mining billion-node graphs: patterns, generators and tools
DESCRIPTION
Mining Billion-node Graphs: Patterns, Generators and Tools. Christos Faloutsos CMU (on sabbatical at google). Thank you!. José Balcázar, Francesco Bonchi Aristides (Aris) Gionis Michèle Sebag Ricard Gavaldà. Our goal:. Open source system for mining huge graphs: - PowerPoint PPT PresentationTRANSCRIPT
![Page 1: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/1.jpg)
CMU SCS
Mining Billion-node Graphs: Patterns, Generators and Tools
Christos Faloutsos
CMU
(on sabbatical at google)
![Page 2: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/2.jpg)
CMU SCS
Thank you!
• José Balcázar,
• Francesco Bonchi
• Aristides (Aris) Gionis
• Michèle Sebag
• Ricard Gavaldà
ECML/PKDD'10 C. Faloutsos (CMU) 2
![Page 3: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/3.jpg)
CMU SCS
C. Faloutsos (CMU) 3
Our goal:
Open source system for mining huge graphs:
PEGASUS project (PEta GrAph mining System)
• www.cs.cmu.edu/~pegasus
• code and papers
ECML/PKDD'10
![Page 4: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/4.jpg)
CMU SCS
C. Faloutsos (CMU) 4
Outline
• Introduction – Motivation
• Problem#1: Patterns in graphs
• Problem#2: Tools
• Problem#3: Scalability
• Conclusions
ECML/PKDD'10
![Page 5: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/5.jpg)
CMU SCS
C. Faloutsos (CMU) 5
Graphs - why should we care?
Internet Map [lumeta.com]
Food Web [Martinez ’91]
Protein Interactions [genomebiology.com]
Friendship Network [Moody ’01]
ECML/PKDD'10
![Page 6: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/6.jpg)
CMU SCS
C. Faloutsos (CMU) 6
Graphs - why should we care?
• IR: bi-partite graphs (doc-terms)
• web: hyper-text graph
• ... and more:
D1
DN
T1
TM
... ...
ECML/PKDD'10
![Page 7: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/7.jpg)
CMU SCS
C. Faloutsos (CMU) 7
Graphs - why should we care?
• network of companies & board-of-directors members
• ‘viral’ marketing
• web-log (‘blog’) news propagation
• computer network security: email/IP traffic and anomaly detection
• ....
ECML/PKDD'10
![Page 8: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/8.jpg)
CMU SCS
C. Faloutsos (CMU) 8
Outline
• Introduction – Motivation
• Problem#1: Patterns in graphs– Static graphs– Weighted graphs– Time evolving graphs
• Problem#2: Tools
• Problem#3: Scalability
• Conclusions
ECML/PKDD'10
![Page 9: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/9.jpg)
CMU SCS
C. Faloutsos (CMU) 9
Problem #1 - network and graph mining
• What does the Internet look like?• What does FaceBook look like?
• What is ‘normal’/‘abnormal’?• which patterns/laws hold?
ECML/PKDD'10
![Page 10: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/10.jpg)
CMU SCS
C. Faloutsos (CMU) 10
Problem #1 - network and graph mining
• How does the Internet look like?• How does FaceBook look like?
• What is ‘normal’/‘abnormal’?• which patterns/laws hold?
– To spot anomalies (rarities), we have to discover patterns
ECML/PKDD'10
![Page 11: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/11.jpg)
CMU SCS
C. Faloutsos (CMU) 11
Problem #1 - network and graph mining
• How does the Internet look like?• How does FaceBook look like?
• What is ‘normal’/‘abnormal’?• which patterns/laws hold?
– To spot anomalies (rarities), we have to discover patterns
– Large datasets reveal patterns/anomalies that may be invisible otherwise…
ECML/PKDD'10
![Page 12: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/12.jpg)
CMU SCS
C. Faloutsos (CMU) 12
Graph mining
• Are real graphs random?
ECML/PKDD'10
![Page 13: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/13.jpg)
CMU SCS
C. Faloutsos (CMU) 13
Laws and patterns• Are real graphs random?• A: NO!!
