SlideShare uma empresa Scribd logo
1 de 40
Random Walk on Graphs
Pavan Kapanipathi
Reading Group (Kno.e.sis)
Referred: Purnamrita Sarkar, Random
Walks on Graphs: An Overview
Agenda
• Introduction
– Motivation

• Background
– Graphs
– Matrices

• Random Walk
– PageRank
– Personalized PageRank
– Topic Sensitive PageRank

• Applications
– Specifically in Recommender Systems
Random Walk

A drunk man will find his way home, but a drunk
bird may get lost forever.
Motivation: Link prediction in social
networks

4
Motivation: Basis for recommendation

5
Since I had very less slides and More
time – Graphs
• Undirected Graphs
Since I had very less slides and more
time in hand-- Graphs
• Directed Graphs
Since I had very less slides and more
time in hand – Matrix
Rows

Columns

i

j

i
j
k

i,j

k
Adjacency and Transition Matrix

Transition matrix P

Adjacency matrix A

1

1
1

1

1

1

1/2

1/2

9
Markov Property (Basic)
• Given the present state, the future and past
states are independent
Stochastic
• Wikipedia: In probability theory, a purely
stochastic system is one whose state is nondeterministic so that the subsequent state of
the system is determined probabilistically.
1
1

• Matrix
– Stochastic Row/Column?

1/2

1/2
Random Walk on Graphs
Random Walk on Graphs
Random Walk on Graphs
Random Walk on Graphs

The random sequence of points selected this
way is a random walk on the graph
Again: Transition Matrix
j
k
i
i
j
k

Transition matrix P
Probability?
1
1

1/2

1/2

16
Probability Distributions
• xt(i) = probability that the surfer is at node i at
time t
• xt+1(i) = ∑j(Probability of being at node j)*Pr(j->i)
=∑jxt(j)*P(j,i)

• xt+1 = xtP = xt-1*P*P= xt-2*P*P*P = …=x0 Pt
Matrix Multiplication?

• What happens when the surfer keeps walking for
a long time?
17
Property of Adjacency Matrix

Adjacency matrix A

1
1

1

1
What is a stationary distribution?
Intuitively and Mathematically
• The stationary distribution at a node is related to the amount
of time a random walker spends visiting that node.
• Remember that we can write the probability distribution at a
node as
– xt+1 = xtP

• For the stationary distribution v0 we have
– v0 = v0 P
• Whoa! that’s just the left eigenvector of the transition matrix !

19
Eigen Value and Eigen Vector?
Interesting questions
• Does a stationary distribution always exist? Is it unique?
– Yes, if the graph is “well-behaved”.

• What is “well-behaved”?
– We shall talk about this soon.

• How fast will the random surfer approach this stationary
distribution?
– Mixing Time!

21
Well behaved graphs
• Irreducible: There is a path from every node to every other
node.

What about connected undirected Graph?

Irreducible

Not irreducible

22
Well behaved graphs
• Aperiodic: The GCD of all cycle lengths is 1. The GCD is also
called period.

Periodicity is 3

Aperiodic

23
Implications of the Perron Frobenius Theorem
• If a markov chain is irreducible and aperiodic then the largest
eigenvalue of the transition matrix will be equal to 1 and all
the other eigenvalues will be strictly less than 1.
– Let the eigenvalues of P be {σi| i=0:n-1} in non-increasing order of
σi .
– σ0 = 1 > σ1 > σ2 >= ……>= σn

• These results imply that for a well behaved graph there exists
an unique stationary distribution.

• More details when we discuss pagerank.

24
Some fun stuff about undirected graphs
• A connected undirected graph is irreducible
• A connected non-bipartite undirected graph has a stationary
distribution proportional to the degree distribution!

• Makes sense, since larger the degree of the node more likely a
random walk is to come back to it.

25
Proximity measures from random walks
b

a

• How long does it take to hit node b in a random walk starting
at node a? --- Hitting time.
• How long does it take to hit node b and come back to node a?
--- Commute time.
26
Hitting and Commute times
b

a

• Hitting time from node i to node j
– Expected number of hops to hit node j starting at node i.
– Is not symmetric. h(a,b) > h(a,b)
– h(i,j) = 1 + ΣkЄnbs(A) p(i,k)h(k,j)

27
Hitting and Commute times
b

a

• Commute time between node i and j
– Is expected time to hit node j and come back to i

– c(i,j) = h(i,j) + h(j,i)
– Is symmetric. c(a,b) = c(b,a)

28
Random Walk (versions)
• PageRank
– Personalized PageRank
– Topic Sensitive PageRank

• Recommender Systems (My interests)
Recommender Networks
• For a customer node i define similarity as
– H(i,j)
– C(i,j)
– Or the cosine similarity


Lij

Lii Ljj

• Now the question is how to compute these quantities quickly
for very large graphs.
– Fast iterative techniques (Brand 2005)
– Fast Random Walk with Restart (Tong, Faloutsos 2006)
– Finding nearest neighbors in graphs (Sarkar, Moore 2007)

30
PageRank (Initial)
• Intuition
– PageRank of “A” is higher if the pages that links to
“A” has higher PageRank
• User behavior where a surfer clicks on links at random
with no regard towards content

– One page's PageRank is not completely passed on
to a page it links to, but is divided by the number
of links on the page.
PageRank
• Intuitively
v (i ) 

v (j )

i degout ( j )
j


• v works out to be the stationary distribution of
the markov chain corresponding to the web.
Pagerank & Perron-frobenius
• Perron Frobenius only holds if the graph is irreducible and
aperiodic.
• But how can we guarantee that for the web graph?
– Do it with a small restart probability c.

• At any time-step the random surfer
– jumps (teleport) to any other node with probability c
– jumps to its direct neighbors with total probability 1-c.
~

P  (1  c )P  cU
Uij 

1

n

i , j

~

P  cP  (1  c)U
U ij 

1
i, j
n
33
Pagerank
• We are looking for the vector v s.t.
v  (1  c ) vP  cr

• r is a distribution over web-pages.

