1 Taming the long tail Identify Filtering in Social Media Avner May, Nitish Korula, Silvio Lattanzi A. Chaintreau Columbia University, Google
2
3 Are social media sustainable?
4 From the trenches: no! ¡ Users’s dilemma - May I be missing something? ¡ Content producters - May I be missing my audience?
5 From the faculty lounge: of course! ¡ Socializing is essential for information - To find about jobs [Gr74] , innovation [CKM57] “It pays to know / It hurts to be unaware.” ¡ When looking for good content, most of the time is wasted, but some gems are priceless - This process is more efficient collectively - And curating is at least informally rewarded ¡ In this talk, we focus on news dissemination
6 What is the role of intermediaries?
7 Understanding these intermediaries
8 2013: two interesting works Twitter “precision” 40.5% average - Encouraging! got a low relevance score for thi user 15 random twee t at rando planation of the high precision we see in this tria ear that inspection paradox 2 alone coul t ) s t e e w t ) o f ) n o i s c i e r p d ) e t a r r . e s U Timeline" 1" Random" Homogeneous or 0.9" 0.8" 0.7" 0.6" ) n o 0.5" i s 10" i c 0.4" 9" e r P 8" 0.3" 7" 6" 0.2" 5" structured e 1: Comparison of self-reported precision b 4" User)number) 0.1" interests leads to 3" 0" 2" from a user’s timeline and tweets c 1" efficient networks ecision
9 Can we find evidence of filtering?
10 Looking for filtering Intermediaries URLs Posted Data Sets Source User URL s s NY Times Links Twitter 330k 33k Bin Laden Death Twitter 700k 545k Occupy Wall Twitter 354k 316k Street Steve Jobs Death Twitter 719k 251k iPhone 5 Launch Twitter 81k 37k iPhone 5 Launch Facebook 330k 193k All Spinn3r blogs Spinn3r 68k 441k Obama Spinn3r 13k 85k Facebook Spinn3r 12k 70k Euro Spinn3r 10k 53k Mubarak Spinn3r 7k 43k
11 Evidence of information filtering “Filtering law” Not an artefact of M - replacement O R E A C T - exposure I V I T Y à à L E S S P O P U L A R C O N T E N T
12 Many open questions ¡ Can we find more evidence of precision? - Using click (Twitter data grant, more partners) - Does selectivity correlate with success? ¡ Current models somewhat at odds - Discrete topics + continuous popularity range - Are there more general models ¡ Can crowd-curation be improved? - In principle (no friction etc.), already efficient. - With incentive? With new mechanism?
13 Thank you!
14 Back-Up Slides
15 Theoretical Results Audience Pure Strategy Price of Strategy Equilibrium? Anarchy Greedy No -- Satisficing Yes 2 Satisficing w/ Yes 2 blogger ability
16 Filtering Law Consistent Across Data Sets
17 INACTIVE ACTIVE < 2 / month < 2 /day 5% VERY 35% ACTIVE >= 2 / day 60% M O R E A C C O T N I V T E E N à à T L E S S P O P U L A R
Simply explained by replacement effect? 18 INACTIVE ACTIV VERY < 2 / month E ACTIVE 5% < 2 / >= 2 / day day 60% 35% M O R E A C C O T N I V T E E N à à T L E S S P O NO! P U L A R
19 In Summary … ¡ Previous work: Intermediaries play key role in information dissemination. ¡ We provided theoretical and empirical justification for intermediaries as information filters. ¡ Come see my poster! - Results not shown: Role of MORE ACTIVE à LESS filtering on success of POPULAR CONTENT intermediary
Recommend
More recommend