– Diameter– in- and out- degree distributions– other (surprising) patterns
• So, let’s look at the data
ECML/PKDD'10
![Page 14: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/14.jpg)
CMU SCS
C. Faloutsos (CMU) 14
Solution# S.1
• Power law in the degree distribution [SIGCOMM99]
log(rank)
log(degree)
internet domains
att.com
ibm.com
ECML/PKDD'10
![Page 15: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/15.jpg)
CMU SCS
C. Faloutsos (CMU) 15
Solution# S.1
• Power law in the degree distribution [SIGCOMM99]
log(rank)
log(degree)
-0.82
internet domains
att.com
ibm.com
ECML/PKDD'10
![Page 16: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/16.jpg)
CMU SCS
C. Faloutsos (CMU) 16
Solution# S.2: Eigen Exponent E
• A2: power law in the eigenvalues of the adjacency matrix
E = -0.48
Exponent = slope
Eigenvalue
Rank of decreasing eigenvalue
May 2001
ECML/PKDD'10
![Page 17: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/17.jpg)
CMU SCS
C. Faloutsos (CMU) 17
Solution# S.2: Eigen Exponent E
• [Mihail, Papadimitriou ’02]: slope is ½ of rank exponent
E = -0.48
Exponent = slope
Eigenvalue
Rank of decreasing eigenvalue
May 2001
ECML/PKDD'10
![Page 18: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/18.jpg)
CMU SCS
C. Faloutsos (CMU) 18
But:
How about graphs from other domains?
ECML/PKDD'10
![Page 19: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/19.jpg)
CMU SCS
C. Faloutsos (CMU) 19
More power laws:
• web hit counts [w/ A. Montgomery]
Web Site Traffic
in-degree (log scale)
Count(log scale)
Zipf
userssites
``ebay’’
ECML/PKDD'10
![Page 20: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/20.jpg)
CMU SCS
C. Faloutsos (CMU) 20
epinions.com• who-trusts-whom
[Richardson + Domingos, KDD 2001]
(out) degree
count
trusts-2000-people user
ECML/PKDD'10
![Page 21: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/21.jpg)
CMU SCS
And numerous more
• # of sexual contacts
• Income [Pareto] –’80-20 distribution’
• Duration of downloads [Bestavros+]
• Duration of UNIX jobs (‘mice and elephants’)
• Size of files of a user
• …
• ‘Black swans’ECML/PKDD'10 C. Faloutsos (CMU) 21
![Page 22: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/22.jpg)
CMU SCS
C. Faloutsos (CMU) 22
Outline
• Introduction – Motivation
• Problem#1: Patterns in graphs– Static graphs
• degree, diameter, eigen,
• triangles
• cliques
– Weighted graphs– Time evolving graphs
• Problem#2: ToolsECML/PKDD'10
![Page 23: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/23.jpg)
CMU SCS
C. Faloutsos (CMU) 23
Solution# S.3: Triangle ‘Laws’
• Real social networks have a lot of triangles
ECML/PKDD'10
![Page 24: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/24.jpg)
CMU SCS
C. Faloutsos (CMU) 24
Solution# S.3: Triangle ‘Laws’
• Real social networks have a lot of triangles– Friends of friends are friends
• Any patterns?
ECML/PKDD'10
![Page 25: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/25.jpg)
CMU SCS
C. Faloutsos (CMU) 25
Triangle Law: #S.3 [Tsourakakis ICDM 2008]
ASNHEP-TH
Epinions X-axis: # of participatingtrianglesY: count (~ pdf)
ECML/PKDD'10
![Page 26: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/26.jpg)
CMU SCS
C. Faloutsos (CMU) 26
Triangle Law: #S.3 [Tsourakakis ICDM 2008]
ASNHEP-TH
Epinions
ECML/PKDD'10
X-axis: # of participatingtrianglesY: count (~ pdf)
![Page 27: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/27.jpg)
CMU SCS
C. Faloutsos (CMU) 27
Triangle Law: #S.4 [Tsourakakis ICDM 2008]
SNReuters
EpinionsX-axis: degreeY-axis: mean # trianglesn friends -> ~n1.6 triangles
ECML/PKDD'10
![Page 28: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/28.jpg)
CMU SCS
C. Faloutsos (CMU) 28
Triangle Law: Computations [Tsourakakis ICDM 2008]
But: triangles are expensive to compute(3-way join; several approx. algos)
Q: Can we do that quickly?
details
ECML/PKDD'10
![Page 29: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/29.jpg)
CMU SCS
C. Faloutsos (CMU) 29
Triangle Law: Computations [Tsourakakis ICDM 2008]
But: triangles are expensive to compute(3-way join; several approx. algos)
Q: Can we do that quickly?A: Yes!
#triangles = 1/6 Sum ( i3 )
(and, because of skewness (S2) , we only need the top few eigenvalues!
details
ECML/PKDD'10
![Page 30: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/30.jpg)
CMU SCS
C. Faloutsos (CMU) 30
Triangle Law: Computations [Tsourakakis ICDM 2008]
1000x+ speed-up, >90% accuracy
details
ECML/PKDD'10
![Page 31: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/31.jpg)
CMU SCS
EigenSpokes
B. Aditya Prakash, Mukund Seshadri, Ashwin Sridharan, Sridhar Machiraju and Christos Faloutsos: EigenSpokes: Surprising Patterns and Scalable Community Chipping in Large Graphs, PAKDD 2010, Hyderabad, India, 21-24 June 2010.
