business network analytics - uzha02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · business network...
TRANSCRIPT
![Page 1: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/1.jpg)
Business Network
Analytics
Sep, 2017
Daning Hu
Department of Informatics
University of Zurich
Business Intelligence Research Group
F Schweitzer et al. Science 2009
![Page 2: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/2.jpg)
What: Social Network Analysis
Social network analysis (SNA) is a set of metrics and
methods for systematically describing, modelling, and
analyzing relationships among actors.
Social network analysis (SNA)
is motivated by a structural intuition based on ties linking
social actors
is grounded in systematic empirical data
draws heavily on graphic imagery
relies on the use of mathematical and/or computational
models.
![Page 3: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/3.jpg)
3
Social Network Analysis: Topology Analysis
• Network level topology analysis takes a macro perspective to
describe the physical properties of network structures. Typical
network topological measures include:
– Centrality Measures: Degree, Betweenness, and Closeness
– Size and Density,
– Average Degree,
– Average Path Length: on average, the number of steps it takes to get
from one member of the network to another.
– Diameter
– Clustering Coefficient: a measure of an "all-my-friends-know-each-
other" property; small-world feature
1
)(
)1(
2)(
i
ii
i
iCoeffClusteringCC
kk
EiCC ki = Cd(i) = # of neighbors of node i
Ei = # of links actually exist between ki nodes
![Page 4: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/4.jpg)
Node Level Analysis: Node Centrality
• Node Centrality can be viewed as a measure of influence or
importance of nodes in a network.
• Degree
– the number of links that a node possesses in a network. In a directed
network, one must differentiate between in-links and out-links by calculating
in-degree and out-degree.
• Betweeness
– the number of shortest paths in a network that traverse through that node.
• Closeness
– the average distance that each node is from all other nodes in the network
![Page 5: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/5.jpg)
Node Level Analysis: Degree Centrality
From Steve Borgatti
![Page 6: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/6.jpg)
Node Level Analysis: Betweenness Centrality
From Steve Borgatti
![Page 7: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/7.jpg)
Node Level Analysis: Closeness Centrality
From Steve Borgatti
1
1
),()(
g
j
jiic nndnC
)1))((()(' gnCnC iCiC
Normalized Closeness:
![Page 8: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/8.jpg)
Node Level Analysis: Eigenvector Centrality
From Steve Borgatti
![Page 9: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/9.jpg)
Comparison of Centrality Measures
Low
Degree
Low
Closeness
Low
Betweenness
High Degree Embedded in
cluster that is far
from the rest of
the network
Ego's connections
are redundant -
communication
bypasses him/her
High Closeness Key player tied to
important
important/active
alters
Ego is near many
people, but so are
many others
High
Betweenness
Ego's few ties
are crucial for
network flow
Very rare cell.
Would mean
that ego
monopolizes the
ties from a small
number of
people to many
others.
![Page 10: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/10.jpg)
Identifying Key Players
10
(Node size proportional to betweenness centrality )
![Page 11: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/11.jpg)
Link Level Analysis: Length and Distance
![Page 12: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/12.jpg)
Link Level Analysis: Cut Points and Bridge
From Steve
Borgatti
![Page 13: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/13.jpg)
The Strength of Weak Tie (Granovetter 1973)
• Strong ties create transitivity
– Two nodes linked by a strong tie will have mutual acquaitances
• Ties that are part of transitive triples cannot be bridges
• Therefore, only weak ties can be bridges
– the value of weak ties!!