• If r is the uniform distribution we get pagerank.
• What happens if r is non-uniform?
Personalization

34
Personalized Pagerank1,2,3
• The only difference is that we use a non-uniform teleportation
distribution, i.e. at any time step teleport to a set of
webpages.
• In other words we are looking for the vector v s.t.
v  (1  c ) vP  cr

• r is a non-uniform preference vector specific to an user.
• v gives “personalized views” of the web.
1. Scaling Personalized Web Search, Jeh, Widom. 2003
2. Topic-sensitive PageRank, Haveliwala, 2001

3. Towards scaling fully personalized pagerank, D. Fogaras and B. Racz, 2004

35
Topic-sensitive pagerank (Haveliwala’01)
• Divide the webpages into 16 broad categories
• For each category compute the biased personalized pagerank
vector by uniformly teleporting to websites under that
category.
• At query time the probability of the query being from any of
the above classes is computed, and the final page-rank vector
is computed by a linear combination of the biased pagerank
vectors computed offline.

36
Random Walk for Recommendations
• Collaborative Filtering by Shang et.al
• Graph
– Vertices: Users (U), Items (I), Item Information (T) and
User Profiles (P)
– Edges with weights
•
•
•
•

u has a rating for i
i has a tag for t
u belongs to profile category p
u to u if they are connected in the social network

• Edge weights assignments
Random Walk for Recommendations
Connectivity/Transition

Generally 0.85

Preference Vector
Most of it from
• Purnamrita Sarkar, Random Walks on Graphs:
An Overview
• Random Walks on Graphs: A Survey, Laszlo
Lov'asz
• OBVIOUSLY: Wikipedia :D
• Random Walk on Graphs: Ankit Agarwal
Thanks

Mais conteúdo relacionado

Mais procurados

Optimization/Gradient Descent
Optimization/Gradient DescentOptimization/Gradient Descent
Optimization/Gradient Descentkandelin
 
Beginners Guide to Non-Negative Matrix Factorization
Beginners Guide to Non-Negative Matrix FactorizationBeginners Guide to Non-Negative Matrix Factorization
Beginners Guide to Non-Negative Matrix FactorizationBenjamin Bengfort
 
Model selection and cross validation techniques
Model selection and cross validation techniquesModel selection and cross validation techniques
Model selection and cross validation techniquesVenkata Reddy Konasani
 
Bias and variance trade off
Bias and variance trade offBias and variance trade off
Bias and variance trade offVARUN KUMAR
 
Decision tree and instance-based learning for label ranking
Decision tree and instance-based learning for label rankingDecision tree and instance-based learning for label ranking
Decision tree and instance-based learning for label rankingroywwcheng
 
Multi-Layer Perceptrons
Multi-Layer PerceptronsMulti-Layer Perceptrons
Multi-Layer PerceptronsESCOM
 
Opinion Dynamics on Networks
Opinion Dynamics on NetworksOpinion Dynamics on Networks
Opinion Dynamics on NetworksMason Porter
 
Feedforward neural network
Feedforward neural networkFeedforward neural network
Feedforward neural networkSopheaktra YONG
 
Natural language processing (nlp)
Natural language processing (nlp)Natural language processing (nlp)
Natural language processing (nlp)Kuppusamy P
 
Linear regression with gradient descent
Linear regression with gradient descentLinear regression with gradient descent
Linear regression with gradient descentSuraj Parmar
 
Linear Regression vs Logistic Regression | Edureka
Linear Regression vs Logistic Regression | EdurekaLinear Regression vs Logistic Regression | Edureka
Linear Regression vs Logistic Regression | EdurekaEdureka!
 
Evolutionary Algorithms
Evolutionary AlgorithmsEvolutionary Algorithms
Evolutionary AlgorithmsReem Alattas
 
Multi-Armed Bandit and Applications
Multi-Armed Bandit and ApplicationsMulti-Armed Bandit and Applications
Multi-Armed Bandit and ApplicationsSangwoo Mo
 
Rules of inference
Rules of inferenceRules of inference
Rules of inferenceharman kaur
 
Graph Neural Networks for Recommendations
Graph Neural Networks for RecommendationsGraph Neural Networks for Recommendations
Graph Neural Networks for RecommendationsWQ Fan
 

Mais procurados (20)

Optimization/Gradient Descent
Optimization/Gradient DescentOptimization/Gradient Descent
Optimization/Gradient Descent
 
Beginners Guide to Non-Negative Matrix Factorization
Beginners Guide to Non-Negative Matrix FactorizationBeginners Guide to Non-Negative Matrix Factorization
Beginners Guide to Non-Negative Matrix Factorization
 
Model selection and cross validation techniques
Model selection and cross validation techniquesModel selection and cross validation techniques
Model selection and cross validation techniques
 
Bias and variance trade off
Bias and variance trade offBias and variance trade off
Bias and variance trade off
 
Markov chain
Markov chainMarkov chain
Markov chain
 
Decision tree and instance-based learning for label ranking
Decision tree and instance-based learning for label rankingDecision tree and instance-based learning for label ranking
Decision tree and instance-based learning for label ranking
 
Clustering
ClusteringClustering
Clustering
 
Multi-Layer Perceptrons
Multi-Layer PerceptronsMulti-Layer Perceptrons
Multi-Layer Perceptrons
 
Opinion Dynamics on Networks
Opinion Dynamics on NetworksOpinion Dynamics on Networks
Opinion Dynamics on Networks
 
lattice
 lattice lattice
lattice
 
Feedforward neural network
Feedforward neural networkFeedforward neural network
Feedforward neural network
 
K - Nearest neighbor ( KNN )
K - Nearest neighbor  ( KNN )K - Nearest neighbor  ( KNN )
K - Nearest neighbor ( KNN )
 
Natural language processing (nlp)
Natural language processing (nlp)Natural language processing (nlp)
Natural language processing (nlp)
 
Linear regression with gradient descent
Linear regression with gradient descentLinear regression with gradient descent
Linear regression with gradient descent
 
Linear Regression vs Logistic Regression | Edureka
Linear Regression vs Logistic Regression | EdurekaLinear Regression vs Logistic Regression | Edureka
Linear Regression vs Logistic Regression | Edureka
 