C. Faloutsos (CMU) 31ECML/PKDD'10
![Page 32: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/32.jpg)
CMU SCS
EigenSpokes• Eigenvectors of adjacency matrix
equivalent to singular vectors (symmetric, undirected graph)
32C. Faloutsos (CMU)ECML/PKDD'10
![Page 33: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/33.jpg)
CMU SCS
EigenSpokes• Eigenvectors of adjacency matrix
equivalent to singular vectors (symmetric, undirected graph)
33C. Faloutsos (CMU)ECML/PKDD'10
N
N
details
![Page 34: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/34.jpg)
CMU SCS
EigenSpokes• Eigenvectors of adjacency matrix
equivalent to singular vectors (symmetric, undirected graph)
34C. Faloutsos (CMU)ECML/PKDD'10
N
N
details
![Page 35: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/35.jpg)
CMU SCS
EigenSpokes• Eigenvectors of adjacency matrix
equivalent to singular vectors (symmetric, undirected graph)
35C. Faloutsos (CMU)ECML/PKDD'10
N
N
details
![Page 36: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/36.jpg)
CMU SCS
EigenSpokes
• EE plot:
• Scatter plot of scores of u1 vs u2
• One would expect– Many points @
origin– A few scattered
~randomly
C. Faloutsos (CMU) 36
u1
u2
ECML/PKDD'10
1st Principal component
2nd Principal component
![Page 37: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/37.jpg)
CMU SCS
EigenSpokes
• EE plot:
• Scatter plot of scores of u1 vs u2
• One would expect– Many points @
origin– A few scattered
~randomly
C. Faloutsos (CMU) 37
u1
u290o
ECML/PKDD'10
![Page 38: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/38.jpg)
CMU SCS
EigenSpokes - pervasiveness
•Present in mobile social graph across time and space
•Patent citation graph
38C. Faloutsos (CMU)ECML/PKDD'10
![Page 39: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/39.jpg)
CMU SCS
EigenSpokes - explanation
Near-cliques, or near-bipartite-cores, loosely connected
39C. Faloutsos (CMU)ECML/PKDD'10
![Page 40: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/40.jpg)
CMU SCS
EigenSpokes - explanation
Near-cliques, or near-bipartite-cores, loosely connected
40C. Faloutsos (CMU)ECML/PKDD'10
![Page 41: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/41.jpg)
CMU SCS
EigenSpokes - explanation
Near-cliques, or near-bipartite-cores, loosely connected
41C. Faloutsos (CMU)ECML/PKDD'10
![Page 42: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/42.jpg)
CMU SCS
EigenSpokes - explanation
Near-cliques, or near-bipartite-cores, loosely connected
So what? Extract nodes with high
scores high connectivity Good “communities”
spy plot of top 20 nodes
42C. Faloutsos (CMU)ECML/PKDD'10
![Page 43: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/43.jpg)
CMU SCS
Bipartite Communities!
magnified bipartite community
patents fromsame inventor(s)
cut-and-pastebibliography!
43C. Faloutsos (CMU)ECML/PKDD'10
![Page 44: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/44.jpg)
CMU SCS
C. Faloutsos (CMU) 44
Outline
• Introduction – Motivation
• Problem#1: Patterns in graphs– Static graphs
• degree, diameter, eigen,
• triangles
• cliques
– Weighted graphs– Time evolving graphs
• Problem#2: ToolsECML/PKDD'10
![Page 45: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/45.jpg)
CMU SCS
C. Faloutsos (CMU) 45
Observations on weighted graphs?
• A: yes - even more ‘laws’!
M. McGlohon, L. Akoglu, and C. Faloutsos Weighted Graphs and Disconnected Components: Patterns and a Generator. SIG-KDD 2008
ECML/PKDD'10
![Page 46: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/46.jpg)
CMU SCS
C. Faloutsos (CMU) 46
Observation W.1: Fortification
Q: How do the weights
of nodes relate to degree?
ECML/PKDD'10
![Page 47: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/47.jpg)
CMU SCS
C. Faloutsos (CMU) 47
Observation W.1: Fortification
More donors, more $ ?
