SlideShare a Scribd company logo
Spectral Analysis of Signed Graphs for
Clustering, Prediction and Visualization
Jérôme Kunegis¹, Stephan Schmidt¹, Andreas Lommatzsch¹ & Jürgen Lerner²
¹DAI Lab, Technische Universität Berlin, ²Universität Konstanz, Germany
10th
SIAM International Conference on Data Mining, April 29–May 1, Columbus, Ohio
Kunegis et al. Spectral Analysis of Signed Graphs 2
Introduction: Negative Edges
Some websites allow you to have foes:
Example: Slashdot Zoo (Kunegis 2009)
Kunegis et al. Spectral Analysis of Signed Graphs 3
Introduction: Signed Graphs
• The resulting social network is signed
• Edges are positive or negative
• In this talk: we use the graph Laplacian to study signed graphs
Example:
Slashdot Zoo (Kunegis 2009)
me
Friend ofFoe of
Fan ofFreak of
Kunegis et al. Spectral Analysis of Signed Graphs 4
Outline
Introduction: Signed Graphs
1. Negative Edges and the Laplacian
2. Balance, Conflict and the Graph Spectrum
3. Communities, Cuts and Clustering
4. Resistance, Conductivity and Link Prediction
Discussion
Kunegis et al. Spectral Analysis of Signed Graphs 5
1. Negative Edges and the Laplacian
Graph drawing: Place each node at the center of its neighbors
v0
= (1/3) (v1
+ v2
+ v3
)
Algebraically: D v = A v
Solution 1: Upper eigenvectors of D−1
A using A = {0, 1}n×n
Solution 2: Lower eigenvectors of D – A and Dii
= Σj
Aij
We look at solution 2: L = D − A is the Laplacian matrix
v0
v1
v2
v3
Kunegis et al. Spectral Analysis of Signed Graphs 6
Drawing Signed Graphs
• Replace ‘negative’ neighbors by their
antipodal points
v0
= (1/3) (−v1
+ v2
+ v3
)
Solution: lower eigenvectors of L = D − A
Using A = {0, −1, +1}n×n
And Dii
= Σj
| Aij
|
v0
v1
v2
v3
−v1
Kunegis et al. Spectral Analysis of Signed Graphs 7
Example: Synthetic Graph
Unsigned Graph Drawing → Signed Graph Drawing
Kunegis et al. Spectral Analysis of Signed Graphs 8
2. Balance, Conflict and the Graph Spectrum
• ‘Balanced’ graphs have a
perfect 2-clustering
• Invert all negative edges
• Effect on the Laplacian
decomposition: Inversion of
all eigenvectors of one
cluster
• Therefore: The spectrum of a
balanced graph is the same
as for the underlying unsigned
graph (λ₁ = 0)
Kunegis et al. Spectral Analysis of Signed Graphs 9
The Laplacian Spectrum of Unbalanced Graphs
• Networks with conflict contain odd
cycles
• The Laplacian is always positive
semidefinite
xT
Lx = Σij
|Aij
|(xi
− sgn(Aij
) xj
)² ≥ 0
• In unbalanced networks: λ₁ > 0
Kunegis et al. Spectral Analysis of Signed Graphs 10
Algebraic Conflict
• λ₁ denotes conflict
Network λ₁₁₁₁
MovieLens 100k 0.4285
MovieLens 1M 0.3761
Jester 0.06515
MovieLens 10M 0.006183
Slashdot Zoo 0.006183
Epinions 0.004438
Conflict
For effect of size, see Appendix
Kunegis et al. Spectral Analysis of Signed Graphs 11
3. Communities, Cuts and Clustering
The tribal groups of the Eastern Central Highlands of New Guinea can
be friends (‘rova’) or enemies (‘hina’)
Graphic uses two lower eigenvectors of L
Kunegis et al. Spectral Analysis of Signed Graphs 12
Finding Communities
The Laplacian matrix finds communities:
• Communities are
connected by many
positive edges
• Community are
separated by many
negative edges
Kunegis et al. Spectral Analysis of Signed Graphs 13
Signed Spectral Clustering
• Compute the d lower eigenvectors of L
• Use k-means to cluster nodes in this d-dimensional space
• Minimize signed normalized cut between communities X and Y
SNC(X, Y) = (|X|−1
+ |Y|−1
) · (2 pos(X, Y) + neg(X, X) + neg(Y, Y))
pos/neg: number of positive/negative edges between communities
Kunegis et al. Spectral Analysis of Signed Graphs 14
Example: Wikipedia Reverts
• Users revert users on controversial Wikipedia article ‘Criticism of
Prem Rawat’
• All edges are negative
• Distance to center normalized
to unit
• Four clusters are apparent
Kunegis et al. Spectral Analysis of Signed Graphs 15
4. Resistance, Conductivity and Link Prediction
• Consider a network of electrical resistances:
• Between any two nodes, the network has an effective resistance
• The resistance distance is a squared Euclidean metric
Kunegis et al. Spectral Analysis of Signed Graphs 16
Link Prediction
• The resistance distance can be used for link prediction:
– Long paths count less
– Parallel paths count more
dist(i,j) = (L+
)ii
+ (L+
)jj
− (L+
)ij
− (L+
)ji
• Problem: How to handle negative edges?
Kunegis et al. Spectral Analysis of Signed Graphs 17
Voltage Inversion
• Solution: inverting amplifier
dist(i,j) = (L+
)ii
+ (L+
)jj
− (L+
)ij
− (L+
)ji
• Using signed Laplacian L
• Is squared Euclidean because L is positive semidefinite
−w
w −
Kunegis et al. Spectral Analysis of Signed Graphs 18
• Task: Predict the sign of new links
• Problem: Find a function F(A) = B
Evaluation: Link Sign Prediction
Known positive links (A)
Links to be predicted (B)
Known negative links (A)
Kunegis et al. Spectral Analysis of Signed Graphs 19
Graph Kernels
Link prediction functions using the Laplacian:
• L+ – Signed Laplacian kernel
• (I + αL)−1
– Signed regularized Laplacian kernel
• exp(−αL) – Signed ‘heat diffusion’
Other link prediction functions:
• (A)k
– Rank reduction
• exp(A) – Matrix exponential
• Poly(A) – Path counting
Kunegis et al. Spectral Analysis of Signed Graphs 20
Evaluation Results
• MovieLens: predict good / bad rating
• Best rating prediction: signed regularized Laplacian graph kernel
Link Prediction RMSE
Rank reduction 0.838
Path counting 0.840
Matrix exponential 0.839
Signed resistance distance 0.812
Signed regularized Laplacian 0.778
Signed heat diffusion 0.789
Kunegis et al. Spectral Analysis of Signed Graphs 21
Summary
The Laplacian matrix applies to signed graphs
The Laplacian spectrum denotes graph conflict
The signed Laplacian arises in several ways:
For graph drawing, the Laplacian implements antipodal proximity
For clustering, the Laplacian implements signed cuts
As an interpretation of negation as inversion of electrical
potential
Thank You
Kunegis et al. Spectral Analysis of Signed Graphs 23
References
P. Hage, F. Harary. Structural models in anthropology, Cambridge
University Press, 1983.
F. Harary. On the notion of balance of a signed graph, Michigan Math.
J., 2:143–146, 1953.
J. Kunegis, A. Lommatzsch, C. Bauckhage, The Slashdot Zoo: Mining
a social network with negative edges, Proc. Int. World Wide Web
Conf., pages 741–750, 2009.
J. Leskovec, Daniel Huttenlocher, Jon Kleinberg, Predicting positive
and negative links in online social networks, Proc. Int. World Wide
Web Conf., 2010.
Kunegis et al. Spectral Analysis of Signed Graphs 24
Appendix – Balance vs Volume
Kunegis et al. Spectral Analysis of Signed Graphs 25
Appendix – Scalability
• Evaluation results in function of reduced rank k
Kunegis et al. Spectral Analysis of Signed Graphs 26
Appendix -- Balance, Conflict and the Graph Spectrum
Look at triads of users (Harary 1953):
• In balanced triangles, the multiplication rule holds
• If it doesn't, there is conflict
Balance:
Conflict:
Kunegis et al. Spectral Analysis of Signed Graphs 27
The Signed Clustering Coefficient
• How many triangles are balanced?
Cs
= (#balanced − #unbalanced) / #possible
• This measure is local, not global (Kunegis 2009)
± uv ?
u v
Kunegis et al. Spectral Analysis of Signed Graphs 28
Introduction: Networks
• Many web sites allow you to have friends:
Example: Facebook

