Real-World Networks And their common properties 1. Macro-level - - PDF document
9/22/20 DCS/CSCI 2350 Social & Economic Networks What does a real-world network look like? Reading: Ch 2 of EK, Ch 2 & 3 of Jackson Graph visualization using Gephi Mohammad T . Irfan Email: mirfan@bowdoin.edu 1 Real-World Networks
9/22/20 1
1
3
9/22/20 2
4
u Intuitive example– world acquaintance
network
u Questions
u Is it connected? u How many giant components are there?
5
9/22/20 3
u Actor network
u Edge between two actors iff they appear together
in a movie
u 98% of 449,913 actors belong to the giant
component (IMDB, May 2000)
6
u Instant messaging
u Microsoft IM: one giant component in a network of
240 million users (2008) u Co-author network u Email u Biological networks (neural networks) u Technology networks (power grid) u The Internet (web of links)
Can you think of a network that doesn’t have any giant component?
8
9/22/20 4
High school relationships (1993-95)
9
u Proposition
u The average shortest path between any two nodes
in a connected component is “small” u Intuition
Also known as distance
11
9/22/20 5
u Hungarian author Frigyes Karinthy (1929
short story “Chain-Links”)
u John Guare’s play (1990)
& later movie
“A fascinating game grew out of this discussion. One of us suggested performing the following experiment to prove that the population of the Earth is closer together now than they have ever been before. We should select any person from the 1.5 billion inhabitants of the Earth – anyone, anywhere at all. He bet us that, using no more than five individuals, one of whom is a personal acquaintance, he could contact the selected individual using nothing except the network of personal acquaintances.”
12
14
9/22/20 6
15
u Only 64 out of 296 cases were successful u How useful? What is the implication?
u Milgram: “six worlds apart”
16
9/22/20 7
17
u Microsoft instant messenger (2008)
u 240M node network u Edge: Two-way conversation at some point during a
month-long observation period
u Average distance: 6.6
Fraction of pairs
this distance
18
9/22/20 8
u How to find the “right 6 people?”
u Breadth-first search (BFS) algorithm to find the
shortest path u Fun application– Bacon number
u Bacon number of an actor = distance from Kevin
Bacon
u Average Bacon number: 2.9 u https://oracleofbacon.org/
19
20
9/22/20 9
u Resulting graph: BFS tree
AKA "root" Other existing edges within a layer are not drawn here. Draw only the edges explored.
Yo u
Nodes whose distance have not yet been calculated and who have edges to nodes in the previous layer Distance = 1 Distance = 2 Distance = 3 Distance = 0 Your friends Friends of friends Friends of friends
22
ARPANET (1970)
23
9/22/20 10
Animal House Apollo 13 The Wild River Da Vinci Code Titanic Holiday Joe vs. Volcano High Noon Dial M for Murder The Eagle Has Landed Cold Mountain Hamlet Portrait of a Lady
Bill Paxton Tom Hanks Paul Herbert Yves Aubert Kate Winslet Kevin Bacon Meryl Streep Donald Sutherland John Belushi Kathleen Quinlan Lloyd Bridges Grace Kelly Patrick Allen Nicole Kidman John Gielgud
Charlie’s Angels
Bill Murray Cameron Diaz
Network among movie actors
24
u When all the edges have the same "weight"/dist. u Negative example: Frankfurt—Kassel—Munchen not shortest path
25
9/22/20 11
u Tree
u Connected, acyclic graph u Example: BFS tree
u Bipartite graph
u Two sets of nodes with no edge within the same set of
nodes
u Example: Network between movies and actors
Actors Movies
26
u What’s the probability of finding a node with degree k? u What fraction of nodes have degree k? Call it Pk.
27
9/22/20 12
u Power law distribution (or Pareto distrib.)
u Mathematical formulation u Scale-free networks
Extremely important Please take note
29
u Clustering coeff = Average probability that
two friends of a node are also friends
u How to calculate?
u Local clustering coeff. of node i,
Ci =
u Clustering coefficient of the whole network
= average Ci of all the nodes i
Actual # of edges among i’s friends Max possible # of edges among i’s friends Need to count di (di -1) / 2 where di = degree of i
30
9/22/20 13
u What is the clustering coefficient of this
network?
2 3 1 4 5
31
u “High” clustering coefficient is observed in
real-world networks
32
9/22/20 14
u Uzzi et al., 2007 u https://www.kellogg.northwestern.edu/facu
lty/uzzi/ftp/Uzzi_EuropeanManReview_2007. pdf
u N = # of nodes
k = Avg degree L = Avg shortest path length CC = Clustering coefficient
33 34
9/22/20 15
35
u Six Degrees, pg. 51
36
9/22/20 16
1.
Degree centrality
2.
Closeness centrality
3.
Betweenness centrality
4.