$10
$5
ECML/PKDD'10
‘Reagan’
‘Clinton’$7
![Page 48: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/48.jpg)
CMU SCS
Edges (# donors)
In-weights($)
C. Faloutsos (CMU) 48
Observation W.1: fortification:Snapshot Power Law
• Weight: super-linear on in-degree • exponent ‘iw’: 1.01 < iw < 1.26
Orgs-Candidates
e.g. John Kerry, $10M received,from 1K donors
More donors, even more $
$10
$5
ECML/PKDD'10
![Page 49: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/49.jpg)
CMU SCS
C. Faloutsos (CMU) 49
Outline
• Introduction – Motivation
• Problem#1: Patterns in graphs– Static graphs – Weighted graphs– Time evolving graphs
• Problem#2: Tools
• …
ECML/PKDD'10
![Page 50: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/50.jpg)
CMU SCS
C. Faloutsos (CMU) 50
Problem: Time evolution• with Jure Leskovec (CMU ->
Stanford)
• and Jon Kleinberg (Cornell – sabb. @ CMU)
ECML/PKDD'10
![Page 51: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/51.jpg)
CMU SCS
C. Faloutsos (CMU) 51
T.1 Evolution of the Diameter
• Prior work on Power Law graphs hints at slowly growing diameter:– diameter ~ O(log N)– diameter ~ O(log log N)
• What is happening in real data?
ECML/PKDD'10
![Page 52: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/52.jpg)
CMU SCS
C. Faloutsos (CMU) 52
T.1 Evolution of the Diameter
• Prior work on Power Law graphs hints at slowly growing diameter:– diameter ~ O(log N)– diameter ~ O(log log N)
• What is happening in real data?
• Diameter shrinks over time
ECML/PKDD'10
![Page 53: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/53.jpg)
CMU SCS
C. Faloutsos (CMU) 53
T.1 Diameter – “Patents”
• Patent citation network
• 25 years of data
• @1999– 2.9 M nodes– 16.5 M edges
time [years]
diameter
ECML/PKDD'10
![Page 54: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/54.jpg)
CMU SCS
C. Faloutsos (CMU) 54
T.2 Temporal Evolution of the Graphs
• N(t) … nodes at time t
• E(t) … edges at time t
• Suppose thatN(t+1) = 2 * N(t)
• Q: what is your guess for E(t+1) =? 2 * E(t)
ECML/PKDD'10
![Page 55: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/55.jpg)
CMU SCS
C. Faloutsos (CMU) 55
T.2 Temporal Evolution of the Graphs
• N(t) … nodes at time t• E(t) … edges at time t• Suppose that
N(t+1) = 2 * N(t)
• Q: what is your guess for E(t+1) =? 2 * E(t)
• A: over-doubled!– But obeying the ``Densification Power Law’’
ECML/PKDD'10
![Page 56: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/56.jpg)
CMU SCS
C. Faloutsos (CMU) 56
T.2 Densification – Patent Citations
• Citations among patents granted
• @1999– 2.9 M nodes– 16.5 M edges
• Each year is a datapoint
N(t)
E(t)
1.66
ECML/PKDD'10
![Page 57: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/57.jpg)
CMU SCS
C. Faloutsos (CMU) 57
Outline
• Introduction – Motivation
• Problem#1: Patterns in graphs– Static graphs – Weighted graphs– Time evolving graphs
• Problem#2: Tools
• …
ECML/PKDD'10
![Page 58: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/58.jpg)
CMU SCS
C. Faloutsos (CMU) 58
More on Time-evolving graphs
M. McGlohon, L. Akoglu, and C. Faloutsos Weighted Graphs and Disconnected Components: Patterns and a Generator. SIG-KDD 2008
ECML/PKDD'10
![Page 59: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/59.jpg)
CMU SCS
C. Faloutsos (CMU) 59
Observation T.3: NLCC behaviorQ: How do NLCC’s emerge and join with the
GCC?
(``NLCC’’ = non-largest conn. components)
– Do they continue to grow in size?
– or do they shrink?
– or stabilize?
ECML/PKDD'10
![Page 60: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/60.jpg)
CMU SCS
C. Faloutsos (CMU) 60
Observation T.3: NLCC behaviorQ: How do NLCC’s emerge and join with the
GCC?
(``NLCC’’ = non-largest conn. components)
– Do they continue to grow in size?
– or do they shrink?
– or stabilize?
ECML/PKDD'10
![Page 61: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/61.jpg)
CMU SCS
C. Faloutsos (CMU) 61
Observation T.3: NLCC behavior
• After the gelling point, the GCC takes off, but NLCC’s remain ~constant (actually, oscillate).