More Related Content

Similar to Spectral Analysis of Signed Graphs for Clustering, Prediction and Visualization

Graph Evolution Models
Graph Evolution ModelsGraph Evolution Models
Graph Evolution Models
Carlos Castillo (ChaTo)
 
240401_JW_labseminar[LINE: Large-scale Information Network Embeddin].pptx
240401_JW_labseminar[LINE: Large-scale Information Network Embeddin].pptx240401_JW_labseminar[LINE: Large-scale Information Network Embeddin].pptx
240401_JW_labseminar[LINE: Large-scale Information Network Embeddin].pptx
thanhdowork
 
Spectral clustering with motifs and higher-order structures
Spectral clustering with motifs and higher-order structuresSpectral clustering with motifs and higher-order structures
Spectral clustering with motifs and higher-order structures
David Gleich
 
Simplicial closure and higher-order link prediction
Simplicial closure and higher-order link predictionSimplicial closure and higher-order link prediction
Simplicial closure and higher-order link prediction
Austin Benson
 
Random graph models
Random graph modelsRandom graph models
Random graph models
networksuw
 
Higher-order Link Prediction GraphEx
Higher-order Link Prediction GraphExHigher-order Link Prediction GraphEx
Higher-order Link Prediction GraphEx
Austin Benson
 
“Solving QCD: from BG/P to BG/Q”. Prof. Dr. Attilio Cucchieri – IFSC/USP.
“Solving QCD: from BG/P to BG/Q”. Prof. Dr. Attilio Cucchieri – IFSC/USP.“Solving QCD: from BG/P to BG/Q”. Prof. Dr. Attilio Cucchieri – IFSC/USP.
“Solving QCD: from BG/P to BG/Q”. Prof. Dr. Attilio Cucchieri – IFSC/USP.
lccausp
 
Simplicial closure & higher-order link prediction
Simplicial closure & higher-order link predictionSimplicial closure & higher-order link prediction
Simplicial closure & higher-order link prediction
Austin Benson
 
Quantum persistent k cores for community detection
Quantum persistent k cores for community detectionQuantum persistent k cores for community detection
Quantum persistent k cores for community detection
Colleen Farrelly
 
Higher-order spectral graph clustering with motifs
Higher-order spectral graph clustering with motifsHigher-order spectral graph clustering with motifs
Higher-order spectral graph clustering with motifs
Austin Benson
 
Sun_MAPL_GNN.pptx
Sun_MAPL_GNN.pptxSun_MAPL_GNN.pptx
Sun_MAPL_GNN.pptx
ssuser1760c0
 
Graph Partitioning and Spectral Methods
Graph Partitioning and Spectral MethodsGraph Partitioning and Spectral Methods
Graph Partitioning and Spectral Methods
Carlos Castillo (ChaTo)
 
kaleem arshad-1.pptx
kaleem arshad-1.pptxkaleem arshad-1.pptx
kaleem arshad-1.pptx
aliraza2732
 
Recreation mathematics ppt
Recreation mathematics pptRecreation mathematics ppt
Recreation mathematics ppt
Pawan Yadav
 
