and Detection System to Study its Effect on B...
Development of Stress Induction and Detection System to Study its Effect on B...
 
A Framework for Automatic Question Answering in Indian Languages
A Framework for Automatic Question Answering in Indian LanguagesA Framework for Automatic Question Answering in Indian Languages
A Framework for Automatic Question Answering in Indian Languages
 

Recently uploaded

Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
WSO2
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
?#DUbAI#??##{{(☎️+971_581248768%)**%*]'#abortion pills for sale in dubai@
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
panagenda
 

Recently uploaded (20)

MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024MINDCTI Revenue Release Quarter One 2024
MINDCTI Revenue Release Quarter One 2024
 
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ..."I see eyes in my soup": How Delivery Hero implemented the safety system for ...
"I see eyes in my soup": How Delivery Hero implemented the safety system for ...
 
Real Time Object Detection Using Open CV
Real Time Object Detection Using Open CVReal Time Object Detection Using Open CV
Real Time Object Detection Using Open CV
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost SavingRepurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
Repurposing LNG terminals for Hydrogen Ammonia: Feasibility and Cost Saving
 
Automating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps ScriptAutomating Google Workspace (GWS) & more with Apps Script
Automating Google Workspace (GWS) & more with Apps Script
 
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
Apidays New York 2024 - Accelerating FinTech Innovation by Vasa Krishnan, Fin...
 
ICT role in 21st century education and its challenges
ICT role in 21st century education and its challengesICT role in 21st century education and its challenges
ICT role in 21st century education and its challenges
 
Architecting Cloud Native Applications
Architecting Cloud Native ApplicationsArchitecting Cloud Native Applications
Architecting Cloud Native Applications
 
A Beginners Guide to Building a RAG App Using Open Source Milvus
A Beginners Guide to Building a RAG App Using Open Source MilvusA Beginners Guide to Building a RAG App Using Open Source Milvus
A Beginners Guide to Building a RAG App Using Open Source Milvus
 
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
+971581248768>> SAFE AND ORIGINAL ABORTION PILLS FOR SALE IN DUBAI AND ABUDHA...
 
presentation ICT roal in 21st century education
presentation ICT roal in 21st century educationpresentation ICT roal in 21st century education
presentation ICT roal in 21st century education
 
Ransomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdfRansomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdf
 
How to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected WorkerHow to Troubleshoot Apps for the Modern Connected Worker
How to Troubleshoot Apps for the Modern Connected Worker
 
Data Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt RobisonData Cloud, More than a CDP by Matt Robison
Data Cloud, More than a CDP by Matt Robison
 
Why Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire businessWhy Teams call analytics are critical to your entire business
Why Teams call analytics are critical to your entire business
 
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, AdobeApidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
Apidays New York 2024 - Scaling API-first by Ian Reasor and Radu Cotescu, Adobe
 
Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024
 
A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?A Year of the Servo Reboot: Where Are We Now?
A Year of the Servo Reboot: Where Are We Now?
 
Apidays Singapore 2024 - Scalable LLM APIs for AI and Generative AI Applicati...
Apidays Singapore 2024 - Scalable LLM APIs for AI and Generative AI Applicati...Apidays Singapore 2024 - Scalable LLM APIs for AI and Generative AI Applicati...
Apidays Singapore 2024 - Scalable LLM APIs for AI and Generative AI Applicati...
 

Faking Sandy: Characterizing and Identifying Fake Images on Twitter during Hurricane Sandy