IMDB
CC size
Time-stampECML/PKDD'10
![Page 62: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/62.jpg)
CMU SCS
C. Faloutsos (CMU) 62
Timing for Blogs
• with Mary McGlohon (CMU->google)
• Jure Leskovec (CMU->Stanford)
• Natalie Glance (now at Google)
• Mat Hurst (now at MSR)
[SDM’07]
ECML/PKDD'10
![Page 63: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/63.jpg)
CMU SCS
C. Faloutsos (CMU) 63
T.4 : popularity over time
Post popularity drops-off – exponentially?
lag: days after post
# in links
1 2 3
@t
@t + lag
ECML/PKDD'10
![Page 64: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/64.jpg)
CMU SCS
C. Faloutsos (CMU) 64
T.4 : popularity over time
Post popularity drops-off – exponentially?POWER LAW!Exponent?
# in links(log)
1 2 3 days after post(log)
ECML/PKDD'10
![Page 65: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/65.jpg)
CMU SCS
C. Faloutsos (CMU) 65
T.4 : popularity over time
Post popularity drops-off – exponentially?POWER LAW!Exponent? -1.6 • close to -1.5: Barabasi’s stack model• and like the zero-crossings of a random walk
# in links(log)
1 2 3
-1.6
days after post(log)
ECML/PKDD'10
![Page 66: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/66.jpg)
CMU SCS
ECML/PKDD'10 C. Faloutsos (CMU) 66
-1.5 slope
J. G. Oliveira & A.-L. Barabási Human Dynamics: The Correspondence Patterns of Darwin and Einstein. Nature 437, 1251 (2005) . [PDF]
![Page 67: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/67.jpg)
CMU SCS
T.5: duration of phonecalls
Surprising Patterns for the Call Duration Distribution of Mobile Phone Users
Pedro O. S. Vaz de Melo, Leman
Akoglu, Christos Faloutsos, Antonio A. F. Loureiro
PKDD 2010
ECML/PKDD'10 C. Faloutsos (CMU) 67
![Page 68: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/68.jpg)
CMU SCS
Probably, power law (?)
ECML/PKDD'10 C. Faloutsos (CMU) 68
??
![Page 69: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/69.jpg)
CMU SCS
No Power Law!
ECML/PKDD'10 C. Faloutsos (CMU) 69
![Page 70: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/70.jpg)
CMU SCS
‘TLaC: Lazy Contractor’
• The longer a task (phonecall) has taken,
• The even longer it will take
ECML/PKDD'10 C. Faloutsos (CMU) 70
Odds ratio=
Casualties(<x):Survivors(>=x)
== power law
![Page 71: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/71.jpg)
CMU SCS
71
Data Description
Data from a private mobile operator of a large city 4 months of data 3.1 million users more than 1 billion phone records
ECML/PKDD'10 C. Faloutsos (CMU)
![Page 72: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/72.jpg)
CMU SCS
C. Faloutsos (CMU) 72
Outline
• Introduction – Motivation
• Problem#1: Patterns in graphs
• Problem#2: Tools– CenterPiece Subgraphs– OddBall (anomaly detection)– PEGASUS
• Problem#3: Scalability
• Conclusions
ECML/PKDD'10
![Page 73: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/73.jpg)
CMU SCS
CenterPiece Subgraphs
• Hanghang TONG et al, KDD’06
C. Faloutsos (CMU) 73ECML/PKDD'10
![Page 74: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/74.jpg)
CMU SCS
Center-Piece Subgraph Discovery [Tong+ KDD 06]
Original Graph
Q: Who is the most central node wrt the black nodes?
(e.g., master-mind criminal, common advisor/collaborator, etc)
Input
B
A
C
74C. Faloutsos (CMU)ECML/PKDD'10
![Page 75: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/75.jpg)
CMU SCS
75
B
A
C B
A
C
Center-Piece Subgraph Discovery [Tong+ KDD 06]
Q: How to find hub for the query nodes?
Input: original graph Output: CePS
CePS Node
C. Faloutsos (CMU)A: Combine proximity scores (RWR)ECML/PKDD'10
![Page 76: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/76.jpg)
CMU SCS
CePS: Example (AND Query)
76
?
