Human Social Dynamics vs. The Data We Can Get Aaron Clauset - - PowerPoint PPT Presentation

Aaron Clauset Assistant Professor, Computer Science and BioFrontiers Institute, University of Colorado Boulder External Faculty, Santa Fe Institute

Human Social Dynamics vs. The Data We Can Get

4 June 2013 NetSci 2013 Social Dynamics Workshop

1

• Twitter, Facebook, Google+, Pinterest, etc.
• Academic coauthorships & citations
• World Wide Web
• etc.

data sources for social network dynamics

2

but...

• links often have low or no cost = unrealistic
• system structure drives social dynamics
• few sources capture “real” social networks

(face-to-face time)

• Twitter, Facebook, Google+, Pinterest, etc.
• Academic coauthorships & citations
• World Wide Web
• etc.

data sources for social network dynamics

3

the data we want

[detailed, individual traces, over time, about specific social processes]

are not the data we can get

[detailed, massive electronic traces that may be only vaguely relevant to any real human social processes]

the more general problem

4

an illustration

5

Halo: Reach (Bungie, 2010)

• played online via XBox Live platform
• team combat simulation (FPS)
• 20TB of game data, spanning
• 18 months of time
• 17+ million players
• 1 billion competitions
• 70% are team competitions
• complex spatial environments
• complex social interactions

an illustration

6

• join “party” (of 0-3 friends)
• choose game type and

subtype (“competitive / team 4v4”)

• Xbox Live places parties into

matches (matchmaking)

• play! (for roughly 10 minutes)
• repeat (1 billion times)

how it works

7

• observed interactions =

F(game matchmaking , friendships)

• mean interaction degree 330
• how to distinguish latent friendships

from observed interactions?

• plus, no demographic information

the problem

≈

8

• anonymous web survey

a small solution

Mason and Clauset, CSCW 2013 9

• anonymous web survey
• 847 participants
• demographic questions

age, sex, location, education

• psychometric questions

attitudes, play style, etc.

• friendship survey
• 14,405 labeled friends
• 7,159,989 labeled non-friends

Mason and Clauset, CSCW 2013

a small solution

10

we can observe a sequence of pairwise interactions can we robustly distinguish friendships from non-friendships? recovering friendships from interactions

σij = (i, j, t1), (i, j, t2), . . .

11

we can observe a sequence of pairwise interactions can we robustly distinguish friendships from non-friendships? recovering friendships from interactions

σij = (i, j, t1), (i, j, t2), . . .

• volume of data varies widely by individual =

heavy-tailed distribution in

• friendships are sparse in large networks
• survey data provide “subjective” truth only

problems:

|σij|

12

friendship = periodic + prosocial interactions diurnal cycle modulates all interactions what is a friendship? social interactions: recovering latent friendship ties supervised learning define 9 statistical features which do well?

Merrit, Jacobs, Mason and Clauset, ICWSM 2013 13

• AUC vs. how much data
• n a person we have
• periodic + prosocial

interactions highly robust and efficient

• total interaction count

also good, eventually accuracy vs. amount of interaction data 9 features of pairwise interactions

periodic or prosocial interactions volume of interactions

Merrit, Jacobs, Mason and Clauset, ICWSM 2013 14

• friendships easy to recover from interactions
• mean degree (interactions) 330
• mean degree (friendships) 4
• friendship graph very different from interaction graph
• results likely to generalize [see Jones et al. PLoS ONE (2013)]
• clarifies “friendship” = periodic + prosocial interactions
• privacy concerns

recovering friendships from interactions

Merrit, Jacobs, Mason and Clauset, ICWSM 2013 Jones et al. PLOS ONE 8(1):e52168 (2013).

≈ ≈

15

general outlook

• electronic data
• new window into human social dynamics!
• big and detailed!
• but, dumb. (not the data we want)
• computational social science
• know limits of dumb data
• keep eye on underlying social processes
• be willing to commit a “social science”
• a general prescription
• 1. obtain electronic data (big and dumb)
• 2. seek out user labeled data (small and smart)
• 3. model latent variables from observed data (supervised)
• 4. extract underlying social dynamics

16

thanks

Abigail Z Jacobs (Colorado) Sears Merritt (Colorado) Winter Mason (Stevens Inst. Tech.)

funded in part by references

• Mason and Clauset, “Friends FTW! Friendship, Collaboration and Competition in Halo: Reach.”

CSCW (2013)

• Merritt, Jacobs, Mason and Clauset, “Detecting friendship within dynamic online interaction networks.”

ICWSM (2013).

• Merrit and Clauset, “Environmental structure and competitive scoring advantages in team competitions.”

arXiv:1304.1039 (2013)

17

fin

18