Disintegration of the small world property with increasing diversity of chemi...
Disintegration of the small world property with increasing diversity of chemi...Disintegration of the small world property with increasing diversity of chemi...
Disintegration of the small world property with increasing diversity of chemi...
N. Sukumar
 
08 Exponential Random Graph Models (ERGM)
08 Exponential Random Graph Models (ERGM)08 Exponential Random Graph Models (ERGM)
08 Exponential Random Graph Models (ERGM)
dnac
 
08 Exponential Random Graph Models (2016)
08 Exponential Random Graph Models (2016)08 Exponential Random Graph Models (2016)
08 Exponential Random Graph Models (2016)
Duke Network Analysis Center
 
Socialnetworkanalysis (Tin180 Com)
Socialnetworkanalysis (Tin180 Com)Socialnetworkanalysis (Tin180 Com)
Socialnetworkanalysis (Tin180 Com)
Tin180 VietNam
 
On the Scalability of Graph Kernels Applied to Collaborative Recommenders
On the Scalability of Graph Kernels Applied to Collaborative RecommendersOn the Scalability of Graph Kernels Applied to Collaborative Recommenders
On the Scalability of Graph Kernels Applied to Collaborative Recommenders
Jérôme KUNEGIS
 

Similar to Spectral Analysis of Signed Graphs for Clustering, Prediction and Visualization (20)

Graph Evolution Models
Graph Evolution ModelsGraph Evolution Models
Graph Evolution Models
 
240401_JW_labseminar[LINE: Large-scale Information Network Embeddin].pptx
240401_JW_labseminar[LINE: Large-scale Information Network Embeddin].pptx240401_JW_labseminar[LINE: Large-scale Information Network Embeddin].pptx
240401_JW_labseminar[LINE: Large-scale Information Network Embeddin].pptx
 
Spectral clustering with motifs and higher-order structures
Spectral clustering with motifs and higher-order structuresSpectral clustering with motifs and higher-order structures
Spectral clustering with motifs and higher-order structures
 
Simplicial closure and higher-order link prediction
Simplicial closure and higher-order link predictionSimplicial closure and higher-order link prediction
Simplicial closure and higher-order link prediction
 
Random graph models
Random graph modelsRandom graph models
Random graph models
 
Higher-order Link Prediction GraphEx
Higher-order Link Prediction GraphExHigher-order Link Prediction GraphEx
Higher-order Link Prediction GraphEx
 
“Solving QCD: from BG/P to BG/Q”. Prof. Dr. Attilio Cucchieri – IFSC/USP.
“Solving QCD: from BG/P to BG/Q”. Prof. Dr. Attilio Cucchieri – IFSC/USP.“Solving QCD: from BG/P to BG/Q”. Prof. Dr. Attilio Cucchieri – IFSC/USP.
“Solving QCD: from BG/P to BG/Q”. Prof. Dr. Attilio Cucchieri – IFSC/USP.
 
Simplicial closure & higher-order link prediction
Simplicial closure & higher-order link predictionSimplicial closure & higher-order link prediction
Simplicial closure & higher-order link prediction
 
Quantum persistent k cores for community detection
Quantum persistent k cores for community detectionQuantum persistent k cores for community detection
Quantum persistent k cores for community detection
 
Graph cluster randomization
Graph cluster randomizationGraph cluster randomization
Graph cluster randomization
 
Higher-order spectral graph clustering with motifs
Higher-order spectral graph clustering with motifsHigher-order spectral graph clustering with motifs
Higher-order spectral graph clustering with motifs
 
Sun_MAPL_GNN.pptx
Sun_MAPL_GNN.pptxSun_MAPL_GNN.pptx
Sun_MAPL_GNN.pptx
 
Graph Partitioning and Spectral Methods
Graph Partitioning and Spectral MethodsGraph Partitioning and Spectral Methods
Graph Partitioning and Spectral Methods
 
kaleem arshad-1.pptx
kaleem arshad-1.pptxkaleem arshad-1.pptx
kaleem arshad-1.pptx
 
Recreation mathematics ppt
Recreation mathematics pptRecreation mathematics ppt
Recreation mathematics ppt
 
Disintegration of the small world property with increasing diversity of chemi...
Disintegration of the small world property with increasing diversity of chemi...Disintegration of the small world property with increasing diversity of chemi...
Disintegration of the small world property with increasing diversity of chemi...
 
08 Exponential Random Graph Models (ERGM)
08 Exponential Random Graph Models (ERGM)08 Exponential Random Graph Models (ERGM)
08 Exponential Random Graph Models (ERGM)
 
08 Exponential Random Graph Models (2016)
08 Exponential Random Graph Models (2016)08 Exponential Random Graph Models (2016)
08 Exponential Random Graph Models (2016)
 
Socialnetworkanalysis (Tin180 Com)
Socialnetworkanalysis (Tin180 Com)Socialnetworkanalysis (Tin180 Com)
Socialnetworkanalysis (Tin180 Com)
 
On the Scalability of Graph Kernels Applied to Collaborative Recommenders
On the Scalability of Graph Kernels Applied to Collaborative RecommendersOn the Scalability of Graph Kernels Applied to Collaborative Recommenders
On the Scalability of Graph Kernels Applied to Collaborative Recommenders
 

More from Jérôme KUNEGIS

Succinct Summarisation of Large Networks via Small Synthetic Representative G...
Succinct Summarisation of Large Networks via Small Synthetic Representative G...Succinct Summarisation of Large Networks via Small Synthetic Representative G...
Succinct Summarisation of Large Networks via Small Synthetic Representative G...
Jérôme KUNEGIS
 
Title: What Is the Difference between a Social and a Hyperlink Network? -- Ho...
Title: What Is the Difference between a Social and a Hyperlink Network? -- Ho...Title: What Is the Difference between a Social and a Hyperlink Network? -- Ho...
Title: What Is the Difference between a Social and a Hyperlink Network? -- Ho...
Jérôme KUNEGIS
 
