Week 13 How is date being used Predict Presidential Election Nate Silver httpadagecomarticlecampaigntrailnatesilverselectionpredictionsawinbigdatayorktimes238182 Predict Pregnancy Target ID: 565685
Download Presentation The PPT/PDF document "Big Data, Network Analysis" is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Slide1
Big Data, Network Analysis
Week 13Slide2
How is date being used
Predict Presidential Election - Nate Silver
–
http://adage.com/article/campaign-trail/nate-silver-s-election-predictions-a-win-big-data-york-times/238182
/
Predict Pregnancy - Target
–
http://www.forbes.com/sites/kashmirhill/2012/02/16/how-target-figured-out-a-teen-girl-was-pregnant-before-her-father-did
/Slide3
Why Networks?
Why is the role of networks in CS, Info Science, Social Science, Physics, Economics, and Biology expanding?
More Data
Rise of the Web and Social Media
Shared vocabulary between (very different fields)Slide4
Reasoning about Networks
How do we reason about networks?
Empirical
: Look at large networks and see what we find
Mathematical Models
: probabilistic, graph theory
Algorithms
for
analyzing graphs
What do we hope to achieve from the networks?
Patterns and statistical properties of network data
Design principles and models
Understand why networks are organized the way they are (predict behaviors or networked systems)Slide5
Why networks?
Network data is increasingly available:
Large on-line computing applications where data can naturally be represented as a network
Online communities:
Facebook
Communications:
Instant Messenger
News and Social Media:
Blogging
Also in systems biology, health, medicine, …Slide6
Networks: Rich Data
a – Internet
b
– Citation network
c
– World Wide Web
d – sexual network
e – dating networkSlide7
Networks
Information networks:
World Wide Web: hyperlinks
Citation
Blog
Social networks:
Organizational
Communication
Collaboration
Sexual
Collaboration
Technological networks:
Power grid
Airline, road, river
Telephone
InternetAutonomous systemsSlide8
What is Social Network Analysis?
Network analysis is the study of social relations among a set of actors. It is a
field of study, not just a method.
“
Social network analysis involves theorizing, model building and empirical research focused on uncovering the patterning of links among actors. It is concerned also with uncovering the antecedents and consequences of recurrent patterns.
”
(Linton Freeman
)Slide9
The Network Perspective
People
(nodes)
Ties
(edges)Slide10
Ties in a social network
Directed or undirected
Simplex or multiplex
Valued or unvalued
7Slide11
What is a Social Network?
A set of dyadic ties, all of the same type, among a set of actors
Actors can be persons, organizations, groups
A tie is an instance of a
specific
social relationship
11Slide12
Network Relations
Among
Individuals
Kinship
Role-based (friend of)
Cognitive/Perceptual (knows, aware of)
Affiliations
Affective (likes, trusts)
Communication
Among Organizations
Buy from / Sell to
Owns shares of
Joint ventures
12Slide13
Key Perspectives in Social Network Analysis
Focus on
relationships
between actors rather than just the attributes of actors.
Interdependent view
rather than atomistic (individualist) view of social processes and effects.
Social structure
affects substantive outcomes (which is a philosophical departure from other traditions)
13Slide14
Interdisciplinary Field of Study
Computer Science
Designing and understanding complex network structures
Mathematics, Physics
Methods, complex systems analysis
Social Science (Sociology, Social Psychology, Economics)
Theories and measurement of social networks, using networks to understand human behavior
14Slide15
Multiple Levels of Analysis
Individual Level
How does individual position in a network affect various outcomes for the individual?
Systems Level
How does the network structure as a whole affect outcomes for various tasks?
15Slide16
Network Data Collection
Common Types:
Survey
Interviews
Affiliation/membership records
Behavioral (e.g., observation of communication patterns)
Experiments
16
Data obtained through manyeyes and graphed: http://www.esv.org/blog/2007/01/mapping.nt.social.networksSlide17
Types of Network Data
One mode Two mode
Whole network Egocentric
17
A
B
C
A
B
School ASlide18
Non-directed versus Directed Graphs
18
A
B
C
A
B
CSlide19
Analyzing Social Networks
A
B
C
D
A
-
1
1
1
B
1
-
1
0
C
1
1
-
1
D
1
0
1
-
19
A
D
B
C
Simple Adjacency MatrixSlide20
Some Key Principles in Social Networks
Degree
The degree to which actors are connected directly to each other by
cohesive bonds
Density
The
proportion of direct ties in a network relative to the total number
possible
Centrality
a group of metrics that aim to quantify the "importance" or "influence" (in a variety of senses) of a particular node (or group) within a
network
20Slide21
Degree in Social Networks
21Slide22
Density in Social Networks
22
Low Density
High Density / Integrated
“
Radial
”
(Valente)Slide23
Centrality in Social Networks
Degree Centrality
Closeness Centrality
Betweeness Centrality
23Slide24
Why all of this sudden interest?
The strength of the
“
Strength of Weak Ties
”
argument.
Granovetter (1973)
Argues that
‘
weaker
’
peripheral ties build heterogeneous networks, which in turn provide access to new and useful information.
Heterogeneity through weak-ties widely accepted as a
“
good thing
”
for communication
Access to jobs
Access to other opportunities
Helps distribute ideas, innovations
24Slide25
Ted Talk
The
hidden
influence
of social
networks
http
://youtu.be/2U-tOghblfESlide26
Social Networks
http://www.youtube.com/watch?v=
5etSid8G6EU
http://www.youtube.com/watch?v=PThAriHjk10&playnext=1&list=PL05CC28C66163B00D&feature=
results_mainSlide27
Slides adapted from:
Jure
Leskovec
, Stanford CS224W: Social and Information Network Analysis
ure
Leskovec
, Stanford CS224W: Social and Information Network Analysis
http
://bit.ly/
Y7fALp