Example 3 (Adjacency matrix)

• [[ 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0. 0.]
• [ 0. 0. 0. 0. 1. 0. 1. 0. 0. 0. 0. 0. 0. 0. 0.]
• [ 1. 0. 0. 0. 1. 0. 0. 1. 1. 1. 0. 0. 0. 0. 0.]
• [ 0. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 1. 0. 0.]
• [ 0. 1. 1. 0. 0. 0. 0. 0. 0. 0. 1. 0. 1. 1. 0.]
• [ 0. 0. 0. 1. 0. 0. 1. 0. 0. 0. 0. 0. 1. 0. 1.]

• [ 0. 1. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0. 1. 1.]
• [ 0. 0. 1. 0. 0. 0. 0. 0. 1. 1. 0. 0. 0. 0. 0.]
• [ 0. 0. 1. 0. 0. 0. 0. 1. 0. 0. 0. 1. 0. 0. 1.]
• [ 0. 0. 1. 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 0.]
• [ 0. 0. 0. 0. 1. 0. 0. 0. 0. 0. 0. 1. 1. 1. 0.]
• [ 0. 0. 0. 0. 0. 0. 0. 0. 1. 0. 1. 0. 0. 0. 0.]
• [ 0. 0. 0. 1. 1. 1. 0. 0. 0. 0. 1. 0. 0. 0. 0.]
• [ 0. 0. 0. 0. 1. 0. 1. 0. 0. 0. 1. 0. 0. 0. 0.]

• [ 0. 0. 0. 0. 0. 1. 1. 0. 1. 0. 0. 0. 0. 0. 0.]]

1
Example 3 (Eigenvector centrality)
• 0: 0.08448651593556764,
• 1: 0.1928608426462633,
• 2: 0.3011603786470362,
• 3: 0.17530527234882679,
• 4: 0.40835121533077895,
• 5: 0.2865100597893966,
• 6: 0.2791290343288953,
• 7: 0.1931920790704947,
• 8: 0.24881035953707603,
• 9: 0.13868390351302598,
• 10: 0.336067959653752,
• 11: 0.16407815738375933,
• 12: 0.33838887484747293,
• 13: 0.2871391639624871,
• 14: 0.22848023925633135

2
Example 3 (Eigenvector centrality)
• 2: 0.3011603786470362,
• 4: 0.40835121533077895,
• 5: 0.2865100597893966,
• 6: 0.2791290343288953,
• 8: 0.24881035953707603,
• 10: 0.336067959653752,
• 12: 0.33838887484747293,
• 13: 0.2871391639624871,
• 14: 0.22848023925633135

Also, deg(4) = deg(2) = Δ(G), yet…

3
Example 3 (Eigenvector centrality)
• 2: 0.3011603786470362,
• 4: 0.40835121533077895,
• 5: 0.2865100597893966,
• 6: 0.2791290343288953,
• 8: 0.24881035953707603,
• 10: 0.336067959653752,
• 12: 0.33838887484747293,
• 13: 0.2871391639624871,
• 14: 0.22848023925633135

Also, deg(4) = deg(2) = Δ(G), yet 𝐶! > 𝐶" .

4
 Example 3 (Eigenvector centrality)
C2 = 0.3011603786470362 C4 = 0.40835121533077895

Adjacent to vertices of small degree Adjacent to vertices of large degree

5
Example 3 (Eigenvector centrality)
• {0: 0.08448651593556764,
• 1: 0.1928608426462633,
• 2: 0.3011603786470362,
• 3: 0.17530527234882679,
• 4: 0.40835121533077895,
• 5: 0.2865100597893966,
• 6: 0.2791290343288953,
• 7: 0.1931920790704947,
• 8: 0.24881035953707603,
• 9: 0.13868390351302598,
• 10: 0.336067959653752,
• 11: 0.16407815738375933,
• 12: 0.33838887484747293,
• 13: 0.2871391639624871,
deg(13) < deg(8)
• 14: 0.22848023925633135}
6
Example 3 (Eigenvector centrality)
• {0: 0.08448651593556764,
• 1: 0.1928608426462633,
• 2: 0.3011603786470362,
• 3: 0.17530527234882679,
• 4: 0.40835121533077895,
• 5: 0.2865100597893966,
• 6: 0.2791290343288953,
• 7: 0.1931920790704947,
• 8: 0.24881035953707603,
• 9: 0.13868390351302598,
• 10: 0.336067959653752,
• 11: 0.16407815738375933,
• 12: 0.33838887484747293,
• 13: 0.2871391639624871,
• 14: 0.22848023925633135}
7
 Eigenvector Centrality
• Define the centrality 𝑥′! of 𝑖 recursively in terms of
the centrality of its neighbors
𝑥!" = # 𝑥# 𝑜𝑟 𝑥!" = # 𝐴!( 𝑥(
#∈%(!) (
With initial vertex centrality 𝑥( = 1, ∀𝑗 (𝑖𝑛𝑐𝑙𝑢𝑑𝑖𝑛𝑔 𝑖)—we’ll
see why on next slide
• That is equivalent to:
𝑥! (𝑡) = # 𝐴!( 𝑥( (𝑡 − 1) The centrality of vertices
𝑖 and 𝑗 at time t and t-1,
( respectively
with the centrality at time t=0 being 𝑥( 0 = 1, ∀𝑗