Measuring the Conflict in a Social Network with Friends and Foes: A Recent Al...
Measuring the Conflict in a Social Network with Friends and Foes: A Recent Al...Measuring the Conflict in a Social Network with Friends and Foes: A Recent Al...
Measuring the Conflict in a Social Network with Friends and Foes: A Recent Al...
Jérôme KUNEGIS
 
Schach und Computer
Schach und ComputerSchach und Computer
Schach und Computer
Jérôme KUNEGIS
 
Winning Science Slam by Jérôme Kunegis – First Prize at ICWSM 2016
Winning Science Slam by Jérôme Kunegis – First Prize at ICWSM 2016Winning Science Slam by Jérôme Kunegis – First Prize at ICWSM 2016
Winning Science Slam by Jérôme Kunegis – First Prize at ICWSM 2016
Jérôme KUNEGIS
 
Algebraic Graph-theoretic Measures of Conflict
Algebraic Graph-theoretic Measures of ConflictAlgebraic Graph-theoretic Measures of Conflict
Algebraic Graph-theoretic Measures of Conflict
Jérôme KUNEGIS
 
Generating Networks with Arbitrary Properties
Generating Networks with Arbitrary PropertiesGenerating Networks with Arbitrary Properties
Generating Networks with Arbitrary PropertiesJérôme KUNEGIS
 
Karriere Lounge – INFORMATIK 2013
Karriere Lounge – INFORMATIK 2013Karriere Lounge – INFORMATIK 2013
Karriere Lounge – INFORMATIK 2013
Jérôme KUNEGIS
 
Eight Formalisms for Defining Graph Models
Eight Formalisms for Defining Graph ModelsEight Formalisms for Defining Graph Models
Eight Formalisms for Defining Graph Models
Jérôme KUNEGIS
 
What Is the Added Value of Negative Links in Online Social Networks?
What Is the Added Value of Negative Links in Online Social Networks?What Is the Added Value of Negative Links in Online Social Networks?
What Is the Added Value of Negative Links in Online Social Networks?
Jérôme KUNEGIS
 
KONECT – The Koblenz Network Collection
KONECT – The Koblenz Network CollectionKONECT – The Koblenz Network Collection
KONECT – The Koblenz Network Collection
Jérôme KUNEGIS
 
Preferential Attachment in Online Networks: Measurement and Explanations
Preferential Attachment in Online Networks:  Measurement and ExplanationsPreferential Attachment in Online Networks:  Measurement and Explanations
Preferential Attachment in Online Networks: Measurement and Explanations
Jérôme KUNEGIS
 
Predicting Directed Links using Nondiagonal Matrix Decompositions
Predicting Directed Links using Nondiagonal Matrix DecompositionsPredicting Directed Links using Nondiagonal Matrix Decompositions
Predicting Directed Links using Nondiagonal Matrix DecompositionsJérôme KUNEGIS
 
Online Dating Recommender Systems: The Split-complex Number Approach
Online Dating Recommender Systems: The Split-complex Number ApproachOnline Dating Recommender Systems: The Split-complex Number Approach
Online Dating Recommender Systems: The Split-complex Number Approach
Jérôme KUNEGIS
 
Why Beyoncé Is More Popular Than Me – Fairness, Diversity and Other Measures
Why Beyoncé Is More Popular Than Me – Fairness, Diversity and Other MeasuresWhy Beyoncé Is More Popular Than Me – Fairness, Diversity and Other Measures
Why Beyoncé Is More Popular Than Me – Fairness, Diversity and Other MeasuresJérôme KUNEGIS
 
Fairness on the Web: Alternatives to the Power Law (WebSci 2012)
Fairness on the Web:  Alternatives to the Power Law (WebSci 2012)Fairness on the Web:  Alternatives to the Power Law (WebSci 2012)
Fairness on the Web: Alternatives to the Power Law (WebSci 2012)Jérôme KUNEGIS
 
Fairness on the Web: Alternatives to the Power Law
Fairness on the Web:  Alternatives to the Power LawFairness on the Web:  Alternatives to the Power Law
Fairness on the Web: Alternatives to the Power Law
Jérôme KUNEGIS
 
KONECT Cloud – Large Scale Network Mining in the Cloud
KONECT Cloud – Large Scale Network Mining in the CloudKONECT Cloud – Large Scale Network Mining in the Cloud
KONECT Cloud – Large Scale Network Mining in the Cloud
Jérôme KUNEGIS
 
On the Spectral Evolution of Large Networks (PhD Thesis by Jérôme Kunegis)
On the Spectral Evolution of Large Networks (PhD Thesis by Jérôme Kunegis)On the Spectral Evolution of Large Networks (PhD Thesis by Jérôme Kunegis)
On the Spectral Evolution of Large Networks (PhD Thesis by Jérôme Kunegis)
Jérôme KUNEGIS
 
Searching Microblogs: Coping with Sparsity and Document Quality
Searching Microblogs: Coping with Sparsity and Document QualitySearching Microblogs: Coping with Sparsity and Document Quality
Searching Microblogs: Coping with Sparsity and Document QualityJérôme KUNEGIS
 

More from Jérôme KUNEGIS (20)

Succinct Summarisation of Large Networks via Small Synthetic Representative G...
Succinct Summarisation of Large Networks via Small Synthetic Representative G...Succinct Summarisation of Large Networks via Small Synthetic Representative G...
Succinct Summarisation of Large Networks via Small Synthetic Representative G...
 
Title: What Is the Difference between a Social and a Hyperlink Network? -- Ho...
Title: What Is the Difference between a Social and a Hyperlink Network? -- Ho...Title: What Is the Difference between a Social and a Hyperlink Network? -- Ho...
Title: What Is the Difference between a Social and a Hyperlink Network? -- Ho...
 