• Strong ties embeded in tight homophilous clusters, while
weak ties connect to diversity
– Weak ties is a major source of novel information
![Page 14: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/14.jpg)
14
Network Level Analysis
• Network Topology Analysis takes a macro perspective to study
the physical properties of network structures. Network topological
measures include:
– Size, i.e., number of nodes and links, Average Degree,
– Degree Distribution
– Average Path Length, Diameter
– Clustering Coefficient: a measure of an "all-my-friends-know-each-
other" property; small-world feature
– Centralization and Density
![Page 15: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/15.jpg)
15
Network Level Analysis: Degree Distribution
![Page 16: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/16.jpg)
16
Network Level Analysis: Diameter and
Average Path Length
![Page 17: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/17.jpg)
17
Network Level Analysis: Clustering Coefficient
![Page 18: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/18.jpg)
18
Network Level Analysis: Clustering Coefficient
![Page 19: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/19.jpg)
19
A Real Network: The MSN Messenger
![Page 20: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/20.jpg)
The MSN Global Communication Network
![Page 21: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/21.jpg)
The MSN Global Communication Network
![Page 22: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/22.jpg)
Degree Distribution
![Page 23: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/23.jpg)
Log-Log Degree Distribution
![Page 24: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/24.jpg)
Clustering
![Page 25: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/25.jpg)
Diameter
![Page 26: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/26.jpg)
To Sum Up
![Page 27: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/27.jpg)
The Null Model: Random Network (Graph)
![Page 28: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/28.jpg)
The Null Model: Random Network (Graph)
![Page 29: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/29.jpg)
Degree Distribution of Gnp
![Page 30: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/30.jpg)
Clustering Coefficient of Gnp
![Page 31: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/31.jpg)
The MSN Network vs. Gnp
![Page 32: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/32.jpg)
Real (MSN) Networks vs. Gnp
![Page 33: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/33.jpg)
Real (MSN) Networks vs. Gnp
![Page 34: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/34.jpg)
Evolution of a Random Network (Graph)
![Page 35: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/35.jpg)
Gnp Simulation Experiment
![Page 36: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/36.jpg)
Network Level Analysis: Centralization
Variance of the individual centrality scores.
gCnCSg
i
diDD /))((1
22
Freeman’s general formula for centralization (which ranges from 0 to 1):
)]2)(1[(
)()(1
*
gg
nCnCC
g
i iDD
D
Freeman: 1.0
Variance: 3.9
Freeman: .02
Variance: .17Freeman: 0.0
Variance: 0.0
![Page 37: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/37.jpg)
37
Network Level Analysis: Centralization
• Centralization: Degree to which network revolves around a
single node.
From Steve Borgatti
![Page 38: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/38.jpg)
Network Level Analysis: Network Density
• The density D of a network is the percentage of the links L
divided by the number of all possible links N(N-1)/2
![Page 39: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/39.jpg)
Our Approach
39
What Why How
Social network
analysis (Metrics)
Describe the changes
in network evolution
Temporal changes in
network topological
measures
Dynamic network
recovery
(Relational) data
mining
Econometric
identification of
casual Social and
Economic influence
Distinguishing
homophily
Confounding
factors
PSM, DID, RD,
etc.
Explanations
Combine social
science methods,
data mining, machine
learning with
econometric analysis
Predict link formation
Simulate the
evolution of networks
![Page 40: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/40.jpg)
Describe Network Evolution: A Global Terrorist Network
40
The Global Salafi Jihad (GSJ) network data is compiled by a
former CIA operation officer Dr. Marc Sageman - 366 terrorists
friendship, kinship, same religious leader, operational interactions, etc.
geographical origins, socio-economic status, education, etc.
when they join and leave GSJ
The goal of dynamic analysis
gain insights about the evolution of GSJ network
develop effective attack strategies to break down GSJ network
Sample data of GSJ terrorists
![Page 41: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/41.jpg)
41
a
![Page 42: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/42.jpg)
42
Stretching
Or Not?
Leonardo da Vinci
Vitruvian Man, 1487
![Page 43: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/43.jpg)
43
![Page 44: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/44.jpg)
Temporal Changes in Network-level Measures
Average Degree <k >
0
2
4
6
8
10
12
14
16
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
2003
deg
ree
<k>
Fig.1. The temporal changes in the (a)
average degree, (b) and (c) degree
distribution
Degree = number of links a node has
a
b
c
0.00
0.03
0.06
0.09
0.12
0.15
0.18
0.21
0.24
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49
pro
bab
ilit
y o
f d
eg
ree
1990
1991
1993
Poisson
0.00
0.03
0.06
0.09
0.12
0.15
0.18
0.21
0.24
1 4 7 10 13 16 19 22 25 28 31 34 37 40 43 46 49 52
pro
bab
ilit
y o
f d
eg
ree
1995
1997
1999
![Page 45: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/45.jpg)
Findings
45
There are three stages for the evolution of the GSJ network:
1989 - 1993 The emerging stage:
The network grows in size
Accelerated Growth - No. of edges increases faster than nodes
Random network topology (Poisson degree distribution)
1994 - 2000 The mature stage:
The size of the network reached its peak in 2000
Scale-free topology (Power-law degree distribution)
2001 - 2003 The disintegration stage:
Falling into small disconnected components after 9/11
![Page 46: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/46.jpg)
Temporal Changes in Node Centrality Measures
46
0
10
20
30
40
50
60
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
Degree
0
500
1000
1500
2000
2500
3000
3500
4000
4500
5000
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001
2002
Betweenness
Figure.2. Temporal changes in
Degree and Betweenness
centrality of Osama Bin Laden
Degree: No. of links a node has
Betweenness of a node i
No. of shortest paths from all nodes to
all others that pass through node i
Measure i’s influence on the traffic
(information, resource) flowing through it
![Page 47: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/47.jpg)
Findings and Possible Explanations
47
1994 – 1996: A sharp decrease in Bin Laden’s Betweenness
1994: Saudi revoked his citizenship and expelled him
1995: Went to Sudan and was expelled again under U.S. pressure
1996: Went to Afghanistan and established camps there
1998 –1999: Another sharp decrease in his Betweenness
After 1998 bombings of U.S. embassies, Bill Clinton ordered a freeze
on assets linked to bin Laden (top 10 most wanted)
August 1998: A failed assassination on him from U.S.
