Scribd
Upload
42 views35 pages

Advanced Topics in Data Mining Special Focus: Social Networks

This document provides an overview of key concepts and analysis techniques for social networks. It discusses measuring properties like degree distributions, clustering coefficients, small world phenomena, and more. The class will focus on recent papers analyzing social network data. Students will present papers, work on projects, and learn about topics like information propagation, expertise location, and privacy in networks. The introductory lectures will cover measurements, generative models, and algorithms for social network analysis.

Uploaded by

smjain
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
42 views35 pages

Advanced Topics in Data Mining Special Focus: Social Networks

This document provides an overview of key concepts and analysis techniques for social networks. It discusses measuring properties like degree distributions, clustering coefficients, small world phenomena, and more. The class will focus on recent papers analyzing social network data. Students will present papers, work on projects, and learn about topics like information propagation, expertise location, and privacy in networks. The introductory lectures will cover measurements, generative models, and algorithms for social network analysis.

Uploaded by

smjain
Copyright
© Attribution Non-Commercial (BY-NC)
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPT, PDF, TXT or read online on Scribd
You are on page 1/ 35

Advanced Topics in Data Mining Special focus: Social Networks

Goal of the class


Address major trends in the analysis of socialnetwork data Get you involved and interested Do something fun and cool

What is a social network?


Facebook LinkedIn . The network of your friends and acquaintances Social network is a graph G=(V,E)
V: set of users E: connections/friendships among users

Social Networks
Links denote a social interaction
Networks of acquaintances collaboration networks
actor networks co-authorship networks director networks

phone-call networks e-mail networks IM networks Bluetooth networks sexual networks home page/blog networks

Themes in data analysis for social networks


Measure characteristics of social networks (Measurements)
How many hops apart are two random Facebook users

Design models that capture the generation process of network data (Generative Models)
Generate graphs with the same properties as real social network graphs

Algorithmic problems related to (Algorithmic SN analysis)


Information propagation Advertising Expertise finding Privacy

Structure and function of the class


Material: Mostly based on recent papers related to social-network analysis.
Some papers and links are already posted on the website of the class Other interesting papers can be found in the proceedings of : KDD, WWW, WSDM, ICDM conferences

Goal: Understand the material in these papers and (hopefully) extend it

Structure and function of the class


Introductory lectures Paper presentations (20%) Projects and Project Presentation (50%) Project Report (otherwise called reaction paper) (20%) Class Participation (10%)

Introductory Lectures
Measurements in networks Generative models Algorithmic topics
Introduction to information propagation Expertise location Privacy

Measuring Networks
Degree distributions Small world phenomena Clustering Coefficient Mixing patterns Degree correlations Communities and clusters

Degree distributions

frequency

fk = fraction of nodes with degree k = probability of a randomly selected node to have degree k

fk

degree

Problem: find the probability distribution that best fits the observed data

Power-law distributions
The degree distributions of most real-life networks follow a power law

p(k) = Ck-
Right-skewed/Heavy-tail distribution
there is a non-negligible fraction of nodes that has very high degree (hubs) scale-free: no characteristic scale, average is not informative

In stark contrast with the random graph model!


Poisson degree distribution, z=np

zk z p(k) P(k; z) e k!
highly concentrated around the mean the probability of very high degree nodes is exponentially small

Power-law signature
Power-law distribution gives a line in the log-log plot
log p(k) = - logk + logC
frequency log frequency

degree

log degree

: power-law exponent (typically 2 3)

Examples

Taken from [Newman 2003]

Exponential distribution
Observed in some technological or collaboration networks
p(k) = e-k

Identified by a line in the log-linear plot


log p(k) = - k + log
log frequency

degree

The basic random graph model


The measurements on real networks are usually compared against those on random networks The basic Gn,p (Erds-Renyi) random graph model:
n : the number of vertices 0p1 for each pair (i,j), generate the edge (i,j) independently with probability p

A random graph example

Average/Expected degree
For random graphs z = np For power-law distributed degree
if 2, it is a constant if < 2, it diverges

Maximum degree
For random graphs, the maximum degree is highly concentrated around the average degree z For power law graphs
k max n1/(1)

Clustering (Transitivity) coefficient


Measures the density of triangles (local clusters) in the graph Two different ways to measure it:
C (1)

triangles centered at node i triples centered at node i


i i

The ratio of the means

Example
1 4

3 2 5

(1)

3 3 1 1 6 8

Clustering (Transitivity) coefficient


Clustering coefficient for node i
triangles centered at node i Ci triples centered at node i

(2)

1 Ci n

The mean of the ratios

Example
1 4

(2)

1 13 1 1 1 6 5 30 3 8

C (1)

The two clustering coefficients give different measures C(2) increases with nodes with low degree

Clustering coefficient for random graphs


The probability of two of your neighbors also being neighbors is p, independent of local structure
clustering coefficient C = p when z is fixed C = z/n =O(1/n)

The C(k) distribution


The C(k) distribution is supposed to capture the hierarchical nature of the network
when constant: no hierarchy when power-law: hierarchy
C(k) = average clustering coefficient of nodes with degree k

C(k)

degree

The small-world experiment


Milgram 1967 Picked 300 people at random from Nebraska Asked them to get the letter to a stockbroker in Boston they could bypass the letter through friends they knew on a first-name basis How many steps does it take?
Six degrees of separation: (play of John Guare)

Six Degrees of Kevin Bacon


Bacon number:
Create a network of Hollywood actors Connect two actors if they co-appeared in some movie Bacon number: number of steps to Kevin Bacon

As of Dec 2007, the highest (finite) Bacon number reported is 8 Only approx 12% of all actors cannot be linked to Bacon What is the Bacon number of Elvis Prisley?

Erdos numbers?

The small-world experiment


64 chains completed
6.2 average chain length (thus six degrees of separation)

Further observations
People that owned the stock had shortest paths to the stockbroker than random people People from Boston area have even closer paths

Measuring the small world phenomenon


dij = shortest path between i and j Diameter:
d max dij
i, j

Characteristic path length:


1 dij n(n - 1)/2 i j

Harmonic mean
1 1 -1 d n(n - 1)/2 i j ij

Also, distribution of all shortest paths

Is the path length enough?


Random graphs have diameter
logn d logz

d=logn/loglogn when z=(logn)


Short paths should be combined with other properties
ease of navigation high clustering coefficient

Degree correlations
Do high degree nodes tend to link to high degree nodes? Pastor Satoras et al.
plot the mean degree of the neighbors as a function of the degree

Degree correlations
Newman
compute the correlation coefficient of the degrees of the two endpoints of an edge assortative/disassortative

Connected components
For undirected graphs, the size and distribution of the connected components
is there a giant component?

For directed graphs, the size and distribution of strongly and weakly connected components

Graph eigenvalues
For random graphs
semi-circle law

For the Internet (Faloutsos3)

Next class
What is a good model that generates graphs in which power law degree distribution appears? What is a good model that generates graphs in which small-world phenomena appear?

You might also like