Measuring the Conflict in a Social Network with Friends and Foes: A Recent Al...
Measuring the Conflict in a Social Network with Friends and Foes: A Recent Al...Measuring the Conflict in a Social Network with Friends and Foes: A Recent Al...
Measuring the Conflict in a Social Network with Friends and Foes: A Recent Al...
 
Schach und Computer
Schach und ComputerSchach und Computer
Schach und Computer
 
Winning Science Slam by Jérôme Kunegis – First Prize at ICWSM 2016
Winning Science Slam by Jérôme Kunegis – First Prize at ICWSM 2016Winning Science Slam by Jérôme Kunegis – First Prize at ICWSM 2016
Winning Science Slam by Jérôme Kunegis – First Prize at ICWSM 2016
 
Algebraic Graph-theoretic Measures of Conflict
Algebraic Graph-theoretic Measures of ConflictAlgebraic Graph-theoretic Measures of Conflict
Algebraic Graph-theoretic Measures of Conflict
 
Generating Networks with Arbitrary Properties
Generating Networks with Arbitrary PropertiesGenerating Networks with Arbitrary Properties
Generating Networks with Arbitrary Properties
 
Karriere Lounge – INFORMATIK 2013
Karriere Lounge – INFORMATIK 2013Karriere Lounge – INFORMATIK 2013
Karriere Lounge – INFORMATIK 2013
 
Eight Formalisms for Defining Graph Models
Eight Formalisms for Defining Graph ModelsEight Formalisms for Defining Graph Models
Eight Formalisms for Defining Graph Models
 
What Is the Added Value of Negative Links in Online Social Networks?
What Is the Added Value of Negative Links in Online Social Networks?What Is the Added Value of Negative Links in Online Social Networks?
What Is the Added Value of Negative Links in Online Social Networks?
 
KONECT – The Koblenz Network Collection
KONECT – The Koblenz Network CollectionKONECT – The Koblenz Network Collection
KONECT – The Koblenz Network Collection
 
Preferential Attachment in Online Networks: Measurement and Explanations
Preferential Attachment in Online Networks:  Measurement and ExplanationsPreferential Attachment in Online Networks:  Measurement and Explanations
Preferential Attachment in Online Networks: Measurement and Explanations
 
Predicting Directed Links using Nondiagonal Matrix Decompositions
Predicting Directed Links using Nondiagonal Matrix DecompositionsPredicting Directed Links using Nondiagonal Matrix Decompositions
Predicting Directed Links using Nondiagonal Matrix Decompositions
 
Online Dating Recommender Systems: The Split-complex Number Approach
Online Dating Recommender Systems: The Split-complex Number ApproachOnline Dating Recommender Systems: The Split-complex Number Approach
Online Dating Recommender Systems: The Split-complex Number Approach
 
Why Beyoncé Is More Popular Than Me – Fairness, Diversity and Other Measures
Why Beyoncé Is More Popular Than Me – Fairness, Diversity and Other MeasuresWhy Beyoncé Is More Popular Than Me – Fairness, Diversity and Other Measures
Why Beyoncé Is More Popular Than Me – Fairness, Diversity and Other Measures
 
Fairness on the Web: Alternatives to the Power Law (WebSci 2012)
Fairness on the Web:  Alternatives to the Power Law (WebSci 2012)Fairness on the Web:  Alternatives to the Power Law (WebSci 2012)
Fairness on the Web: Alternatives to the Power Law (WebSci 2012)
 
Fairness on the Web: Alternatives to the Power Law
Fairness on the Web:  Alternatives to the Power LawFairness on the Web:  Alternatives to the Power Law
Fairness on the Web: Alternatives to the Power Law
 
KONECT Cloud – Large Scale Network Mining in the Cloud
KONECT Cloud – Large Scale Network Mining in the CloudKONECT Cloud – Large Scale Network Mining in the Cloud
KONECT Cloud – Large Scale Network Mining in the Cloud
 
On the Spectral Evolution of Large Networks (PhD Thesis by Jérôme Kunegis)
On the Spectral Evolution of Large Networks (PhD Thesis by Jérôme Kunegis)On the Spectral Evolution of Large Networks (PhD Thesis by Jérôme Kunegis)
On the Spectral Evolution of Large Networks (PhD Thesis by Jérôme Kunegis)
 
Searching Microblogs: Coping with Sparsity and Document Quality
Searching Microblogs: Coping with Sparsity and Document QualitySearching Microblogs: Coping with Sparsity and Document Quality
Searching Microblogs: Coping with Sparsity and Document Quality
 

Recently uploaded

Pride Month Slides 2024 David Douglas School District
Pride Month Slides 2024 David Douglas School DistrictPride Month Slides 2024 David Douglas School District
Pride Month Slides 2024 David Douglas School District
David Douglas School District
 
The Accursed House by Émile Gaboriau.pptx
The Accursed House by Émile Gaboriau.pptxThe Accursed House by Émile Gaboriau.pptx
The Accursed House by Émile Gaboriau.pptx
DhatriParmar
 
Thesis Statement for students diagnonsed withADHD.ppt
Thesis Statement for students diagnonsed withADHD.pptThesis Statement for students diagnonsed withADHD.ppt
Thesis Statement for students diagnonsed withADHD.ppt
EverAndrsGuerraGuerr
 
TESDA TM1 REVIEWER FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...
TESDA TM1 REVIEWER  FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...TESDA TM1 REVIEWER  FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...
TESDA TM1 REVIEWER FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...
EugeneSaldivar
 
Synthetic Fiber Construction in lab .pptx
Synthetic Fiber Construction in lab .pptxSynthetic Fiber Construction in lab .pptx
Synthetic Fiber Construction in lab .pptx
Pavel ( NSTU)
 
Supporting (UKRI) OA monographs at Salford.pptx
Supporting (UKRI) OA monographs at Salford.pptxSupporting (UKRI) OA monographs at Salford.pptx
Supporting (UKRI) OA monographs at Salford.pptx
Jisc
 
