Upgrade to Pro — share decks privately, control downloads, hide ads and more …

Do Cascades Recur?

Do Cascades Recur?

Cascades of information-sharing are a primary mechanism by which content reaches its audience on social media, and an active line of research has studied how such cascades, which form as content is reshared from person to person, develop and subside. In this paper, we perform a large-scale analysis of cascades on Facebook over significantly longer time scales, and find that a more complex picture emerges, in which many large cascades recur, exhibiting multiple bursts of popularity with periods of quiescence in between. We characterize recurrence by measuring the time elapsed between bursts, their overlap and proximity in the social network, and the diversity in the demographics of individuals participating in each peak. We discover that content virality, as revealed by its initial popularity, is a main driver of recurrence, with the availability of multiple copies of that content helping to spark new bursts. Still, beyond a certain popularity of content, the rate of recurrence drops as cascades start exhausting the population of interested individuals. We reproduce these observed patterns in a simple model of content recurrence simulated on a real social network. Using only characteristics of a cascade's initial burst, we demonstrate strong performance in predicting whether it will recur in the future.

Presented at WWW 2016.

Justin Cheng

April 15, 2016
Tweet

More Decks by Justin Cheng

Other Decks in Research

Transcript

  1. Do Cascades Recur?
    Justin Cheng, Lada Adamic, Jon Kleinberg, Jure Leskovec

    View Slide

  2. View Slide

  3. Several weeks later…

    View Slide

  4. View Slide

  5. View Slide

  6. View Slide

  7. View Slide

  8. View Slide

  9. View Slide

  10. View Slide

  11. View Slide

  12. Time
    Popularity
    Ahmed et al. (2013), Bauckhage et al. (2013), Matsubara et al. (2012), Yang & Counts (2010)
    Prior work: cascades rise, then fall

    View Slide

  13. Mar Jun Sept Dec
    Cascades rise and fall many times
    Popularity

    View Slide

  14. Mar Jun Sept Dec
    Cascades rise and fall many times
    Popularity

    View Slide

  15. Mar Jun Sept Dec
    Cascades consist of multiple copies
    Popularity

    View Slide

  16. Mar Jun Sept Dec
    Cascades consist of multiple copies
    Popularity

    View Slide

  17. Mar Jun Sept Dec
    Cascades consist of multiple copies
    Popularity

    View Slide

  18. Cascades are complex
    Mar Jun Sept Dec
    Popularity

    View Slide

  19. Cascades are complex
    Time

    View Slide

  20. Individual copies recur
    Time

    View Slide

  21. Studying recurrence on Facebook
    5 billion reshares 100 million photos 76 thousand clusters
    (Sampled over all of 2014 and de-identified)

    View Slide

  22. 2 out of 5 image memes
    on Facebook recur.

    View Slide

  23. 1 out of 3 videos
    on Facebook recur.

    View Slide

  24. Cascades recur.

    View Slide

  25. Cascades recur.
    Why do cascades
    recur?
    How do cascades
    recur?
    What is recurrence?

    View Slide

  26. What is recurrence?

    View Slide

  27. Defining recurrence
    Popularity r
    Time t

    View Slide

  28. Defining recurrence
    p1
    p4
    p2
    Popularity r
    Time t
    p3
    (?)

    View Slide

  29. Defining recurrence
    Popularity r
    Peaks have a minimum
    absolute/relative height
    rp
    ≥ h0,
    rp
    ≥ m·r
    *Not a peak
    Time t
    r
    p4
    p1
    p4
    p2
    pi
    pi
    rp
    p1
    rp
    p2
    rp
    p4

    View Slide

  30. Defining recurrence
    Popularity r
    Peaks are local maxima
    rp
    ≥ max { rj
    | pi
    -w ≤ j ≤ pi
    +w }
    Time t
    p1
    p4
    p2
    pi
    rp
    p1

    View Slide

  31. Defining recurrence
    Popularity r
    Valleys separate peaks
    rp
    , rp
    ≥ v · max { rj
    | pi
    < j < pi+1
    }
    Time t
    p1
    p4
    p2
    pi
    pi+1
    rp
    p1

    View Slide

  32. Defining recurrence
    Popularity r
    Time t
    p1
    p4
    p2
    Recurrences

    View Slide

  33. Defining recurrence
    Popularity r
    Time t
    b1
    b2
    b4
    Recurrences

    View Slide

  34. Recurrence is common
    Number of Bursts
    Empirical CCDF
    0.00
    0.25
    0.50
    0.75
    1.00
    0 5 10 15

    View Slide

  35. 0.00
    0.25
    0.50
    0.75
    1.00
    0 5 10 15
    Recurrence is common
    Number of Bursts
    Empirical CCDF
    40% of cascades recur
    0.40
    2

    View Slide

  36. 0.00
    0.25
    0.50
    0.75
    1.00
    0 5 10 15
    Recurrence is common
    Number of Bursts
    Empirical CCDF
    Cascades have 2.3
    bursts on average
    2.3

    View Slide

  37. Recurrence takes time
    Days Between 1st and 2nd Burst
    Empirical CCDF
    0.00
    0.25
    0.50
    0.75
    1.00
    0 50 100 150 200

    View Slide

  38. 0.00
    0.25
    0.50
    0.75
    1.00
    0 50 100 150 200
    Recurrence takes time
    Days Between 1st and 2nd Burst
    Empirical CCDF
    A meme takes an average
    of 32 days to recur
    32

    View Slide

  39. What is recurrence?
    A cascade recurs when it peaks in popularity more than once.
    Recurrence is common.
    Recurrence takes time.