1999: UN imposed sanctions against Afghanistan to force the Taliban
to extradite him
![Page 48: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/48.jpg)
48
Dynamic Network Recovery
The GSJ Network: Small, Low Frequency (Yearly), AdHoc!
How about large, high-frequency longitudinal network data?
Email communication network
t t +1 t +2
t t +1 t +2
................ ................
Timeline
The relevancy horizon
Which past links are relevant to the current state of the network?
................
How frequent
to sample?
t, t + 2[ ]An instantaneous network
![Page 49: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/49.jpg)
Our Approach
49
What Why How
Social network
analysis (Metrics)
Describe the changes
in network evolution
Temporal changes in
network topological
measures
Dynamic network
recovery
(Relational) data
mining
Econometric
identification of
casual Social and
Economic influence
Distinguishing
homophily
Confounding
factors
PSM, DID, RD,
etc.
Explanations
Combine social
science methods,
data mining, machine
learning with
econometric analysis
Predict link formation
Simulate the
evolution of networks
![Page 50: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/50.jpg)
50
Dynamic Network Recovery
The GSJ Network: Small, Low Frequency (Yearly), AdHoc!
How about large, high-frequency longitudinal network data?
Email communication network
t t +1 t +2
t t +1 t +2
................ ................
Timeline
The relevancy horizon
Which past links are relevant to the current state of the network?
................
How frequent
to sample?
t, t + 2[ ]An instantaneous network
![Page 51: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/51.jpg)
51
Relevancy Horizon and Sampling Period
Recovering a set of instantaneous social networks from
longitudinal network data by setting a sliding window filter
The relevancy horizon : the maximum time length that a past event
(link) has impact on current network.
The sampling period : determines which events were considered to
be simultaneous and independent of each other
Timeline
![Page 52: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/52.jpg)
Research Testbed: A Narcotic Criminal Network
The COPLINK dataset contains 3 million police incident reports from the Tucson Police Department (1990 to 2006).
3 million incident reports and 1.44 million individuals
Their personal and sociological information (age, ethnicity, etc.)
Time information: when two individuals co-offend
AZ Inmate affiliation data: when and where an inmate was housed
A Narcotic Criminal Network
19,608 individuals involved in organized narcotic crimes
29,704 co-offending pairs (links)
52
COPLINK
Narcotic Data
Arizona Inmate
Data
Overlapped (identified by first
name, last name and DOB)
Number of People 36,548 165,540 19,608
Time Span 1990 - 2006 1985 - 2006 17 years
Table 1. Summary of the COPLINK dataset and the Arizona inmate dataset
![Page 53: Business Network Analytics - UZHa02c1b6d-b01f-4d1a-ba5f... · 2017-09-15 · Business Network Analytics Sep, 2017 Daning Hu Department of Informatics University of Zurich ... –](https://reader034.vdocument.in/reader034/viewer/2022042315/5f03fcf07e708231d40bc30f/html5/thumbnails/53.jpg)
Determine Sampling Period and Relevancy Horizon
The sampling period can be calculated based on
Nyquist–Shannon sampling theorem: , where is
the maximum frequency of link formation (i.e., co-offending a crime).
The relevancy horizon is determined by
Within-pair Response Time : the time gap between two
subsequent co-offendings by i and j.
53
d
max2/1 fmaxf
ijt
Cumulative Distribution of within-pair response time
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
1 51 101 165 262 451
tij (days)
p(t
ij)
90% of the response time gaps
are within = 210 days.
Set 210 days as the relevancy
horizon for our empirical analysis.
90.0t