Azure Interview Questions and Answers PDF By ScholarHat
Azure Interview Questions and Answers PDF By ScholarHatAzure Interview Questions and Answers PDF By ScholarHat
Azure Interview Questions and Answers PDF By ScholarHat
Scholarhat
 
Multithreading_in_C++ - std::thread, race condition
Multithreading_in_C++ - std::thread, race conditionMultithreading_in_C++ - std::thread, race condition
Multithreading_in_C++ - std::thread, race condition
Mohammed Sikander
 
Operation Blue Star - Saka Neela Tara
Operation Blue Star   -  Saka Neela TaraOperation Blue Star   -  Saka Neela Tara
Operation Blue Star - Saka Neela Tara
Balvir Singh
 
Lapbook sobre os Regimes Totalitários.pdf
Lapbook sobre os Regimes Totalitários.pdfLapbook sobre os Regimes Totalitários.pdf
Lapbook sobre os Regimes Totalitários.pdf
Jean Carlos Nunes Paixão
 
South African Journal of Science: Writing with integrity workshop (2024)
South African Journal of Science: Writing with integrity workshop (2024)South African Journal of Science: Writing with integrity workshop (2024)
South African Journal of Science: Writing with integrity workshop (2024)
Academy of Science of South Africa
 
How libraries can support authors with open access requirements for UKRI fund...
How libraries can support authors with open access requirements for UKRI fund...How libraries can support authors with open access requirements for UKRI fund...
How libraries can support authors with open access requirements for UKRI fund...
Jisc
 
Acetabularia Information For Class 9 .docx
Acetabularia Information For Class 9  .docxAcetabularia Information For Class 9  .docx
Acetabularia Information For Class 9 .docx
vaibhavrinwa19
 
Chapter 3 - Islamic Banking Products and Services.pptx
Chapter 3 - Islamic Banking Products and Services.pptxChapter 3 - Islamic Banking Products and Services.pptx
Chapter 3 - Islamic Banking Products and Services.pptx
Mohd Adib Abd Muin, Senior Lecturer at Universiti Utara Malaysia
 
Introduction to AI for Nonprofits with Tapp Network
Introduction to AI for Nonprofits with Tapp NetworkIntroduction to AI for Nonprofits with Tapp Network
Introduction to AI for Nonprofits with Tapp Network
TechSoup
 
2024.06.01 Introducing a competency framework for languag learning materials ...
2024.06.01 Introducing a competency framework for languag learning materials ...2024.06.01 Introducing a competency framework for languag learning materials ...
2024.06.01 Introducing a competency framework for languag learning materials ...
Sandy Millin
 
June 3, 2024 Anti-Semitism Letter Sent to MIT President Kornbluth and MIT Cor...
June 3, 2024 Anti-Semitism Letter Sent to MIT President Kornbluth and MIT Cor...June 3, 2024 Anti-Semitism Letter Sent to MIT President Kornbluth and MIT Cor...
June 3, 2024 Anti-Semitism Letter Sent to MIT President Kornbluth and MIT Cor...
Levi Shapiro
 
special B.ed 2nd year old paper_20240531.pdf
special B.ed 2nd year old paper_20240531.pdfspecial B.ed 2nd year old paper_20240531.pdf
special B.ed 2nd year old paper_20240531.pdf
Special education needs
 
"Protectable subject matters, Protection in biotechnology, Protection of othe...
"Protectable subject matters, Protection in biotechnology, Protection of othe..."Protectable subject matters, Protection in biotechnology, Protection of othe...
"Protectable subject matters, Protection in biotechnology, Protection of othe...
SACHIN R KONDAGURI
 
Chapter 4 - Islamic Financial Institutions in Malaysia.pptx
Chapter 4 - Islamic Financial Institutions in Malaysia.pptxChapter 4 - Islamic Financial Institutions in Malaysia.pptx
Chapter 4 - Islamic Financial Institutions in Malaysia.pptx
Mohd Adib Abd Muin, Senior Lecturer at Universiti Utara Malaysia
 

Recently uploaded (20)

Pride Month Slides 2024 David Douglas School District
Pride Month Slides 2024 David Douglas School DistrictPride Month Slides 2024 David Douglas School District
Pride Month Slides 2024 David Douglas School District
 
The Accursed House by Émile Gaboriau.pptx
The Accursed House by Émile Gaboriau.pptxThe Accursed House by Émile Gaboriau.pptx
The Accursed House by Émile Gaboriau.pptx
 
Thesis Statement for students diagnonsed withADHD.ppt
Thesis Statement for students diagnonsed withADHD.pptThesis Statement for students diagnonsed withADHD.ppt
Thesis Statement for students diagnonsed withADHD.ppt
 
TESDA TM1 REVIEWER FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...
TESDA TM1 REVIEWER  FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...TESDA TM1 REVIEWER  FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...
TESDA TM1 REVIEWER FOR NATIONAL ASSESSMENT WRITTEN AND ORAL QUESTIONS WITH A...
 