Evolutionary Algorithms
Evolutionary AlgorithmsEvolutionary Algorithms
Evolutionary Algorithms
 
Multi-Armed Bandit and Applications
Multi-Armed Bandit and ApplicationsMulti-Armed Bandit and Applications
Multi-Armed Bandit and Applications
 
Rules of inference
Rules of inferenceRules of inference
Rules of inference
 
Graph Neural Networks for Recommendations
Graph Neural Networks for RecommendationsGraph Neural Networks for Recommendations
Graph Neural Networks for Recommendations
 
K Nearest Neighbor Algorithm
K Nearest Neighbor AlgorithmK Nearest Neighbor Algorithm
K Nearest Neighbor Algorithm
 

Destaque

Random Walks, Efficient Markets & Stock Prices
Random Walks, Efficient Markets & Stock PricesRandom Walks, Efficient Markets & Stock Prices
Random Walks, Efficient Markets & Stock PricesNEO Empresarial
 
Random Walk Theory- Investment
Random Walk Theory- InvestmentRandom Walk Theory- Investment
Random Walk Theory- InvestmentSaranya karthick
 
Efficient market hypothesis
Efficient market hypothesisEfficient market hypothesis
Efficient market hypothesisKamlesh Pawar
 
A Random Walk Down Wall Street
A Random Walk Down Wall StreetA Random Walk Down Wall Street
A Random Walk Down Wall Streetarcaneadam
 
Personalized and Adaptive Semantic Information Filtering for Social Media
Personalized and Adaptive Semantic Information Filtering for Social MediaPersonalized and Adaptive Semantic Information Filtering for Social Media
Personalized and Adaptive Semantic Information Filtering for Social MediaPavan Kapanipathi
 
An evaluation of SimRank and Personalized PageRank to build a recommender sys...
An evaluation of SimRank and Personalized PageRank to build a recommender sys...An evaluation of SimRank and Personalized PageRank to build a recommender sys...
An evaluation of SimRank and Personalized PageRank to build a recommender sys...Paolo Tomeo
 
Market efficiency presentation
Market efficiency presentationMarket efficiency presentation
Market efficiency presentationMohammed Alashi
 
PageRank for anomaly detection - Hadoop Summit
PageRank for anomaly detection - Hadoop SummitPageRank for anomaly detection - Hadoop Summit
PageRank for anomaly detection - Hadoop SummitOfer Mendelevitch
 
Large-Scale Graph Computation on Just a PC: Aapo Kyrola Ph.D. thesis defense
Large-Scale Graph Computation on Just a PC: Aapo Kyrola Ph.D. thesis defenseLarge-Scale Graph Computation on Just a PC: Aapo Kyrola Ph.D. thesis defense
Large-Scale Graph Computation on Just a PC: Aapo Kyrola Ph.D. thesis defenseAapo Kyrölä
 
corporate finance and market efficiency
corporate finance and market efficiencycorporate finance and market efficiency
corporate finance and market efficiencygeet232
 
Focused Clustering and Outlier Detection in Large Attributed Graphs
Focused Clustering and Outlier Detection in Large Attributed GraphsFocused Clustering and Outlier Detection in Large Attributed Graphs
Focused Clustering and Outlier Detection in Large Attributed GraphsBryan Perozzi
 
A Random Walk Through Search Research
A Random Walk Through Search ResearchA Random Walk Through Search Research
A Random Walk Through Search ResearchNick Watkins
 
Market efficiency presentation
Market efficiency presentationMarket efficiency presentation
Market efficiency presentationMohammad Waqas
 
Data Science view of the KDD 2014
Data Science view of the KDD 2014Data Science view of the KDD 2014
Data Science view of the KDD 2014Jure Leskovec
 
Personalized PageRank based community detection
Personalized PageRank based community detectionPersonalized PageRank based community detection
Personalized PageRank based community detectionDavid Gleich
 
Random Walk Theory: 'Sistemi di grid automatizzati per Metatrader 4'
Random Walk Theory: 'Sistemi di grid automatizzati per Metatrader 4'Random Walk Theory: 'Sistemi di grid automatizzati per Metatrader 4'
Random Walk Theory: 'Sistemi di grid automatizzati per Metatrader 4'Stefano Gianti
 

Destaque (20)

Random walk theory
Random walk theoryRandom walk theory
Random walk theory
 
Random Walks, Efficient Markets & Stock Prices
Random Walks, Efficient Markets & Stock PricesRandom Walks, Efficient Markets & Stock Prices
Random Walks, Efficient Markets & Stock Prices
 
Random Walk Theory- Investment
Random Walk Theory- InvestmentRandom Walk Theory- Investment
Random Walk Theory- Investment
 
Efficient market hypothesis
Efficient market hypothesisEfficient market hypothesis
Efficient market hypothesis
 
A Random Walk Down Wall Street
A Random Walk Down Wall StreetA Random Walk Down Wall Street
A Random Walk Down Wall Street
 
Random Walk Theory
Random Walk Theory Random Walk Theory
Random Walk Theory
 
Personalized and Adaptive Semantic Information Filtering for Social Media
Personalized and Adaptive Semantic Information Filtering for Social MediaPersonalized and Adaptive Semantic Information Filtering for Social Media
Personalized and Adaptive Semantic Information Filtering for Social Media
 
An evaluation of SimRank and Personalized PageRank to build a recommender sys...
An evaluation of SimRank and Personalized PageRank to build a recommender sys...An evaluation of SimRank and Personalized PageRank to build a recommender sys...
An evaluation of SimRank and Personalized PageRank to build a recommender sys...
 