C. Faloutsos (CMU)ECML/PKDD'10
DBLP co-authorship network: -400,000 authors, 2,000,000 edges
Code at: http://www.cs.cmu.edu/~htong/soft.htm
![Page 77: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/77.jpg)
CMU SCS
CePS: Example (AND Query)
R. Agrawal Jiawei Han
V. Vapnik M. Jordan
H.V. Jagadish
Laks V.S. Lakshmanan
Heikki Mannila
Christos Faloutsos
Padhraic Smyth
Corinna Cortes
15 1013
1 1
6
1 1
4 Daryl Pregibon
10
2
11
3
16
77C. Faloutsos (CMU)ECML/PKDD'10
![Page 78: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/78.jpg)
CMU SCS
C. Faloutsos (CMU) 78
Outline
• Introduction – Motivation
• Problem#1: Patterns in graphs
• Problem#2: Tools– CenterPiece Subgraphs– OddBall (anomaly detection)
• Problem#3: Scalability - PEGASUS
• Conclusions
ECML/PKDD'10
![Page 79: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/79.jpg)
CMU SCS
OddBall: Spotting Anomalies in Weighted Graphs
Leman Akoglu, Mary McGlohon, Christos Faloutsos
Carnegie Mellon University
School of Computer Science
PAKDD 2010, Hyderabad, India
![Page 80: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/80.jpg)
CMU SCS
Main idea
For each node,
• extract ‘ego-net’ (=1-step-away neighbors)
• Extract features (#edges, total weight, etc etc)
• Compare with the rest of the population
C. Faloutsos (CMU) 80ECML/PKDD'10
![Page 81: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/81.jpg)
CMU SCS
What is an egonet?
ego
81
egonet
C. Faloutsos (CMU)ECML/PKDD'10
![Page 82: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/82.jpg)
CMU SCS
Selected Features
Ni: number of neighbors (degree) of ego i Ei: number of edges in egonet i Wi: total weight of egonet i λw,i: principal eigenvalue of the weighted
adjacency matrix of egonet I
82C. Faloutsos (CMU)ECML/PKDD'10
![Page 83: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/83.jpg)
CMU SCS
Near-Clique/Star
83ECML/PKDD'10 C. Faloutsos (CMU)
![Page 84: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/84.jpg)
CMU SCS
Near-Clique/Star
84C. Faloutsos (CMU)ECML/PKDD'10
![Page 85: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/85.jpg)
CMU SCS
C. Faloutsos (CMU) 85
Outline
• Introduction – Motivation
• Problem#1: Patterns in graphs
• Problem#2: Tools– CenterPiece Subgraphs– OddBall (anomaly detection)
• Problem#3: Scalability -PEGASUS
• Conclusions
ECML/PKDD'10
![Page 86: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/86.jpg)
CMU SCS
ECML/PKDD'10 C. Faloutsos (CMU) 86
Scalability• Google: > 450,000 processors in clusters of ~2000
processors each [Barroso, Dean, Hölzle, “Web Search for a Planet: The Google Cluster Architecture” IEEE Micro 2003]
• Yahoo: 5Pb of data [Fayyad, KDD’07]• Problem: machine failures, on a daily basis• How to parallelize data mining tasks, then?• A: map/reduce – hadoop (open-source clone)
http://hadoop.apache.org/
![Page 87: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/87.jpg)
CMU SCS
C. Faloutsos (CMU) 87
Centralized Hadoop/PEGASUS
Degree Distr. old old
Pagerank old old
Diameter/ANF old HERE
Conn. Comp old HERE
Triangles done
Visualization started
Outline – Algorithms & results
ECML/PKDD'10
![Page 88: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/88.jpg)
CMU SCS
HADI for diameter estimation
• Radius Plots for Mining Tera-byte Scale Graphs U Kang, Charalampos Tsourakakis, Ana Paula Appel, Christos Faloutsos, Jure Leskovec, SDM’10
• Naively: diameter needs O(N**2) space and up to O(N**3) time – prohibitive (N~1B)
• Our HADI: linear on E (~10B)– Near-linear scalability wrt # machines– Several optimizations -> 5x faster
C. Faloutsos (CMU) 88ECML/PKDD'10
![Page 89: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/89.jpg)
CMU SCS
????????
19+ [Barabasi+]
89C. Faloutsos (CMU)
Radius
Count
ECML/PKDD'10
~1999, ~1M nodes
![Page 90: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/90.jpg)
CMU SCS
YahooWeb graph (120Gb, 1.4B nodes, 6.6 B edges)• Largest publicly available graph ever studied.
????????
19+ [Barabasi+]
90C. Faloutsos (CMU)
Radius
Count
ECML/PKDD'10
??
~1999, ~1M nodes
![Page 91: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/91.jpg)
CMU SCS
YahooWeb graph (120Gb, 1.4B nodes, 6.6 B edges)• Largest publicly available graph ever studied.
????????
19+? [Barabasi+]
91C. Faloutsos (CMU)
Radius
Count
ECML/PKDD'10
14 (dir.)
~7 (undir.)