Synthetic Fiber Construction in lab .pptx
Synthetic Fiber Construction in lab .pptxSynthetic Fiber Construction in lab .pptx
Synthetic Fiber Construction in lab .pptx
 
Supporting (UKRI) OA monographs at Salford.pptx
Supporting (UKRI) OA monographs at Salford.pptxSupporting (UKRI) OA monographs at Salford.pptx
Supporting (UKRI) OA monographs at Salford.pptx
 
Azure Interview Questions and Answers PDF By ScholarHat
Azure Interview Questions and Answers PDF By ScholarHatAzure Interview Questions and Answers PDF By ScholarHat
Azure Interview Questions and Answers PDF By ScholarHat
 
Multithreading_in_C++ - std::thread, race condition
Multithreading_in_C++ - std::thread, race conditionMultithreading_in_C++ - std::thread, race condition
Multithreading_in_C++ - std::thread, race condition
 
Operation Blue Star - Saka Neela Tara
Operation Blue Star   -  Saka Neela TaraOperation Blue Star   -  Saka Neela Tara
Operation Blue Star - Saka Neela Tara
 
Lapbook sobre os Regimes Totalitários.pdf
Lapbook sobre os Regimes Totalitários.pdfLapbook sobre os Regimes Totalitários.pdf
Lapbook sobre os Regimes Totalitários.pdf
 
South African Journal of Science: Writing with integrity workshop (2024)
South African Journal of Science: Writing with integrity workshop (2024)South African Journal of Science: Writing with integrity workshop (2024)
South African Journal of Science: Writing with integrity workshop (2024)
 
How libraries can support authors with open access requirements for UKRI fund...
How libraries can support authors with open access requirements for UKRI fund...How libraries can support authors with open access requirements for UKRI fund...
How libraries can support authors with open access requirements for UKRI fund...
 
Acetabularia Information For Class 9 .docx
Acetabularia Information For Class 9  .docxAcetabularia Information For Class 9  .docx
Acetabularia Information For Class 9 .docx
 
Chapter 3 - Islamic Banking Products and Services.pptx
Chapter 3 - Islamic Banking Products and Services.pptxChapter 3 - Islamic Banking Products and Services.pptx
Chapter 3 - Islamic Banking Products and Services.pptx
 
Introduction to AI for Nonprofits with Tapp Network
Introduction to AI for Nonprofits with Tapp NetworkIntroduction to AI for Nonprofits with Tapp Network
Introduction to AI for Nonprofits with Tapp Network
 
2024.06.01 Introducing a competency framework for languag learning materials ...
2024.06.01 Introducing a competency framework for languag learning materials ...2024.06.01 Introducing a competency framework for languag learning materials ...
2024.06.01 Introducing a competency framework for languag learning materials ...
 
June 3, 2024 Anti-Semitism Letter Sent to MIT President Kornbluth and MIT Cor...
June 3, 2024 Anti-Semitism Letter Sent to MIT President Kornbluth and MIT Cor...June 3, 2024 Anti-Semitism Letter Sent to MIT President Kornbluth and MIT Cor...
June 3, 2024 Anti-Semitism Letter Sent to MIT President Kornbluth and MIT Cor...
 
special B.ed 2nd year old paper_20240531.pdf
special B.ed 2nd year old paper_20240531.pdfspecial B.ed 2nd year old paper_20240531.pdf
special B.ed 2nd year old paper_20240531.pdf
 
"Protectable subject matters, Protection in biotechnology, Protection of othe...
"Protectable subject matters, Protection in biotechnology, Protection of othe..."Protectable subject matters, Protection in biotechnology, Protection of othe...
"Protectable subject matters, Protection in biotechnology, Protection of othe...
 
Chapter 4 - Islamic Financial Institutions in Malaysia.pptx
Chapter 4 - Islamic Financial Institutions in Malaysia.pptxChapter 4 - Islamic Financial Institutions in Malaysia.pptx
Chapter 4 - Islamic Financial Institutions in Malaysia.pptx
 