    View Slide

  40. How do cascades recur?

    View Slide

  41. Do bursts occur in different parts
    of the network?

    View Slide

  42. Bursts are (somewhat) separated
    Time
    Popularity

    View Slide

  43. Bursts are (somewhat) separated
    Time
    Popularity

    View Slide

  44. Bursts are (somewhat) separated
    Time
    Popularity

    View Slide

  45. Bursts are (somewhat) separated
    Time
    Popularity
    3.2 connections within bursts

    View Slide

  46. Bursts are (somewhat) separated
    Time
    Popularity
    1.4 connections across bursts

    View Slide

  47. Are more popular cascades
    more likely to recur?

    View Slide

  48. Moderate popularity increases recurrence
    Size of Initial Burst
    Mean Bursts
    1.5
    2.0
    2.5
    3.0
    3.5
    4.0
    103 104 105 106

    View Slide

  49. Large initial bursts exhaust susceptibles
    Size of Initial Burst
    Proportion Exposed in Second Burst
    0.2
    0.4
    0.6
    0.8
    103 104 105 106

    View Slide

  50. Are cascades with more diverse
    populations more likely to recur?

    View Slide

  51. Moderate diversity increases recurrence
    Country Entropy in Initial Burst
    Mean Bursts
    Gender Entropy in Initial Burst
    2.0
    2.5
    3.0
    3.5
    0 1 2 3 4 5
    1.5
    2.0
    2.5
    3.0
    0.2 0.4 0.6 0.8 1.0

    View Slide

  52. Do new copies of the same
    meme spark recurrence?

    View Slide

  53. New copies can spark recurrence…
    Recurring cascades are
    spread out over more copies
    Recurring
    Non-recurring
    New copies correlate
    with recurrence
    r = 0.66
    Introduction of new copies &
    number of reshares
    4x

    View Slide

  54. …but copies aren’t the only cause!
    Individual copies
    also recur
    18%
    of individual copies recur
    Copies can be traced
    back to other copies
    75%
    of copies can be attributed

    via the friendship graph

    View Slide

  55. How do cascades recur?
    Bursts happen in separate parts of the network.
    Moderately viral/diverse content tends to recur.
    New copies can spark recurrence.

    View Slide

  56. Why do cascades recur?

    View Slide

  57. A model of recurrence

    View Slide

  58. A model of recurrence
    1

    View Slide

  59. A model of recurrence
    1
    2

    View Slide

  60. Low virality
    1
    2

    View Slide

  61. Low virality
    1
    2

    View Slide

  62. High virality
    1
    2

    View Slide

  63. High virality
    1
    2

    View Slide

  64. Moderate virality
    1
    2

    View Slide

  65. Moderate virality
    1
    2

    View Slide

  66. Moderate virality
    1
    2

    View Slide

  67. Popularity
    Low virality Moderate virality High virality
    Overall
    Copy 1
    Copy 2
    Time

    View Slide

  68. Popularity
    Low virality Moderate virality High virality
    Overall
    Copy 1
    Copy 2
    Time

    View Slide

  69. Popularity
    Low virality Moderate virality High virality
    Overall
    Copy 1
    Copy 2
    Time

    View Slide

  70. Popularity
    Low virality Moderate virality High virality
    Overall
    Copy 1
    Copy 2
    Time

    View Slide

  71. Low virality Moderate virality High virality
    Popularity
    Recurrence No Recurrence
    No Recurrence
    Time
    Overall
    Copy 1
    Copy 2

    View Slide

  72. Simulations of this model
    replicate previous key findings.

    View Slide

  73. Can we predict recurrence?

    View Slide

  74. Features
    (e.g., burst length) (e.g., gender) (e.g., # edges) (e.g., # copies)
    Temporal Sharer Network Copy

    View Slide

  75. Predicting recurrence
    Will it recur?
    Existence
    Will it be larger?
    Size
    When will it recur?
    Time

    View Slide

  76. Predicting recurrence
    Will it recur? Will it be larger? When will it recur?
    Existence Size Time
    .89 .78 .58
    AUC
    AUC AUC

    View Slide

  77. Future / Related Work
    • Effect of Multiple Networks, External Stimuli

    Gruhl et al. (2004), Kumar et al. (2005), Myers & Leskovec (2012)
    • Improved Models of Recurrence

    Barabasi (2006), Cha et al. (2012), Matsubara et al. (2012)
    • Other Factors Influencing Recurrence (Seasonality, Sentimentality)

    Altizer et al. (2006), Verdasca et al. (2005)

    View Slide

  78. Cascades Do Recur.
    Justin Cheng, Lada Adamic, Jon Kleinberg, Jure Leskovec
    http://bit.ly/cascades-paper

    View Slide