Prestige/eigenvector centrality Idea Math Example
37
u A node’s centrality = The node’s degree / (n-1) u Who is the most central here? u How about node 4 in this network?
1 2 3 4 6 7 5
38
9/22/20 17
u Idea: node i is very central if it’s pretty close to
the other nodes
u Avg distance from node i to all other nodes = u Closeness centrality of i = 1/Avg distance from i
dist(i, j)
j≠i
n −1
Need to do BFS with root i
39
u Compute the closeness centralities of nodes
1 and 4
1 2 3 4 6 7 5
40
9/22/20 18
u Idea: a node i is very central if a lot of
shortest paths go through i
u Betweenness centrality of i,
βi = Σ
u Florentine marriage: Medici most central
# of shortest paths between j and k passing through i # of shortest paths between j and k, irrespective of passing through i
j,k j ≠k≠i
41
u Compute the between centrality of nodes 1,
2, and 3
u β1 = 0 u β2 = 0 u β3 = ?
1 2 4 5 3 6
42
9/22/20 19
u Images from this tutorial:
http://www.intmath.com/matrices- determinants/3-matrices.php
u 4x1 matrix (AKA vector) u 3x3 matrix
43
u 2x3 matrix multiplied by 3x2 matrix u Result is a 2x2 matrix
must match
44
9/22/20 20
u Transpose operator: superscript T u (A B)T = BT AT A = 1 2 3 4 5 6 ! " # # # $ % & & & AT = 1 2 3 4 5 6 ! " # # $ % & &
45
u Idea (Phillip Bonacich, 1987): A node’s
importance is determined by its friends’ importance
u Mathematical formulation and example
46
9/22/20 21
47
u Tutorial on eigenvector
u Jackson’s Section 2.4 (Appendix)
48
9/22/20 22
49
50
9/22/20 23
u Download
u https://gephi.org/ u Windows: Gephi 0.9.2 will only run with Java 7 or
u How to find Java version in Windows?
https://www.java.com/en/download/help/version_ manual.xml
u Where to get Java for Windows?
https://www.java.com/en/download/
u Mac OS X: Java is bundled with the application so
it doesn't have to be installed separately. u Tutorial: http://bit.ly/gephi_tutorial u Dataset: http://bit.ly/gephi_dataset
51
Ranking: rank and color nodes, edges, and their labels by numeric properties Partition: partition nodes and edges Data Laboratory: Manipulate the input graph files (e.g., apply labels to nodes) Statistics: Computes graph-level properties. Some of them (e.g., Average Degree) must be done before using other features Filters: Filter out nodes/ed ges based
propertie s Useful filter: Topology à Degree Range T: Toggle showing node labels T: Toggle showing edge labels Slider: Tune the size of the node labels "Magnifying glass": Centers the graphics Layout: Select a graph drawing algorithm
Network among the characters of Les Miserables
Slider: Tune edge thickness Preview: Produces a nice visualization (next slide) Color palette for coloring schemes
52
9/22/20 24
Show Labels: Turn it on!
To save the visualization as a pdf file: File à Save
Refresh: Must click this button! Otherwise, nothing will be shown.
53
Term Meaning
betweeness centrality
how often the node appears on the shortest path between nodes in the network closeness centrality of a node average distance from that node to all other nodes in the network degree of a node the number of edges connected to the node (also connectedness); in a directed graph a node can have in- degree and out-degree measures diameter of a graph the longest shortest path between any two nodes in the graph directed graph this means relationships occur one way only (I follow you, but you do not follow me on Twitter); opposite of undirected (we are friends with each other on Facebook) eccentricity of a node the distance (shortest-path length) from the node to the farthest node from it in the network edge a representation of the connection between two nodes, expresses a relationship (a line) eigenvector centrality
in social network analysis, a measure of influence (a node is very influential if it is connected to other influential nodes) layout algorithms also known as graph drawing algorithm; e.g., force-directed drawing where linked nodes attract and non- linked nodes repel leaf node node with a single edge in a “tree-structured” graph modularity a measure of connectedness among groups of nodes (greater than 0.4 is usually considered meaningful) node also called a vertex by mathematicians; a person in a social network graph (a dot or bubble) distance from one node to another the length of the shortest path (counted in the number of edges) from one node to another path length the number of edges in a path singleton node or isolated node node with no edge/connection Red: Graph level Black: Node/edge level
54
9/22/20 25
u Gephi http://bit.ly/gephi_dataset (Les Miserables data) http://bit.ly/gephi_dolphin (Dolphin network data)
Dolphin network data: Social network (by association) among 62 dolphins in Doubtful Sound, New Zealand
55
57
9/22/20 26
See anything interesting?
58
u What are the differences among:
u Degree centrality u Closeness centrality u Betweenness centrality
59