Spectral Analysis of Signed Graphs for Clustering, Prediction and Visualization

  • 1. Spectral Analysis of Signed Graphs for Clustering, Prediction and Visualization Jérôme Kunegis¹, Stephan Schmidt¹, Andreas Lommatzsch¹ & Jürgen Lerner² ¹DAI Lab, Technische Universität Berlin, ²Universität Konstanz, Germany 10th SIAM International Conference on Data Mining, April 29–May 1, Columbus, Ohio
  • 2. Kunegis et al. Spectral Analysis of Signed Graphs 2 Introduction: Negative Edges Some websites allow you to have foes: Example: Slashdot Zoo (Kunegis 2009)
  • 3. Kunegis et al. Spectral Analysis of Signed Graphs 3 Introduction: Signed Graphs • The resulting social network is signed • Edges are positive or negative • In this talk: we use the graph Laplacian to study signed graphs Example: Slashdot Zoo (Kunegis 2009) me Friend ofFoe of Fan ofFreak of
  • 4. Kunegis et al. Spectral Analysis of Signed Graphs 4 Outline Introduction: Signed Graphs 1. Negative Edges and the Laplacian 2. Balance, Conflict and the Graph Spectrum 3. Communities, Cuts and Clustering 4. Resistance, Conductivity and Link Prediction Discussion
  • 5. Kunegis et al. Spectral Analysis of Signed Graphs 5 1. Negative Edges and the Laplacian Graph drawing: Place each node at the center of its neighbors v0 = (1/3) (v1 + v2 + v3 ) Algebraically: D v = A v Solution 1: Upper eigenvectors of D−1 A using A = {0, 1}n×n Solution 2: Lower eigenvectors of D – A and Dii = Σj Aij We look at solution 2: L = D − A is the Laplacian matrix v0 v1 v2 v3
  • 6. Kunegis et al. Spectral Analysis of Signed Graphs 6 Drawing Signed Graphs • Replace ‘negative’ neighbors by their antipodal points v0 = (1/3) (−v1 + v2 + v3 ) Solution: lower eigenvectors of L = D − A Using A = {0, −1, +1}n×n And Dii = Σj | Aij | v0 v1 v2 v3 −v1
  • 7. Kunegis et al. Spectral Analysis of Signed Graphs 7 Example: Synthetic Graph Unsigned Graph Drawing → Signed Graph Drawing
  • 8. Kunegis et al. Spectral Analysis of Signed Graphs 8 2. Balance, Conflict and the Graph Spectrum • ‘Balanced’ graphs have a perfect 2-clustering • Invert all negative edges • Effect on the Laplacian decomposition: Inversion of all eigenvectors of one cluster • Therefore: The spectrum of a balanced graph is the same as for the underlying unsigned graph (λ₁ = 0)
  • 9. Kunegis et al. Spectral Analysis of Signed Graphs 9 The Laplacian Spectrum of Unbalanced Graphs • Networks with conflict contain odd cycles • The Laplacian is always positive semidefinite xT Lx = Σij |Aij |(xi − sgn(Aij ) xj )² ≥ 0 • In unbalanced networks: λ₁ > 0
  • 10. Kunegis et al. Spectral Analysis of Signed Graphs 10 Algebraic Conflict • λ₁ denotes conflict Network λ₁₁₁₁ MovieLens 100k 0.4285 MovieLens 1M 0.3761 Jester 0.06515 MovieLens 10M 0.006183 Slashdot Zoo 0.006183 Epinions 0.004438 Conflict For effect of size, see Appendix
  • 11. Kunegis et al. Spectral Analysis of Signed Graphs 11 3. Communities, Cuts and Clustering The tribal groups of the Eastern Central Highlands of New Guinea can be friends (‘rova’) or enemies (‘hina’) Graphic uses two lower eigenvectors of L
  • 12. Kunegis et al. Spectral Analysis of Signed Graphs 12 Finding Communities The Laplacian matrix finds communities: • Communities are connected by many positive edges • Community are separated by many negative edges
  • 13. Kunegis et al. Spectral Analysis of Signed Graphs 13 Signed Spectral Clustering • Compute the d lower eigenvectors of L • Use k-means to cluster nodes in this d-dimensional space • Minimize signed normalized cut between communities X and Y SNC(X, Y) = (|X|−1 + |Y|−1 ) · (2 pos(X, Y) + neg(X, X) + neg(Y, Y)) pos/neg: number of positive/negative edges between communities
  • 14. Kunegis et al. Spectral Analysis of Signed Graphs 14 Example: Wikipedia Reverts • Users revert users on controversial Wikipedia article ‘Criticism of Prem Rawat’ • All edges are negative • Distance to center normalized to unit • Four clusters are apparent
  • 15. Kunegis et al. Spectral Analysis of Signed Graphs 15 4. Resistance, Conductivity and Link Prediction • Consider a network of electrical resistances: • Between any two nodes, the network has an effective resistance • The resistance distance is a squared Euclidean metric
  • 16. Kunegis et al. Spectral Analysis of Signed Graphs 16 Link Prediction • The resistance distance can be used for link prediction: – Long paths count less – Parallel paths count more dist(i,j) = (L+ )ii + (L+ )jj − (L+ )ij − (L+ )ji • Problem: How to handle negative edges?
  • 17. Kunegis et al. Spectral Analysis of Signed Graphs 17 Voltage Inversion • Solution: inverting amplifier dist(i,j) = (L+ )ii + (L+ )jj − (L+ )ij − (L+ )ji • Using signed Laplacian L • Is squared Euclidean because L is positive semidefinite −w w −
  • 18. Kunegis et al. Spectral Analysis of Signed Graphs 18 • Task: Predict the sign of new links • Problem: Find a function F(A) = B Evaluation: Link Sign Prediction Known positive links (A) Links to be predicted (B) Known negative links (A)
  • 19. Kunegis et al. Spectral Analysis of Signed Graphs 19 Graph Kernels Link prediction functions using the Laplacian: • L+ – Signed Laplacian kernel • (I + αL)−1 – Signed regularized Laplacian kernel • exp(−αL) – Signed ‘heat diffusion’ Other link prediction functions: • (A)k – Rank reduction • exp(A) – Matrix exponential • Poly(A) – Path counting
  • 20. Kunegis et al. Spectral Analysis of Signed Graphs 20 Evaluation Results • MovieLens: predict good / bad rating • Best rating prediction: signed regularized Laplacian graph kernel Link Prediction RMSE Rank reduction 0.838 Path counting 0.840 Matrix exponential 0.839 Signed resistance distance 0.812 Signed regularized Laplacian 0.778 Signed heat diffusion 0.789
  • 21. Kunegis et al. Spectral Analysis of Signed Graphs 21 Summary The Laplacian matrix applies to signed graphs The Laplacian spectrum denotes graph conflict The signed Laplacian arises in several ways: For graph drawing, the Laplacian implements antipodal proximity For clustering, the Laplacian implements signed cuts As an interpretation of negation as inversion of electrical potential
  • 23. Kunegis et al. Spectral Analysis of Signed Graphs 23 References P. Hage, F. Harary. Structural models in anthropology, Cambridge University Press, 1983. F. Harary. On the notion of balance of a signed graph, Michigan Math. J., 2:143–146, 1953. J. Kunegis, A. Lommatzsch, C. Bauckhage, The Slashdot Zoo: Mining a social network with negative edges, Proc. Int. World Wide Web Conf., pages 741–750, 2009. J. Leskovec, Daniel Huttenlocher, Jon Kleinberg, Predicting positive and negative links in online social networks, Proc. Int. World Wide Web Conf., 2010.
  • 24. Kunegis et al. Spectral Analysis of Signed Graphs 24 Appendix – Balance vs Volume
  • 25. Kunegis et al. Spectral Analysis of Signed Graphs 25 Appendix – Scalability • Evaluation results in function of reduced rank k
  • 26. Kunegis et al. Spectral Analysis of Signed Graphs 26 Appendix -- Balance, Conflict and the Graph Spectrum Look at triads of users (Harary 1953): • In balanced triangles, the multiplication rule holds • If it doesn't, there is conflict Balance: Conflict:
  • 27. Kunegis et al. Spectral Analysis of Signed Graphs 27 The Signed Clustering Coefficient • How many triangles are balanced? Cs = (#balanced − #unbalanced) / #possible • This measure is local, not global (Kunegis 2009) ± uv ? u v
  • 28. Kunegis et al. Spectral Analysis of Signed Graphs 28 Introduction: Networks • Many web sites allow you to have friends: Example: Facebook