1 A social network caught in the Web Lada Adamic and Eytan Adar (HP Labs, Palo Alto, CA) Orkut Buyukkokten (Google)
Dec 21, 2015
1
A social network caught in the Web
Lada Adamic and Eytan Adar (HP Labs, Palo Alto, CA)Orkut Buyukkokten (Google)
2
Outline
Intro to Club Nexus
Profiles
Nexus Net
Similarity and distance
Association by similarity
Nexus Karma
Conclusions
7
Profiles:
status (UG or G) yearmajor or departmentresidencegender
Personality (choose 3 exactly):you funny, kind, weird, …friendship honesty/trust, common interests, commitment, …romance - “ -freetime socializing, getting outside, reading, …support unconditional accepters, comic-relief givers, eternal optimists
Interests (choose as many as apply)books mystery & thriller, science fiction, romance, …movies western, biography, horror, …music folk, jazz, techno, …social activities ballroom dancing, barbecuing, bar-hopping, …land sports soccer, tennis, golf, …water sports sailing, kayaking, swimming, …other sports ski diving, weightlifting, billiards, …
8
Finding correlations between user attributes
Are people who consider themselves funny also more likely to enjoy comedies?
518 funny users74 % of users overall like comedies416 (80% of) funny users like comedies,
this is 3.4 standard deviations (=10) above expected (383)
Z score = 3.4
Z scores with absolute value > 2 are significant at the p = 0.05 level.3.4 is significant at the 0.0003 level
small differences (10%) can be significant.
9
book business landsport tennis other weightlifting social barbecuing watersport boating, jet skiing, water skiing
successful
free time fulfilling commitments, catching up on chores and things
book sex movie erotic & softcore, gay & lesbian,
independent music funk, jungle, reggae, trance other skateboarding
not responsible
social raving
book art & photography, philosophy, fiction & literature, classics
music folk, bluegrass/rural, jazz
creative
movie art, documentary, independent
Personality and tastes (just a few examples)
10
Major and personalitypersonality (% of total) major
free time: learning (17%) Physics (46%), Philosophy (37%), Math (31%),
EE (26%), CS (24%)
free time: reading (26%) English (55%)
free time: staying at home (8%) History (24%)
free time: doing anything exciting (52%)
undecided/undeclared (62%)
you: weird (12%) Physics (34%), Math (28%), EE (18%)
you: intelligent (32%) Philosophy (59%), CS (42%)
you: successful (4%) CS (7%)
you: socially adaptable (14%) STS (46%)
you: attractive (16%) Political Science (29%), International Relations (25%)
you: lovable (12%) Political Science (24%)
you: kind (25%) Public Policy (45%)
you: funny (25%) Philosophy (6%)
you: fun (26%) Human Biology (38%)
you: creative (22%) Product Design (62%), English (42%)
you: sexy (8%) English (18%), EE (2%)
11
preference Male users Female users book computers, science fiction, professional &
technical, science, business, politics, philosophy, sports, adventure
romance, fiction & literature, health mind & body, cooking, art & photography, entertainment, mystery & thriller, psychology, classics
landsport football, frisbee golfing, table tennis, golf, baseball, basketball, cricket, fencing, racquetball, squash, tennis, soccer, wrestling
gymnastics, field hockey, softball
movie science fiction, war, action, spy film, erotic & softcore, adventure, anime, sports, western
romance, family, drama, musical, performing arts, comedy, independent
music heavy metal soul/R&B, pop, country/western, rap/hip hop, folk, latin
other computer gaming, weightlifting, billiards, ultimate frisbee, mountain biking, paintballing, laser gaming, bicycling
aerobics, ice skating, jogging
social barbecuing, raving, hot tubbing hip-hop dancing, lating dancing, clubbing
watersport fishing, sailing swimming personality freetime learning, doing physical challenging activities catching up on chores and things,
socializing friendship mutual friends, common interests,
appearance/look, sex laughter, honesty/trust, communication
romance appearance/look, sex, physical attraction laughter, honesty/trust
support the eternal optimists, the give-it-to-you-straight people, i've-been-down-and-dirty-a-few-times-myself people
unconditional accepters, the listeners, chicken-soup people
you intelligent fun, lovable, friendly
Gender Differences
12
0 20 40 60 80 1000
50
100
150
200
250
number of links
nu
mb
er
of u
sers
with
so
ma
ny
links
100
101
102
100
101
102
number of links
num
ber
of u
sers