8
In class: Eigenvector
Adjacency Centrality
matrix A for the graph to the right:
0 1 1 0 1 0
1 0 1 0 1 0
A= 1 1 0 0 0 1
0 0 0 0 0 0
1 1 0 0 0 1
0 0 1 0 1 0

𝑥+
𝑥,
Then the vector x(t) = : gives a random surfer’s behavior. Answer
𝑥-
the following questions based on the information above
9
In class: Eigenvector Centrality
• Q1: Find x(1). What does it represent?
0 1 1 0 1 0 1
1 0 1 0 1 0 1
Answ: 𝑥(1) = 1 1 0 0 0 1 1 = ?
0 0 0 0 0 0 1
1 1 0 0 0 1 1
0 0 1 0 1 0 1

10
In class: Eigenvector Centrality
• Q1: Find x(1). What does it represent?
0 1 1 0 1 0 1 3
1 0 1 0 1 0 1 3
Answ: 𝑥(1) = 1 1 0 0 0 1 1 = 3
0 0 0 0 0 0 1 0
1 1 0 0 0 1 1 3
0 0 1 0 1 0 1 2

The Degree Centrality vector

11
In class: Eigenvector Centrality
• Q2: Find x(2). What does it represent?
0 1 1 0 1 0 3
1 0 1 0 1 0 3
Answ: 𝑥(2) = 1 1 0 0 0 1 3 = ?
0 0 0 0 0 0 0
1 1 0 0 0 1 3
0 0 1 0 1 0 2

12
In class: Eigenvector Centrality
• Q2: Find x(2). What does it represent?
0 1 1 0 1 0 3 9
1 0 1 0 1 0 3 9
Answ: 𝑥(2) = 1 1 0 0 0 1 3 = 8
0 0 0 0 0 0 0 0
1 1 0 0 0 1 3 8
0 0 1 0 1 0 2 6

A weighted Degree Centrality vector (distance 2 or less)

13
In class: Eigenvector Centrality
• Q3: Find x(3). What does it represent?
0 1 1 0 1 0 9
1 0 1 0 1 0 9
Answ: 𝑥(3) = 1 1 0 0 0 1 8 = ?
0 0 0 0 0 0 0
1 1 0 0 0 1 8
0 0 1 0 1 0 6

14
In class: Eigenvector Centrality
• Q3: Find x(3). What does it represent?
0 1 1 0 1 0 9 25
1 0 1 0 1 0 9 25
Answ: 𝑥(3) = 1 1 0 0 0 1 8 = 24
0 0 0 0 0 0 0 0
1 1 0 0 0 1 8 24
0 0 1 0 1 0 6 16

Another weighted Degree Centrality vector (distance 3 or less)

15
In class: Eigenvector Centrality
Eigenvector centrality:
• 0: 0.49122209552166,
• 1: 0.49122209552166,
• 2: 0.4557991200411896,
• 3: 0,
• 4: 0.4557991200411896,
• 5: 0.31921157573304415

16
Discussion: What did you notice?
• What is x(3)?
1
1
Answ: 𝑥 3 = 𝐴(𝐴(𝐴 1 ))) = 𝐴" 𝑥 0
1
1
1
𝑥 3 depends on the centrality of its distance 3 or less
neighbors

17
Discussion: Can you generalize it?
• What is x(t)?
1
1
Answ: 𝑥 t = 𝐴(𝐴 … (𝐴 1 ))) = 𝐴# 𝑥 0 , 𝑡 > 0
1
1
1
𝑥 t depends on the centrality of its distance t or less
neighbors

18
 Eigenvector Centrality Derivation
• We can consolidate the eigenvector centralities of all the nodes in
a recursive formula with vectors:
x 𝑡 = 𝐴 8 𝐱(𝑡 − 1)
with the centrality at time t=0 being x 0 = 𝟏 (as a vector)

• Then, we solve: x 𝑡 = 𝐴. 8 𝐱(0), with x 0 = 𝟏

• Let 𝒗# be the eigenvectors of the adjacency matrix 𝐴
• Let 𝜆+ be the largest eigenvalue.
• Let x 0 = ∑# 𝑐# 𝑣# , be a linear combination of 𝒗# (eigenvectors
are orthogonal since 𝐴 is real and symmetric)
 Eigenvector Centrality Derivation
Facts from previous page:
• x 𝑡 = 𝐴. 8 𝐱(0), with x 0 = 1
• 𝒗𝒌 are the eigenvectors of the adjacency matrix A
• x 0 = ∑# 𝑐# 𝑣# is a linear combination of 𝒗𝒌
• 𝜆+ be the largest eigenvalue.
Then
x 𝑡 = 𝐴. 8 𝐱 0 = 𝐴. ∑# 𝑐# 𝑣# = ∑# 𝑐# 𝜆.# 𝑣# =
0"! 0"# 0"$ 0"%
= 𝜆+. ∑# 𝑐# " 𝑣# = 𝜆+. (𝑐+ 𝑣+ + 𝑐, 𝑣, + 𝑐1 𝑣1 …) x 𝑡 à 𝜆+. 𝑐+ 𝑣+
0# 0"# 0"# 0"#
"
0!
since 0" à0 as t à∞ (as you repeat the process)
#