![Page 92: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/92.jpg)
CMU SCS
YahooWeb graph (120Gb, 1.4B nodes, 6.6 B edges)•7 degrees of separation (!)•Diameter: shrunk
????????
19+? [Barabasi+]
92C. Faloutsos (CMU)
Radius
Count
ECML/PKDD'10
14 (dir.)
~7 (undir.)
![Page 93: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/93.jpg)
CMU SCS
YahooWeb graph (120Gb, 1.4B nodes, 6.6 B edges)• effective diameter: surprisingly small.• Multi-modality: probably mixture of cores .
93C. Faloutsos (CMU)ECML/PKDD'10
![Page 94: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/94.jpg)
CMU SCS
94C. Faloutsos (CMU)
YahooWeb graph (120Gb, 1.4B nodes, 6.6 B edges)• effective diameter: surprisingly small.• Multi-modality: probably mixture of cores .
ECML/PKDD'10
![Page 95: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/95.jpg)
CMU SCS
Radius Plot of GCC of YahooWeb.
95C. Faloutsos (CMU)ECML/PKDD'10
![Page 96: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/96.jpg)
CMU SCS
Running time - Kronecker and Erdos-Renyi Graphs with billions edges.
details
![Page 97: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/97.jpg)
CMU SCS
C. Faloutsos (CMU) 97
Centralized Hadoop/PEGASUS
Degree Distr. old old
Pagerank old old
Diameter/ANF old HERE
Conn. Comp old HERE
Triangles done
Visualization started
Outline – Algorithms & results
ECML/PKDD'10
![Page 98: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/98.jpg)
CMU SCS
Generalized Iterated Matrix Vector Multiplication (GIMV)
C. Faloutsos (CMU) 98
PEGASUS: A Peta-Scale Graph Mining System - Implementation and Observations. U Kang, Charalampos E. Tsourakakis, and Christos Faloutsos. (ICDM) 2009, Miami, Florida, USA. Best Application Paper (runner-up).
ECML/PKDD'10
![Page 99: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/99.jpg)
CMU SCS
Generalized Iterated Matrix Vector Multiplication (GIMV)
C. Faloutsos (CMU) 99
• PageRank• proximity (RWR)• Diameter• Connected components• (eigenvectors, • Belief Prop. • … )
Matrix – vectorMultiplication
(iterated)
ECML/PKDD'10
details
![Page 100: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/100.jpg)
CMU SCS
100
Example: GIM-V At Work
• Connected Components
Size
Count
C. Faloutsos (CMU)ECML/PKDD'10
![Page 101: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/101.jpg)
CMU SCS
101
Example: GIM-V At Work
• Connected Components
Size
Count
C. Faloutsos (CMU)ECML/PKDD'10
~0.7B singleton nodes
![Page 102: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/102.jpg)
CMU SCS
102
Example: GIM-V At Work
• Connected Components
Size
Count
C. Faloutsos (CMU)ECML/PKDD'10
![Page 103: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/103.jpg)
CMU SCS
103
Example: GIM-V At Work
• Connected Components
Size
Count300-size
cmptX 500.Why?
1100-size cmptX 65.Why?
C. Faloutsos (CMU)ECML/PKDD'10
![Page 104: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/104.jpg)
CMU SCS
104
Example: GIM-V At Work
• Connected Components
Size
Count
suspiciousfinancial-advice sites
(not existing now)
C. Faloutsos (CMU)ECML/PKDD'10
![Page 105: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/105.jpg)
CMU SCS
105
GIM-V At Work• Connected Components over Time
• LinkedIn: 7.5M nodes and 58M edges
Stable tail slopeafter the gelling point
C. Faloutsos (CMU)ECML/PKDD'10
![Page 106: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/106.jpg)
CMU SCS
C. Faloutsos (CMU) 106
Outline
• Introduction – Motivation
• Problem#1: Patterns in graphs
• Problem#2: Tools
• Problem#3: Scalability
• Conclusions
ECML/PKDD'10
![Page 107: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/107.jpg)
CMU SCS
C. Faloutsos (CMU) 107
OVERALL CONCLUSIONS – low level:
• Several new patterns (fortification, triangle-laws, conn. components, etc)
• New tools:
– CenterPiece Subgraphs, anomaly detection (OddBall)
• Scalability: PEGASUS / hadoop
ECML/PKDD'10
![Page 108: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/108.jpg)
CMU SCS
C. Faloutsos (CMU) 108