  • 1. Faking Sandy: Characterizing and Identifying Fake Images on Twitter during Hurricane Sandy Adi$  Gupta   Hemank  Lamba   Ponnurangam  Kumaraguru   Anupam  Joshi    
  • 2. Agenda  Background    Mo$va$on    Main  contribu$ons    Data  descrip$on    Methodology    Results    Future  work    precog.iiitd.edu.in   2  
  • 3. Background: Hurricane Sandy  Dates:  Oct  22-­‐  31,  2012    Category  3  storm    Damages  worth  $75  billion    Coast  of  NE  America    precog.iiitd.edu.in   3  
  • 4. Motivation FAKE   IMAGES    precog.iiitd.edu.in   4  
  • 6. Our Contributions   Performed  in-­‐depth  characteriza$on  of  tweets  sharing   fake  images  on  TwiVer  during  Hurricane  Sandy   -  Tweets  containing  the  fake  images  URLs  were  mostly   retweets  (86%)   -  Just  11%  overlap  between  the  retweet  and  follower  graphs  of   tweets  containing  fake  images.         Applied  classifica$on  algorithms  to  dis$nguish  between   tweets  containing  fake  and  real  images.     -  Best  accuracy  of  97%  was  achieved  using  decision  tree   classifier,  using  tweet  based  features.      precog.iiitd.edu.in   6  
  • 7. Methodology Feature   Genera$on   Data  Collec$on   and  Filtering   Evalua$ng   Results   Data   Characteriza$on   Obtaining   Ground  Truth   Classifica$on  Module    precog.iiitd.edu.in   7  
  • 8. Data Description Total  tweets   Total  unique  users   1,174,266   Tweets  with  URLs    precog.iiitd.edu.in   1,782,526   622,860   8  
  • 9. Data Filtering  Reputable  online  resource  to  filter  fake  and  real   images   - Guardian  collected  and  publically  distributed  a  list   of  fake  and  true  images  shared  during  Hurricane   Sandy       Tweets  with  fake  images   10,350     Users  with  fake  images   10,215       Tweets  with  real  images   5,767       Users  with  real  images   5,678      One  of  the  biggest  fake  content  propaga$on   datasets  that  have  been  studied  by  researchers    precog.iiitd.edu.in   9  
  • 10. Characterization – Fake Image Propagation  86%  of  tweets  spreading  the  fake  images  were   retweets    Top  30  users  out  of  10,215  users  (0.3%)  resulted   in  90%  of  the  retweets  of  fake  images    precog.iiitd.edu.in   10  
  • 11. Network Analysis Tweet  –  Retweet  graph  for  the  spread  of   fake  images  at  ‘nth’  and  ‘n+1th‘  hour    precog.iiitd.edu.in   11  
  • 12. Role of Explicit Twitter Network  Analyzing  role  of  follower  network  in  fake   image  propaga$on    We  crawled  the  TwiVer  network  for  all  users   who  tweeted  the  fake  image  URLs    precog.iiitd.edu.in   12  
  • 14. Results Total  edges  in  the  retweet  network   Total  edges  in  the  follower-­‐followee  network   Total  edges  that  exist  in  both  retweet   network  and  the  follower-­‐  followee  network   %age  Overlap    precog.iiitd.edu.in   10,508   10,799,122   1,215   11%   14  
  • 15. Classification  5  fold  cross  valida$on   User  Features  [F1]   Number  of  Friends   Number  of  Followers   Follower-­‐Friend  Ra$o   Number  of  $mes  listed   User  has  a  URL   User  is  a  verified  user   Age  of  user  account    precog.iiitd.edu.in   Tweet  Features  [F2]   Length  of  Tweet   Number  of  Words   Contains  Ques$on  Mark?   Contains  Exclama$on  Mark?   Number  of  Ques$on  Marks   Number  of  Exclama$on  Marks   Contains  Happy  Emo$con   Contains  Sad  Emo$con   Contains  First  Order  Pronoun   Contains  Second  Order  Pronoun   Contains  Third  Order  Pronoun   Number  of  uppercase  characters   Number  of  nega$ve  sen$ment  words   Number  of  posi$ve  sen$ment  words   Number  of  men$ons   Number  of  hashtags   Number  of  URLs   Retweet  count   15  
  • 16. Classification Results F1  (user)   F2  (tweet)   F1+F2   Naïve  Bayes   56.32%   91.97%   91.52%   Decision  Tree   53.24%   97.65%   96.65%   •  Best  results  were  obtained  from  Decision  Tree  classifier,  we  got  97%   accuracy  in  predic$ng  fake  images  from  real.     •  Tweet  based  features  are  very  effec$ve  in  dis$nguishing  fake  images  tweets   from  real,  while  the  performance  of  user  based  features  was  very  poor.       •  Our  results  provided  a  proof  of  concept  that,  automated  techniques  can  be   used  in  iden$fying  real  images  from  fake  images  posted  on  TwiVer.      precog.iiitd.edu.in   16  
  • 17. Future Work Fake  Charity   Rumor    precog.iiitd.edu.in   17  
  • 22. For any further information, please write to pk@iiitd.ac.in precog.iiitd.edu.in 22