Degree Distribution for Nexus Net 2469 users, average degree 8.2
13
1 2 3 4 5 6 7 8 9 10 11 12 130
2
4
6
8
10
12x 10
5
distance
pa
irs
of u
sers
average distance = 4.0
Shortest paths between users
14
Clustering and betweenness
Clustering or transitivity: how many of the user’s friends are friends themselves
C = # links between friends
(# friends)* (# friends - 1)/2
c = 0.17 for Club Nexus
Other findings:
people who list more buddies list more preferences/activities
edges with high betweenness lie between dissimilar people ( = -0.2)
people with high betweenness have more links ( = 0.7)
- “ - have lower clustering coefficients ( = -0.12)
15
Similarity and distance
year is more important for undergradsdepartment is more important for grads
1 2 3 4 5 6 7 80
0.2
0.4
0.6
0.8
1
distance between users in hops
frac
tio
n o
f si
mila
r u
sers
G residenceUG residenceG departmentUG majorG yearUG yearG statusUG status
16
users who like Aall users
Association ratios
p = (# users who like A)/(total #users)L = # connections A users havem = expected number of links to other A users = L*pr = (# links between A users)/m
17
personality association ratio
Z score # users # connections
sexy 1.46 5.47 204 192 talented 1.40 5.17 213 210 fun 1.25 11.22 633 1852 weird 1.25 4.32 286 332 lovable 1.22 4.20 292 406 unique 1.11 4.15 547 1194 funny 1.10 4.06 619 1474 friendly 1.10 7.55 1024 4024 socially adaptable 1.09 2.12 342 482
attractive 1.07 1.76 406 522 creative 1.04 1.48 541 982 intelligent 1.01 0.42 779 1848 responsible 0.99 -0.28 500 686 kind 0.99 -0.44 625 1226 competent 0.92 -1.40 294 226 successful 0.70 -1.57 99 18
Personality and association ratio
18
high association low association
book gay & lesbian, professional & technical, computers, teen, sex, sports
history, fiction & literature, outdoor & nature
movie
genres
gay & lesbian, performing arts, religion, erotic & softcore, sports
drama, mystery, documentary, comedy
music genres
gospel, jungle, bluegrass/rural, heavy metal, trance
pop, classical, rock
land sport lacrosse, field hockey, wrestling, cricket tennis, martial arts, bicycling, racquetball
water sport
synchronized swimming, diving, crew swimming, fishing windsurfing
social raving, ballroom dancing, Latin dancing partying, camping
Interests and association ratios
19
Nexus Karma
Rank how ‘trusty’, ‘nice’, ‘cool’, and ‘sexy’ your buddiesare on a scale of 1 to 4
446 users ranked 1735 different friends
correlations between scores given (users were ranked as‘3,3,3,3’ more often than ‘1,4,2,3’
average scores: nice (3.37), trusty (3.22), cool (3.13), sexy(2.83)
trusty--nice and cool--sexy more highly correlated ( = 0.7) vs.trusty--sexy and nice--sexy ( = 0.4)
no relationship between average score received and # of friendsnegative correlation between average score given and # of friends
20
How users view themselves vs. how others view them
trusty (3.22)nice(3.37)
cool(3.13)
sexy(2.83)
responsible
3.36 3.02 2.67
sexy 3.10 3.23 3.03
attractive 3.09 3.25 2.93
kind 3.34 3.46
friendly 3.44
weird 2.67
funny 3.31
21
Additional insights from Nexus Karma
Users receiving higher ‘nice’ scores give higher ‘trusty’, ‘nice’, and ‘cool’scores ( = 0.14-0.17)
If one user gives another user a higher ‘trusty’ or ‘nice’ score than their other friends, that same friend is more likely to reciprocate.
Users who share friends are more likely to give each other high scores( = 0.10-0.13)
22
Conclusions
Learn about real world social networks from online community
Less effort than traditional social network survey methods,almost a side-effect of digital nature of interactions
Although most results not surprising, data is very rich - opportunity to simulate search and information spread
Karma data can be used to study online reputation mechanisms
Longitudinal data can be used to study network evolution
23
To find out more:
Information dynamics group (IDL) at HP Labs:http://www.hpl.hp.com/shl/
Paper at:http://www.hpl.hp.com/shl/social/
24
free time activity association ratio
Z score # users # connections
fulfilling commitments
1.34 9.30 398 826
socializing 1.12 21.12 1660 11374 catching up on chores and things
1.09 2.71 494 850
learning 1.07 1.82 420 536 doing anything exciting
1.07 8.05 1280 6278
watching TV 1.07 1.85 415 602 reading 1.02 0.66 631 1186 getting outside 1.01 0.97 940 2882 staying at home 0.97 -0.32 209 126 alone 0.96 -0.93 380 398 doing physical challenging activities
0.96 -1.46 577 878
Free time activity and association ratios