Ch10 final
Ch10 finalCh10 final
Ch10 final
 
Market efficiency presentation
Market efficiency presentationMarket efficiency presentation
Market efficiency presentation
 
PageRank for anomaly detection - Hadoop Summit
PageRank for anomaly detection - Hadoop SummitPageRank for anomaly detection - Hadoop Summit
PageRank for anomaly detection - Hadoop Summit
 
Large-Scale Graph Computation on Just a PC: Aapo Kyrola Ph.D. thesis defense
Large-Scale Graph Computation on Just a PC: Aapo Kyrola Ph.D. thesis defenseLarge-Scale Graph Computation on Just a PC: Aapo Kyrola Ph.D. thesis defense
Large-Scale Graph Computation on Just a PC: Aapo Kyrola Ph.D. thesis defense
 
QF-CUTN-FINAL
QF-CUTN-FINALQF-CUTN-FINAL
QF-CUTN-FINAL
 
corporate finance and market efficiency
corporate finance and market efficiencycorporate finance and market efficiency
corporate finance and market efficiency
 
Focused Clustering and Outlier Detection in Large Attributed Graphs
Focused Clustering and Outlier Detection in Large Attributed GraphsFocused Clustering and Outlier Detection in Large Attributed Graphs
Focused Clustering and Outlier Detection in Large Attributed Graphs
 
A Random Walk Through Search Research
A Random Walk Through Search ResearchA Random Walk Through Search Research
A Random Walk Through Search Research
 
Market efficiency presentation
Market efficiency presentationMarket efficiency presentation
Market efficiency presentation
 
Data Science view of the KDD 2014
Data Science view of the KDD 2014Data Science view of the KDD 2014
Data Science view of the KDD 2014
 
Personalized PageRank based community detection
Personalized PageRank based community detectionPersonalized PageRank based community detection
Personalized PageRank based community detection
 
Random Walk Theory: 'Sistemi di grid automatizzati per Metatrader 4'
Random Walk Theory: 'Sistemi di grid automatizzati per Metatrader 4'Random Walk Theory: 'Sistemi di grid automatizzati per Metatrader 4'
Random Walk Theory: 'Sistemi di grid automatizzati per Metatrader 4'
 

Semelhante a Random walk on Graphs

randomwalks_states_figures_events_happenings.ppt
randomwalks_states_figures_events_happenings.pptrandomwalks_states_figures_events_happenings.ppt
randomwalks_states_figures_events_happenings.pptVGaneshKarthikeyan
 
Introduction to Random Walk
Introduction to Random WalkIntroduction to Random Walk
Introduction to Random WalkShuai Zhang
 
Financial Networks III. Centrality and Systemic Importance
Financial Networks III. Centrality and Systemic ImportanceFinancial Networks III. Centrality and Systemic Importance
Financial Networks III. Centrality and Systemic ImportanceKimmo Soramaki
 
Random walks on graphs - link prediction by Rouhollah Nabati
Random walks on graphs - link prediction by Rouhollah NabatiRandom walks on graphs - link prediction by Rouhollah Nabati
Random walks on graphs - link prediction by Rouhollah Nabatinabati
 
Kinematics i
Kinematics iKinematics i
Kinematics ikleybf
 
2.1 Displacement & Velocity
2.1 Displacement & Velocity2.1 Displacement & Velocity
2.1 Displacement & Velocitymlong24
 
Pavan rmc14quantumwalks
Pavan rmc14quantumwalksPavan rmc14quantumwalks
Pavan rmc14quantumwalksPavan Iyengar
 
Unit VI - Graphs.ppt
Unit VI - Graphs.pptUnit VI - Graphs.ppt
Unit VI - Graphs.pptHODElex
 
SPATIAL POINT PATTERNS
SPATIAL POINT PATTERNSSPATIAL POINT PATTERNS
SPATIAL POINT PATTERNSLiemNguyenDuy
 
From Competition to Complementarity: Comparative Influence Diffusion and Maxi...
From Competition to Complementarity: Comparative Influence Diffusion and Maxi...From Competition to Complementarity: Comparative Influence Diffusion and Maxi...
From Competition to Complementarity: Comparative Influence Diffusion and Maxi...Wei Lu
 
AI3391 Artificial Intelligence Session 23 Backtracking CSP's.pptx
AI3391 Artificial Intelligence Session 23 Backtracking CSP's.pptxAI3391 Artificial Intelligence Session 23 Backtracking CSP's.pptx
AI3391 Artificial Intelligence Session 23 Backtracking CSP's.pptxAsst.prof M.Gokilavani
 

Semelhante a Random walk on Graphs (20)

Data Mining Lecture_12.pptx
Data Mining Lecture_12.pptxData Mining Lecture_12.pptx
Data Mining Lecture_12.pptx
 
randomwalks_states_figures_events_happenings.ppt
randomwalks_states_figures_events_happenings.pptrandomwalks_states_figures_events_happenings.ppt
randomwalks_states_figures_events_happenings.ppt
 
randomwalk.ppt
randomwalk.pptrandomwalk.ppt
randomwalk.ppt
 
Introduction to Random Walk
Introduction to Random WalkIntroduction to Random Walk
Introduction to Random Walk
 
Financial Networks III. Centrality and Systemic Importance
Financial Networks III. Centrality and Systemic ImportanceFinancial Networks III. Centrality and Systemic Importance
Financial Networks III. Centrality and Systemic Importance
 
graphs.ppt
graphs.pptgraphs.ppt
graphs.ppt
 
Random walks on graphs - link prediction by Rouhollah Nabati
Random walks on graphs - link prediction by Rouhollah NabatiRandom walks on graphs - link prediction by Rouhollah Nabati
Random walks on graphs - link prediction by Rouhollah Nabati
 
Kinematics i
Kinematics iKinematics i
Kinematics i
 
Locality sensitive hashing
Locality sensitive hashingLocality sensitive hashing
Locality sensitive hashing
 
Clustering - ACM 2013 02-25
Clustering - ACM 2013 02-25Clustering - ACM 2013 02-25
Clustering - ACM 2013 02-25
 
2.1 Displacement & Velocity
2.1 Displacement & Velocity2.1 Displacement & Velocity
2.1 Displacement & Velocity
 
Pavan rmc14quantumwalks
Pavan rmc14quantumwalksPavan rmc14quantumwalks
Pavan rmc14quantumwalks
 
Unit VI - Graphs.ppt
Unit VI - Graphs.pptUnit VI - Graphs.ppt
Unit VI - Graphs.ppt
 
Social (1)
Social (1)Social (1)
Social (1)
 
SPATIAL POINT PATTERNS
SPATIAL POINT PATTERNSSPATIAL POINT PATTERNS
SPATIAL POINT PATTERNS
 
From Competition to Complementarity: Comparative Influence Diffusion and Maxi...
From Competition to Complementarity: Comparative Influence Diffusion and Maxi...From Competition to Complementarity: Comparative Influence Diffusion and Maxi...
From Competition to Complementarity: Comparative Influence Diffusion and Maxi...
 
