Presentazione standard di PowerPointsomeni.di.unimi.it/slide/AA2017_random-real-networks.pdf ·...

Post on 20-Jun-2020

1 views 0 download

transcript

RANDOM-REAL NETWORKS

1

Random networks: model

A random graph is a graph of N nodes where each pair of nodes is connected by probability p: G(N,p)

Random networks: model

3

p=1/6

N=12

L=8 L=10 L=7

The number of links is variable

< L >= LP(L) = pN(N -1)

2L= 0

N(N-1)

2

å

Degree distribution

Real networks are scale-free

Random networks: degree distribution

5

P(k) =N -1

k

æ

è ç

ö

ø ÷ p

k(1- p)(N-1)-k

< k >= p(N -1)

p =< k >

(N -1)

P(k) = e-< k> < k >k

k!

For large N and small p, it can be approximated by a Poisson distribution:

Binomial distribution:

Real networks are scale-free

6

Degree distribution is power-law

Scale-free real networks

7

Degree distribution

Random networks

Poisson distribution

Real networks

Power-law distribution

8

P(k) = e-< k> < k >k

k!

< k >= p(N -1)

A random society would consist of mainly average individuals, with everyone with roughly the same number of friends. It would lack outliers, individuals that are either highly popular or recluse.

Hubs: individuals that are highly “popular”

Connected components

Real networks: giant component and power-law connected components

size distribution

Connected components

10

P=0 disconnected nodes, <k>=0, N_G=1 P=1 fully connected, <k>=N-1, N_G=N One would expect that that the largest component grows

gradually from N_G=1 to N_G=N

Section 3.4

<k>

disconnected nodes NETWORK.

Erdos and Renyi (1959): the condition for the emergence of a giant component is <k>=1. It is evident that one link per node is necessary, but counterintuitive that it also sufficient.

Connected components

I:

Subcritical

<k> < 1

III:

Supercritical

<k> > 1

IV:

Connected

<k> > ln N

II:

Critical

<k> = 1

<k>=0.5 <k>=1 <k>=3 <k>=5

N=

10

0

<k>

I:

Subcritical

<k> < 1

p < pc=1/N

<k>

No giant component.

Isolated clusters, cluster size distribution is exponential

The largest cluster is a tree, its size ~ ln N

II:

Critical

<k> = 1

p=pc=1/N

<k>

Unique giant component: NG~ N2/3

contains a vanishing fraction of all nodes, NG/N~N-1/3

Small components are trees, GC has loops.

Cluster size distribution: p(s)~s-3/2

A jump in the cluster size:

<k>=1,000 ln N~ 6.9; N2/3~95

N=7 109 ln N~ 22; N2/3~3,659,250

<k>=3

<k>

Unique giant component: NG~ (p-pc)N

GC has loops.

Cluster size distribution: exponential

III:

Supercritical

<k> > 1

p > pc=1/N

p(s) ~ s-3 /2e-( k -1)s+(s-1)ln k

IV:

Connected

<k> > ln N

p > (ln N)/N

<k>=5

<k>

Only one cluster: NG=N

GC is dense.

Cluster size distribution: None

Connected components: real networks

19

Connected components: real networks

20

Supercritical: not fully connected Internet: we should have routers that, being disconnected from the giant component, are unable to communicate with other routers. Power grid: some consumers should not get powered Fully connected Social media: no individual disconnected

Diameter and path length Small World

SIX DEGREES 1967: Stanley Milgram

Network Science: Random Graphs

HOW TO TAKE PART IN THIS STUDY

1. ADD YOUR NAME TO THE ROSTER AT THE BOTTOM OF THIS SHEET, so

that the next person who receives this letter will know who it came from.

2. DETACH ONE POSTCARD. FILL IT AND RETURN IT TO HARVARD

UNIVERSITY. No stamp is needed. The postcard is very important. It allows us to keep

track of the progress of the folder as it moves toward the target person.