OVERALL CONCLUSIONS – high level
• Large datasets reveal patterns/outliers that are invisible otherwise
• Terrific opportunities
– Large datasets, easily(*) available PLUS
– s/w and h/w developments
ECML/PKDD'10
![Page 109: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/109.jpg)
CMU SCS
C. Faloutsos (CMU) 109
References• Leman Akoglu, Christos Faloutsos: RTG: A Recursive
Realistic Graph Generator Using Random Typing. ECML/PKDD (1) 2009: 13-28
• Deepayan Chakrabarti, Christos Faloutsos: Graph mining: Laws, generators, and algorithms. ACM Comput. Surv. 38(1): (2006)
ECML/PKDD'10
![Page 110: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/110.jpg)
CMU SCS
C. Faloutsos (CMU) 110
References• Deepayan Chakrabarti, Yang Wang, Chenxi Wang, Jure
Leskovec, Christos Faloutsos: Epidemic thresholds in real networks. ACM Trans. Inf. Syst. Secur. 10(4): (2008)
• Deepayan Chakrabarti, Jure Leskovec, Christos Faloutsos, Samuel Madden, Carlos Guestrin, Michalis Faloutsos: Information Survival Threshold in Sensor and P2P Networks. INFOCOM 2007: 1316-1324
ECML/PKDD'10
![Page 111: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/111.jpg)
CMU SCS
C. Faloutsos (CMU) 111
References• Christos Faloutsos, Tamara G. Kolda, Jimeng Sun:
Mining large graphs and streams using matrix and tensor tools. Tutorial, SIGMOD Conference 2007: 1174
ECML/PKDD'10
![Page 112: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/112.jpg)
CMU SCS
C. Faloutsos (CMU) 112
References• T. G. Kolda and J. Sun. Scalable Tensor
Decompositions for Multi-aspect Data Mining. In: ICDM 2008, pp. 363-372, December 2008.
ECML/PKDD'10
![Page 113: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/113.jpg)
CMU SCS
C. Faloutsos (CMU) 113
References• Jure Leskovec, Jon Kleinberg and Christos Faloutsos
Graphs over Time: Densification Laws, Shrinking Diameters and Possible Explanations, KDD 2005 (Best Research paper award).
• Jure Leskovec, Deepayan Chakrabarti, Jon M. Kleinberg, Christos Faloutsos: Realistic, Mathematically Tractable Graph Generation and Evolution, Using Kronecker Multiplication. PKDD 2005: 133-145
ECML/PKDD'10
![Page 114: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/114.jpg)
CMU SCS
C. Faloutsos (CMU) 114
References• Jimeng Sun, Yinglian Xie, Hui Zhang, Christos
Faloutsos. Less is More: Compact Matrix Decomposition for Large Sparse Graphs, SDM, Minneapolis, Minnesota, Apr 2007.
• Jimeng Sun, Spiros Papadimitriou, Philip S. Yu, and Christos Faloutsos, GraphScope: Parameter-free Mining of Large Time-evolving Graphs ACM SIGKDD Conference, San Jose, CA, August 2007
ECML/PKDD'10
![Page 115: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/115.jpg)
CMU SCS
References
• Jimeng Sun, Dacheng Tao, Christos Faloutsos: Beyond streams and graphs: dynamic tensor analysis. KDD 2006: 374-383
ECML/PKDD'10 C. Faloutsos (CMU) 115
![Page 116: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/116.jpg)
CMU SCS
C. Faloutsos (CMU) 116
References
• Hanghang Tong, Christos Faloutsos, and Jia-Yu Pan, Fast Random Walk with Restart and Its Applications, ICDM 2006, Hong Kong.
• Hanghang Tong, Christos Faloutsos, Center-Piece Subgraphs: Problem Definition and Fast Solutions, KDD 2006, Philadelphia, PA
ECML/PKDD'10
![Page 117: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/117.jpg)
CMU SCS
C. Faloutsos (CMU) 117
References
• Hanghang Tong, Christos Faloutsos, Brian Gallagher, Tina Eliassi-Rad: Fast best-effort pattern matching in large attributed graphs. KDD 2007: 737-746
ECML/PKDD'10
![Page 118: Mining Billion-node Graphs: Patterns, Generators and Tools](https://reader035.vdocuments.us/reader035/viewer/2022062802/56814605550346895db3103c/html5/thumbnails/118.jpg)
CMU SCS
C. Faloutsos (CMU) 118
Project infowww.cs.cmu.edu/~pegasus
Google: pegasus cmu
Akoglu, Leman
Chau, Polo
Kang, U
McGlohon, Mary
Tong, Hanghang
Prakash,Aditya
ECML/PKDD'10
Thanks to: NSF IIS-0705359, IIS-0534205, CTA-INARC; Yahoo (M45), LLNL, IBM, SPRINT, Google, INTEL, HP, iLab