Data Mining Lecture_13.pptx
Data Mining Lecture_13.pptxData Mining Lecture_13.pptx
Data Mining Lecture_13.pptx
 
AI3391 Artificial Intelligence Session 23 Backtracking CSP's.pptx
AI3391 Artificial Intelligence Session 23 Backtracking CSP's.pptxAI3391 Artificial Intelligence Session 23 Backtracking CSP's.pptx
AI3391 Artificial Intelligence Session 23 Backtracking CSP's.pptx
 
16355694.ppt
16355694.ppt16355694.ppt
16355694.ppt
 
Sudoku
SudokuSudoku
Sudoku
 

Mais de Pavan Kapanipathi

Improving Natural Language Inference Using External Knowledge in the Science ...
Improving Natural Language Inference Using External Knowledge in the Science ...Improving Natural Language Inference Using External Knowledge in the Science ...
Improving Natural Language Inference Using External Knowledge in the Science ...Pavan Kapanipathi
 
Knoesis-Semantic filtering-Tutorials
Knoesis-Semantic filtering-TutorialsKnoesis-Semantic filtering-Tutorials
Knoesis-Semantic filtering-TutorialsPavan Kapanipathi
 
Knowledge base enabled Information Filtering on Social Web -- EMC
Knowledge base enabled Information Filtering on Social Web -- EMCKnowledge base enabled Information Filtering on Social Web -- EMC
Knowledge base enabled Information Filtering on Social Web -- EMCPavan Kapanipathi
 
Adressing Volume and Velocity Challenge on the Social Web using Crowd Sourced...
Adressing Volume and Velocity Challenge on the Social Web using Crowd Sourced...Adressing Volume and Velocity Challenge on the Social Web using Crowd Sourced...
Adressing Volume and Velocity Challenge on the Social Web using Crowd Sourced...Pavan Kapanipathi
 
Hierarchical Interest Graphs from Twitter
Hierarchical Interest Graphs from TwitterHierarchical Interest Graphs from Twitter
Hierarchical Interest Graphs from TwitterPavan Kapanipathi
 
User Interests Identification From Twitter using Hierarchical Knowledge Base
User Interests Identification From Twitter using Hierarchical Knowledge BaseUser Interests Identification From Twitter using Hierarchical Knowledge Base
User Interests Identification From Twitter using Hierarchical Knowledge BasePavan Kapanipathi
 
Privacy Aware Semantic Dissemination
Privacy Aware Semantic DisseminationPrivacy Aware Semantic Dissemination
Privacy Aware Semantic DisseminationPavan Kapanipathi
 
Personalized Filtering of Twitter Stream
Personalized Filtering of Twitter StreamPersonalized Filtering of Twitter Stream
Personalized Filtering of Twitter StreamPavan Kapanipathi
 

Mais de Pavan Kapanipathi (9)

Improving Natural Language Inference Using External Knowledge in the Science ...
Improving Natural Language Inference Using External Knowledge in the Science ...Improving Natural Language Inference Using External Knowledge in the Science ...
Improving Natural Language Inference Using External Knowledge in the Science ...
 
Knoesis-Semantic filtering-Tutorials
Knoesis-Semantic filtering-TutorialsKnoesis-Semantic filtering-Tutorials
Knoesis-Semantic filtering-Tutorials
 
Knowledge base enabled Information Filtering on Social Web -- EMC
Knowledge base enabled Information Filtering on Social Web -- EMCKnowledge base enabled Information Filtering on Social Web -- EMC
Knowledge base enabled Information Filtering on Social Web -- EMC
 
Adressing Volume and Velocity Challenge on the Social Web using Crowd Sourced...
Adressing Volume and Velocity Challenge on the Social Web using Crowd Sourced...Adressing Volume and Velocity Challenge on the Social Web using Crowd Sourced...
Adressing Volume and Velocity Challenge on the Social Web using Crowd Sourced...
 
Hierarchical Interest Graphs from Twitter
Hierarchical Interest Graphs from TwitterHierarchical Interest Graphs from Twitter
Hierarchical Interest Graphs from Twitter
 
User Interests Identification From Twitter using Hierarchical Knowledge Base
User Interests Identification From Twitter using Hierarchical Knowledge BaseUser Interests Identification From Twitter using Hierarchical Knowledge Base
User Interests Identification From Twitter using Hierarchical Knowledge Base
 
SemPuSH: ISWC 2011 Poster
SemPuSH: ISWC 2011 PosterSemPuSH: ISWC 2011 Poster
SemPuSH: ISWC 2011 Poster
 
Privacy Aware Semantic Dissemination
Privacy Aware Semantic DisseminationPrivacy Aware Semantic Dissemination
Privacy Aware Semantic Dissemination
 
Personalized Filtering of Twitter Stream
Personalized Filtering of Twitter StreamPersonalized Filtering of Twitter Stream
Personalized Filtering of Twitter Stream
 

Último

Grade 9 Q4-MELC1-Active and Passive Voice.pptx
Grade 9 Q4-MELC1-Active and Passive Voice.pptxGrade 9 Q4-MELC1-Active and Passive Voice.pptx
Grade 9 Q4-MELC1-Active and Passive Voice.pptxChelloAnnAsuncion2
 
Karra SKD Conference Presentation Revised.pptx
Karra SKD Conference Presentation Revised.pptxKarra SKD Conference Presentation Revised.pptx
Karra SKD Conference Presentation Revised.pptxAshokKarra1
 
THEORIES OF ORGANIZATION-PUBLIC ADMINISTRATION
THEORIES OF ORGANIZATION-PUBLIC ADMINISTRATIONTHEORIES OF ORGANIZATION-PUBLIC ADMINISTRATION
THEORIES OF ORGANIZATION-PUBLIC ADMINISTRATIONHumphrey A Beña
 
Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)
Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)
Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)lakshayb543
 
Like-prefer-love -hate+verb+ing & silent letters & citizenship text.pdf
Like-prefer-love -hate+verb+ing & silent letters & citizenship text.pdfLike-prefer-love -hate+verb+ing & silent letters & citizenship text.pdf
Like-prefer-love -hate+verb+ing & silent letters & citizenship text.pdfMr Bounab Samir
 
DATA STRUCTURE AND ALGORITHM for beginners
DATA STRUCTURE AND ALGORITHM for beginnersDATA STRUCTURE AND ALGORITHM for beginners
DATA STRUCTURE AND ALGORITHM for beginnersSabitha Banu
 
Field Attribute Index Feature in Odoo 17
Field Attribute Index Feature in Odoo 17Field Attribute Index Feature in Odoo 17
Field Attribute Index Feature in Odoo 17Celine George
 
Computed Fields and api Depends in the Odoo 17
Computed Fields and api Depends in the Odoo 17Computed Fields and api Depends in the Odoo 17
Computed Fields and api Depends in the Odoo 17Celine George
 
Science 7 Quarter 4 Module 2: Natural Resources.pptx
Science 7 Quarter 4 Module 2: Natural Resources.pptxScience 7 Quarter 4 Module 2: Natural Resources.pptx
Science 7 Quarter 4 Module 2: Natural Resources.pptxMaryGraceBautista27
 
Q4 English4 Week3 PPT Melcnmg-based.pptx
Q4 English4 Week3 PPT Melcnmg-based.pptxQ4 English4 Week3 PPT Melcnmg-based.pptx
Q4 English4 Week3 PPT Melcnmg-based.pptxnelietumpap1
 
ACC 2024 Chronicles. Cardiology. Exam.pdf
ACC 2024 Chronicles. Cardiology. Exam.pdfACC 2024 Chronicles. Cardiology. Exam.pdf
ACC 2024 Chronicles. Cardiology. Exam.pdfSpandanaRallapalli
 
How to do quick user assign in kanban in Odoo 17 ERP
How to do quick user assign in kanban in Odoo 17 ERPHow to do quick user assign in kanban in Odoo 17 ERP
How to do quick user assign in kanban in Odoo 17 ERPCeline George
 
Procuring digital preservation CAN be quick and painless with our new dynamic...
Procuring digital preservation CAN be quick and painless with our new dynamic...Procuring digital preservation CAN be quick and painless with our new dynamic...
Procuring digital preservation CAN be quick and painless with our new dynamic...Jisc
 
Full Stack Web Development Course for Beginners
Full Stack Web Development Course  for BeginnersFull Stack Web Development Course  for Beginners
Full Stack Web Development Course for BeginnersSabitha Banu
 
ENGLISH 7_Q4_LESSON 2_ Employing a Variety of Strategies for Effective Interp...
ENGLISH 7_Q4_LESSON 2_ Employing a Variety of Strategies for Effective Interp...ENGLISH 7_Q4_LESSON 2_ Employing a Variety of Strategies for Effective Interp...
ENGLISH 7_Q4_LESSON 2_ Employing a Variety of Strategies for Effective Interp...JhezDiaz1
 
MULTIDISCIPLINRY NATURE OF THE ENVIRONMENTAL STUDIES.pptx
MULTIDISCIPLINRY NATURE OF THE ENVIRONMENTAL STUDIES.pptxMULTIDISCIPLINRY NATURE OF THE ENVIRONMENTAL STUDIES.pptx
MULTIDISCIPLINRY NATURE OF THE ENVIRONMENTAL STUDIES.pptxAnupkumar Sharma
 
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPTECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPTiammrhaywood
 

Último (20)

Grade 9 Q4-MELC1-Active and Passive Voice.pptx
Grade 9 Q4-MELC1-Active and Passive Voice.pptxGrade 9 Q4-MELC1-Active and Passive Voice.pptx
Grade 9 Q4-MELC1-Active and Passive Voice.pptx
 
Karra SKD Conference Presentation Revised.pptx
Karra SKD Conference Presentation Revised.pptxKarra SKD Conference Presentation Revised.pptx
Karra SKD Conference Presentation Revised.pptx
 
THEORIES OF ORGANIZATION-PUBLIC ADMINISTRATION
THEORIES OF ORGANIZATION-PUBLIC ADMINISTRATIONTHEORIES OF ORGANIZATION-PUBLIC ADMINISTRATION
THEORIES OF ORGANIZATION-PUBLIC ADMINISTRATION
 
Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)
Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)
Visit to a blind student's school🧑‍🦯🧑‍🦯(community medicine)
 
Like-prefer-love -hate+verb+ing & silent letters & citizenship text.pdf
Like-prefer-love -hate+verb+ing & silent letters & citizenship text.pdfLike-prefer-love -hate+verb+ing & silent letters & citizenship text.pdf
Like-prefer-love -hate+verb+ing & silent letters & citizenship text.pdf
 
DATA STRUCTURE AND ALGORITHM for beginners
DATA STRUCTURE AND ALGORITHM for beginnersDATA STRUCTURE AND ALGORITHM for beginners
DATA STRUCTURE AND ALGORITHM for beginners
 
Field Attribute Index Feature in Odoo 17
Field Attribute Index Feature in Odoo 17Field Attribute Index Feature in Odoo 17
Field Attribute Index Feature in Odoo 17
 
Computed Fields and api Depends in the Odoo 17
Computed Fields and api Depends in the Odoo 17Computed Fields and api Depends in the Odoo 17
Computed Fields and api Depends in the Odoo 17
 
Science 7 Quarter 4 Module 2: Natural Resources.pptx
Science 7 Quarter 4 Module 2: Natural Resources.pptxScience 7 Quarter 4 Module 2: Natural Resources.pptx
Science 7 Quarter 4 Module 2: Natural Resources.pptx
 