3. IF YOU KNOW THE TARGET PERSON ON A PERSONAL BASIS, MAIL THIS

FOLDER DIRECTLY TO HIM (HER). Do this only if you have previously met the target

person and know each other on a first name basis.

4. IF YOU DO NOT KNOW THE TARGET PERSON ON A PERSONAL BASIS,

DO NOT TRY TO CONTACT HIM DIRECTLY. INSTEAD, MAIL THIS FOLDER (POST

CARDS AND ALL) TO A PERSONAL ACQUAINTANCE WHO IS MORE LIKELY THAN

YOU TO KNOW THE TARGET PERSON. You may send the folder to a friend, relative or

acquaintance, but it must be someone you know on a first name basis.

SIX DEGREES 1967: Stanley Milgram

Network Science: Random Graphs

SIX DEGREES 1991: John Guare

Network Science: Random Graphs

"Everybody on this planet is separated by only six other people.

Six degrees of separation. Between us and everybody else on

this planet. The president of the United States. A gondolier in

Venice…. It's not just the big names. It's anyone. A native in a

rain forest. A Tierra del Fuegan. An Eskimo. I am bound to

everyone on this planet by a trail of six people. It's a profound

thought. How every person is a new door, opening up into other

worlds."

Random graph: diameter

25

Random graphs tend to have a tree-like topology with almost constant node degrees.

dmax =logN

log k

N =1+ k + k2

+ ...+ kdmax =

kdmax +1

-1

k -1» k

dmax

Small World

26

< d >=logN

log k

Small world phenomena: the property that the average path length or the diameter depends logarithmically on the system size. ”Small” means that ⟨d⟩ is proportional to log N

In most networks this offers a better approximation to the average distance between two randomly chosen nodes, ⟨d⟩, than to dmax .

Given the huge differences in scope, size, and average degree, the agreement is excellent.

Social networks

28

< d >=ln(N)

ln k= 3.28

N=7*10^9, <k>=1000

Using Facebook’s social graph of May 2011, consisting of 721 million active users and 68 billion symmetric friendship links, researchers found an average distance 4.74 between the users. Therefore, the study detected only ‘four degrees of separation’ .

WWW: 19 DEGREES OF SEPARATION

Image by Matthew Hurst

Blogosphere Network Science: Random Graphs

Smaller average degree and larger order than the social network [d=ln N/ln <k>]

Clustering coefficient

Real networks: triadic closure

Random networks: clustering coefficient

31

•The clustering coefficient of random graphs is small.

• For fixed degree, C decreases with the system size N.

• C is independent of a node’s degree k.

Clustering coefficient: real networks

32

In May 2011, Facebook had an average clustering coefficient of 0.5 for individuals who had 2 friends.

Real networks: clustering coefficient

33

Random:

(a) Green line: random network. The average clustering coefficient decreases as 1/N Circles: real networks The average clustering coefficient is independent of N. (b) –(d) Dependence of the local clustering coefficient on node’s degree. Green line: average clustering coefficient in random networks

Clustering coefficient

Random networks Real networks

A much higher clustering coefficient than expected for a random network of similar N and L.

Independent of N

High-degree nodes tend to have a smaller clustering coefficient than low-degree nodes.

34

The clustering coefficient of random graphs is small.

For fixed degree C decreases with the system size N.

C is independent of a node’s degree k.

Small World Model

Small-world Model

• Small-world Model also known as the Watts and Strogatz model is a special type of random graphs with small-world properties, including:

– Short average path length

– High clustering.

• It was proposed by Duncan J. Watts and Steven Strogatz in their joint 1998 Nature paper

Constructing Small World Networks

As in many network generating algorithms • Disallow self-edges • Disallow multiple edges

regular ring lattice of degree c: nodes are connected to their previous c/2 and following c/2 neighbors.

Real-World Network and Simulated Graphs

Small world model

Random networks Real networks

39

Low average path length High clustering coefficient Clustering coefficient independent of k Poisson-like degree distribution

Low average path length High clustering coefficient Clustering coefficient dependent on k Power-law degree distribution

How can human networks be both clumpy and have short distances? The strength of weak ties