mitigating misinformation spread on micro-blogging web services using tweetcred

Post on 12-Apr-2017

385 Views

Category:

Technology

0 Downloads

Preview:

Click to see full reader

TRANSCRIPT

Mitigating Misinformation Spread on Micro-blogging Web Services using

TweetCred

CENS  Workshop  on  Social  Media  in  Communica7on,  Governance  and  Security  

Nov  6,  2015    

Ponnurangam  Kumaraguru  (“PK”)  Associate  Professor,  Founding  Head  

Cybersecurity  Educa7on  and  Research  Centre  (CERC)  O/ponnurangam.kumaraguru,  @ponguru  

Who am I? � Associate  Professor,  IIIT-­‐Delhi      � Ph.D.  from  School  of  Computer  Science,          Carnegie  Mellon  University  (CMU)      � Research  interests    - Privacy,  e-­‐crime,  online  social  media,  and  usable  security    

� Founding  Head,  CERC@IIITD    � Co-­‐ordinate  and  manage  Precog,  precog.iiitd.edu.in    

�  I  conduct  Government  /  Intelligence  /  Police  focused  training  programs  on  OSM  

2  

Harvard (1839) – Harvard – Harvard – Harvard – MIT – Northwestern – UIUC – WUSL – CMU (2009) – IIITD (2015)

3  

Real World Events

4  

Misinformation on Social Media

5  

Misinformation on Social Media

6  

Misinformation on Social Media

7  

Misinformation Tweets

FAKE  

RUMORS  

8  

$  

Temporal Patterns

9  

Fake  content  /  rumors  becomes  viral  in  first  7-­‐8  hours  just  a`er  the  event.      

Sample Fake Tweets

10  

>  50,000  RTs  

>  30,000  RTs  

Architecture

11  

Features for Real-time Analysis

12  

Feature  set      Features  (45)    

Tweet  meta-­‐data     Number  of  seconds  since  the  tweet;  Source  of  tweet  (mobile  /  web/  etc);  Tweet  contains  geo-­‐coordinates  

Tweet  content  (simple)    

Number  of  characters;  Number  of  words;  Number  of  URLs;  Number  of  hashtags;  Number  of  unique  characters;  Presence  of  stock  symbol;  Presence  of  happy  smiley;  Presence  of  sad  smiley;  Tweet  contains  `via';  Presence  of  colon  symbol  

Tweet  content  (linguis7c)    Presence  of  swear  words;  Presence  of  nega7ve  emo7on  words;  Presence  of  posi7ve  emo7on  words;  Presence  of  pronouns;  Men7on  of  self  words  in  tweet  (I;  my;  mine)  

Tweet  author     Number  of  followers;  friends;  7me  since  the  user  if  on  Twiger;  etc.  

Tweet  network     Number  of  retweets;  Number  of  men7ons;  Tweet  is  a  reply;  Tweet  is  a  retweet  

Tweet  links     WOT  score  for  the  URL;  Ra7o  of  likes  /  dislikes  for  a  YouTube  video  

Implementation

Feedback by Users

14  

Usage Statistics

Date  of  launch  of  TweetCred    27  Apr,  2014  

Credibility  score  requests  received   14,234,131  

Unique  Twiger  users   1,808  

Feedback  was  given  for  tweets   1,654  

Unique  users  who  gave  feedback   364  

15  

*  Data  as  on  April’15.  hgp://precog.iiitd.edu.in/Publica7ons_files/socinfo_paper_102.pdf  

Users of TweetCred

Sample  users:  - Emergency  responders  - Firefighters  - Journalists  /  news  media  - General  users  - Researchers  (Requested  API  tokens)  

16  

v

Media / Popular reference

TweetCred

� Available  as  a  Chrome  Extension  � Rest  API  

hgps://chrome.google.com/webstore/detail/tweetcred/Ookljinlogeihdnkikeeneiankdgikg?hl=en  

Takeaways

� Technology  can  be  built  to  reduce  the  effect  of  rumors  /  misinforma7on  on  Online  Social  Media    

� Will  be  good  to  have  some  interna7onal  collabora7on  in  using  OSM  for  Govt.      - Given  that  I  am  in  Delhi,  will  be  happy  to  help  in  anyways  possible    

19  

Thank  you!      

pk@iiitd.ac.in  precog.iiitd.edu.in  

top related