Q4 English4 Week3 PPT Melcnmg-based.pptx
Q4 English4 Week3 PPT Melcnmg-based.pptxQ4 English4 Week3 PPT Melcnmg-based.pptx
Q4 English4 Week3 PPT Melcnmg-based.pptx
 
ACC 2024 Chronicles. Cardiology. Exam.pdf
ACC 2024 Chronicles. Cardiology. Exam.pdfACC 2024 Chronicles. Cardiology. Exam.pdf
ACC 2024 Chronicles. Cardiology. Exam.pdf
 
How to do quick user assign in kanban in Odoo 17 ERP
How to do quick user assign in kanban in Odoo 17 ERPHow to do quick user assign in kanban in Odoo 17 ERP
How to do quick user assign in kanban in Odoo 17 ERP
 
Procuring digital preservation CAN be quick and painless with our new dynamic...
Procuring digital preservation CAN be quick and painless with our new dynamic...Procuring digital preservation CAN be quick and painless with our new dynamic...
Procuring digital preservation CAN be quick and painless with our new dynamic...
 
FINALS_OF_LEFT_ON_C'N_EL_DORADO_2024.pptx
FINALS_OF_LEFT_ON_C'N_EL_DORADO_2024.pptxFINALS_OF_LEFT_ON_C'N_EL_DORADO_2024.pptx
FINALS_OF_LEFT_ON_C'N_EL_DORADO_2024.pptx
 
Full Stack Web Development Course for Beginners
Full Stack Web Development Course  for BeginnersFull Stack Web Development Course  for Beginners
Full Stack Web Development Course for Beginners
 
TataKelola dan KamSiber Kecerdasan Buatan v022.pdf
TataKelola dan KamSiber Kecerdasan Buatan v022.pdfTataKelola dan KamSiber Kecerdasan Buatan v022.pdf
TataKelola dan KamSiber Kecerdasan Buatan v022.pdf
 
YOUVE GOT EMAIL_FINALS_EL_DORADO_2024.pptx
YOUVE GOT EMAIL_FINALS_EL_DORADO_2024.pptxYOUVE GOT EMAIL_FINALS_EL_DORADO_2024.pptx
YOUVE GOT EMAIL_FINALS_EL_DORADO_2024.pptx
 
ENGLISH 7_Q4_LESSON 2_ Employing a Variety of Strategies for Effective Interp...
ENGLISH 7_Q4_LESSON 2_ Employing a Variety of Strategies for Effective Interp...ENGLISH 7_Q4_LESSON 2_ Employing a Variety of Strategies for Effective Interp...
ENGLISH 7_Q4_LESSON 2_ Employing a Variety of Strategies for Effective Interp...
 
MULTIDISCIPLINRY NATURE OF THE ENVIRONMENTAL STUDIES.pptx
MULTIDISCIPLINRY NATURE OF THE ENVIRONMENTAL STUDIES.pptxMULTIDISCIPLINRY NATURE OF THE ENVIRONMENTAL STUDIES.pptx
MULTIDISCIPLINRY NATURE OF THE ENVIRONMENTAL STUDIES.pptx
 
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPTECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
ECONOMIC CONTEXT - LONG FORM TV DRAMA - PPT
 

Random walk on Graphs

  • 1. Random Walk on Graphs Pavan Kapanipathi Reading Group (Kno.e.sis) Referred: Purnamrita Sarkar, Random Walks on Graphs: An Overview
  • 2. Agenda • Introduction – Motivation • Background – Graphs – Matrices • Random Walk – PageRank – Personalized PageRank – Topic Sensitive PageRank • Applications – Specifically in Recommender Systems
  • 3. Random Walk A drunk man will find his way home, but a drunk bird may get lost forever.
  • 4. Motivation: Link prediction in social networks 4
  • 5. Motivation: Basis for recommendation 5
  • 6. Since I had very less slides and More time – Graphs • Undirected Graphs
  • 7. Since I had very less slides and more time in hand-- Graphs • Directed Graphs
  • 8. Since I had very less slides and more time in hand – Matrix Rows Columns i j i j k i,j k
  • 9. Adjacency and Transition Matrix Transition matrix P Adjacency matrix A 1 1 1 1 1 1 1/2 1/2 9
  • 10. Markov Property (Basic) • Given the present state, the future and past states are independent
  • 11. Stochastic • Wikipedia: In probability theory, a purely stochastic system is one whose state is nondeterministic so that the subsequent state of the system is determined probabilistically. 1 1 • Matrix – Stochastic Row/Column? 1/2 1/2
  • 12. Random Walk on Graphs
  • 13. Random Walk on Graphs
  • 14. Random Walk on Graphs
  • 15. Random Walk on Graphs The random sequence of points selected this way is a random walk on the graph
  • 16. Again: Transition Matrix j k i i j k Transition matrix P Probability? 1 1 1/2 1/2 16
  • 17. Probability Distributions • xt(i) = probability that the surfer is at node i at time t • xt+1(i) = ∑j(Probability of being at node j)*Pr(j->i) =∑jxt(j)*P(j,i) • xt+1 = xtP = xt-1*P*P= xt-2*P*P*P = …=x0 Pt Matrix Multiplication? • What happens when the surfer keeps walking for a long time? 17
  • 18. Property of Adjacency Matrix Adjacency matrix A 1 1 1 1
  • 19. What is a stationary distribution? Intuitively and Mathematically • The stationary distribution at a node is related to the amount of time a random walker spends visiting that node. • Remember that we can write the probability distribution at a node as – xt+1 = xtP • For the stationary distribution v0 we have – v0 = v0 P • Whoa! that’s just the left eigenvector of the transition matrix ! 19
  • 20. Eigen Value and Eigen Vector?
  • 21. Interesting questions • Does a stationary distribution always exist? Is it unique? – Yes, if the graph is “well-behaved”. • What is “well-behaved”? – We shall talk about this soon. • How fast will the random surfer approach this stationary distribution? – Mixing Time! 21
  • 22. Well behaved graphs • Irreducible: There is a path from every node to every other node. What about connected undirected Graph? Irreducible Not irreducible 22
  • 23. Well behaved graphs • Aperiodic: The GCD of all cycle lengths is 1. The GCD is also called period. Periodicity is 3 Aperiodic 23
  • 24. Implications of the Perron Frobenius Theorem • If a markov chain is irreducible and aperiodic then the largest eigenvalue of the transition matrix will be equal to 1 and all the other eigenvalues will be strictly less than 1. – Let the eigenvalues of P be {σi| i=0:n-1} in non-increasing order of σi . – σ0 = 1 > σ1 > σ2 >= ……>= σn • These results imply that for a well behaved graph there exists an unique stationary distribution. • More details when we discuss pagerank. 24
  • 25. Some fun stuff about undirected graphs • A connected undirected graph is irreducible • A connected non-bipartite undirected graph has a stationary distribution proportional to the degree distribution! • Makes sense, since larger the degree of the node more likely a random walk is to come back to it. 25
  • 26. Proximity measures from random walks b a • How long does it take to hit node b in a random walk starting at node a? --- Hitting time. • How long does it take to hit node b and come back to node a? --- Commute time. 26
  • 27. Hitting and Commute times b a • Hitting time from node i to node j – Expected number of hops to hit node j starting at node i. – Is not symmetric. h(a,b) > h(a,b) – h(i,j) = 1 + ΣkЄnbs(A) p(i,k)h(k,j) 27
  • 28. Hitting and Commute times b a • Commute time between node i and j – Is expected time to hit node j and come back to i – c(i,j) = h(i,j) + h(j,i) – Is symmetric. c(a,b) = c(b,a) 28
  • 29. Random Walk (versions) • PageRank – Personalized PageRank – Topic Sensitive PageRank • Recommender Systems (My interests)
  • 30. Recommender Networks • For a customer node i define similarity as – H(i,j) – C(i,j) – Or the cosine similarity  Lij  Lii Ljj • Now the question is how to compute these quantities quickly for very large graphs. – Fast iterative techniques (Brand 2005) – Fast Random Walk with Restart (Tong, Faloutsos 2006) – Finding nearest neighbors in graphs (Sarkar, Moore 2007) 30
  • 31. PageRank (Initial) • Intuition – PageRank of “A” is higher if the pages that links to “A” has higher PageRank • User behavior where a surfer clicks on links at random with no regard towards content – One page's PageRank is not completely passed on to a page it links to, but is divided by the number of links on the page.
  • 32. PageRank • Intuitively v (i )  v (j ) i degout ( j ) j  • v works out to be the stationary distribution of the markov chain corresponding to the web.
  • 33. Pagerank & Perron-frobenius • Perron Frobenius only holds if the graph is irreducible and aperiodic. • But how can we guarantee that for the web graph? – Do it with a small restart probability c. • At any time-step the random surfer – jumps (teleport) to any other node with probability c – jumps to its direct neighbors with total probability 1-c. ~ P  (1  c )P  cU Uij  1 n i , j ~ P  cP  (1  c)U U ij  1 i, j n 33
  • 34. Pagerank • We are looking for the vector v s.t. v  (1  c ) vP  cr • r is a distribution over web-pages. • If r is the uniform distribution we get pagerank. • What happens if r is non-uniform? Personalization 34
  • 35. Personalized Pagerank1,2,3 • The only difference is that we use a non-uniform teleportation distribution, i.e. at any time step teleport to a set of webpages. • In other words we are looking for the vector v s.t. v  (1  c ) vP  cr • r is a non-uniform preference vector specific to an user. • v gives “personalized views” of the web. 1. Scaling Personalized Web Search, Jeh, Widom. 2003 2. Topic-sensitive PageRank, Haveliwala, 2001 3. Towards scaling fully personalized pagerank, D. Fogaras and B. Racz, 2004 35
  • 36. Topic-sensitive pagerank (Haveliwala’01) • Divide the webpages into 16 broad categories • For each category compute the biased personalized pagerank vector by uniformly teleporting to websites under that category. • At query time the probability of the query being from any of the above classes is computed, and the final page-rank vector is computed by a linear combination of the biased pagerank vectors computed offline. 36
  • 37. Random Walk for Recommendations • Collaborative Filtering by Shang et.al • Graph – Vertices: Users (U), Items (I), Item Information (T) and User Profiles (P) – Edges with weights • • • • u has a rating for i i has a tag for t u belongs to profile category p u to u if they are connected in the social network • Edge weights assignments
  • 38. Random Walk for Recommendations Connectivity/Transition Generally 0.85 Preference Vector
  • 39. Most of it from • Purnamrita Sarkar, Random Walks on Graphs: An Overview • Random Walks on Graphs: A Survey, Laszlo Lov'asz • OBVIOUSLY: Wikipedia :D • Random Walk on Graphs: Ankit Agarwal

Notas do Editor

  1. Probability of a random surfer to go from node k, j
  2. Continuing the transition Matrix How many know how matrix multiplication
  3. Probability of reaching node “j” from “I” node in path 2
  4. Almost all vectors change direction when multiplied by a matrix A. Certain exceptional vectors x are in the same direction when multiplied with A. They are called Eigen Vectors.
  5. 0 in P-Matrix will reoccur– Connected undirected graph is irreducible
  6. Why is periodicity important?Because if it is periodic then the nodes will be reached infinite number of timesI will get into the details if time and necessary Periodicity[edit]A state i has period k if any return to state i must occur in multiples of k time steps. Formally, the period of a state is defined as(where "gcd" is the greatest common divisor). Note that even though a state has period k, it may not be possible to reach the state in k steps. For example, suppose it is possible to return to the state in {6, 8, 10, 12, ...} time steps; k would be 2, even though 2 does not appear in this list.If k = 1, then the state is said to be aperiodic: returns to state i can occur at irregular times. In other words, a state i is aperiodic if there exists n such that for all n' ≥ n,Otherwise (k > 1), the state is said to be periodic with period k. A Markov chain is aperiodic if every state is aperiodic. An irreducible markov chain only needs one aperiodic state to imply all states are aperiodic.Every state of a bipartite graph has an even period.
  7. It is more intuitive
  8. Very simple intuition
  9. C is the damping factor
  10. vP should not change the value
  11. Manipulating these