MobiHoc 2014 MINIMUM-SIZED INFLUENTIAL NODE SET SELECTION FOR SOCIAL NETWORKS UNDER THE INDEPENDENT CASCADE MODEL

Similar documents
Minimum-sized Positive Influential Node Set Selection for Social Networks: Considering Both Positive and Negative Influences

CSCI 3210: Computational Game Theory. Cascading Behavior in Networks Ref: [AGT] Ch 24

Web Structure Mining Nodes, Links and Influence

Personalized Social Recommendations Accurate or Private

Efficient Respondents Selection for Biased Survey using Online Social Networks

Maximizing the Spread of Influence through a Social Network. David Kempe, Jon Kleinberg, Éva Tardos SIGKDD 03

DS504/CS586: Big Data Analytics Graph Mining II

Diffusion of Innovation and Influence Maximization

DS504/CS586: Big Data Analytics Graph Mining II

An Efficient reconciliation algorithm for social networks

Diffusion of Innovation

Interact with Strangers

Modeling, Analysis, and Control of Information Propagation in Multi-layer and Multiplex Networks. Osman Yağan

ECS 253 / MAE 253, Lecture 15 May 17, I. Probability generating function recap

Influence Maximization in Dynamic Social Networks

Performance Evaluation. Analyzing competitive influence maximization problems with partial information: An approximation algorithmic framework

A Note on Maximizing the Spread of Influence in Social Networks

Cost and Preference in Recommender Systems Junhua Chen LESS IS MORE

Analysis of Multiview Legislative Networks with Structured Matrix Factorization: Does Twitter Influence Translate to the Real World?

Adaptive Rumor Spreading

Mining Newsgroups Using Networks Arising From Social Behavior by Rakesh Agrawal et al. Presented by Will Lee

Social Influence in Online Social Networks. Epidemiological Models. Epidemic Process

Kristina Lerman USC Information Sciences Institute

CS224W: Analysis of Networks Jure Leskovec, Stanford University

Finding central nodes in large networks

Structure based Data De-anonymization of Social Networks and Mobility Traces

Maximizing the Spread of Influence through a Social Network

ORIE 4741: Learning with Big Messy Data. Spectral Graph Theory

Time-Critical Influence Maximization in Social Networks with Time-Delayed Diffusion Process

Wiki Definition. Reputation Systems I. Outline. Introduction to Reputations. Yury Lifshits. HITS, PageRank, SALSA, ebay, EigenTrust, VKontakte

Determining the Diameter of Small World Networks

Stability and Robustness in Influence Maximization

Influence Blocking Maximization in Social Networks under the Competitive Linear Threshold Model

6.207/14.15: Networks Lecture 7: Search on Networks: Navigation and Web Search

CS224W: Social and Information Network Analysis Jure Leskovec, Stanford University

Analytically tractable processes on networks

CS224W: Social and Information Network Analysis Jure Leskovec, Stanford University

Info-Cluster Based Regional Influence Analysis in Social Networks

Lab 8: Measuring Graph Centrality - PageRank. Monday, November 5 CompSci 531, Fall 2018

Reducing Computation Time for the Analysis of Large Social Science Datasets

Friends or Foes: Detecting Dishonest Recommenders in Online Social Networks

On the Efficiency of Influence-and-Exploit Strategies for Revenue Maximization under Positive Externalities

Greedy Maximization Framework for Graph-based Influence Functions

Evaluating Physical, Chemical, and Biological Impacts from the Savannah Harbor Expansion Project Cooperative Agreement Number W912HZ

Distributed Systems Gossip Algorithms

Mining Triadic Closure Patterns in Social Networks

Influence Spreading Path and its Application to the Time Constrained Social Influence Maximization Problem and Beyond

When Social Influence Meets Item Inference

Polyhedral Approaches to Online Bipartite Matching

CS224W: Social and Information Network Analysis Jure Leskovec, Stanford University

9. Submodular function optimization

WITH the recent advancements of information technologies,

Least cost influence propagation in (social) networks

Continuous Influence Maximization: What Discounts Should We Offer to Social Network Users?

From Competition to Complementarity: Comparative Influence Diffusion and Maximization

Ant Colony Optimization: an introduction. Daniel Chivilikhin

Jure Leskovec Stanford University

Multi-Round Influence Maximization

1. REPRESENTATIVE PROBLEMS

CSI 445/660 Part 6 (Centrality Measures for Networks) 6 1 / 68

Cascading Behavior in Networks: Algorithmic and Economic Issues

Mining Structural Hole Spanners Through Information Diffusion in Social Networks

EE595A Submodular functions, their optimization and applications Spring 2011

A Bivariate Point Process Model with Application to Social Media User Content Generation

Memory-Efficient Low Rank Approximation of Massive Graphs

Online Social Networks and Media. Opinion formation on social networks

Using Crowdsourced Data in Location-based Social Networks to Explore Influence Maximization

Lecture 11 October 11, Information Dissemination through Social Networks

LEAD FROM THE FRONT MEDIA KIT

Minimizing Seed Set Selection with Probabilistic Coverage Guarantee in a Social Network

Modelling self-organizing networks

Densest subgraph computation and applications in finding events on social media

Social Networks with Competing Products

ECEN 689 Special Topics in Data Science for Communications Networks

F o r u m v e n u e :

4. How to prove a problem is NPC

1 Maximizing a Submodular Function

Dominating Set. Chapter 26

Dominating Set. Chapter Sequential Greedy Algorithm 294 CHAPTER 26. DOMINATING SET

Influence and Homophily

Purnamrita Sarkar (Carnegie Mellon) Deepayan Chakrabarti (Yahoo! Research) Andrew W. Moore (Google, Inc.)

ELEC6910Q Analytics and Systems for Social Media and Big Data Applications Lecture 3 Centrality, Similarity, and Strength Ties

CS224W: Social and Information Network Analysis Jure Leskovec, Stanford University

Whom to Ask? Jury Selection for Decision Making Tasks on Micro-blog Services

A note on modeling retweet cascades on Twitter

Online Influence Maximization

Profit Maximization for Viral Marketing in Online Social Networks

INFO 2950 Intro to Data Science. Lecture 18: Power Laws and Big Data

MS&E 233 Lecture 9 & 10: Network models

Modeling Social Media Memes as a Contagious Process

Online to Offline : Translating Media Usage To Real Life Public

Viral Marketing and the Diffusion of Trends on Social Networks

Evaluating Physical, Chemical, and Biological Impacts from the Savannah Harbor Expansion Project Cooperative Agreement Number W912HZ

Facebook Friends! and Matrix Functions

Distributed Optimization. Song Chong EE, KAIST

Viewing the minimum dominating set and maximum coverage problems motivated by word of mouth marketing in a problem decomposition context

Social Computing and Its Application in Query Suggestion

Influence Maximization in Social Networks: An Ising-model-based Approach

the open-source sky survey David W. Hogg (NYU)

Collaborative Filtering. Radek Pelánek

Transcription:

MobiHoc 2014 MINIMUM-SIZED INFLUENTIAL NODE SET SELECTION FOR SOCIAL NETWORKS UNDER THE INDEPENDENT CASCADE MODEL Jing (Selena) He Department of Computer Science, Kennesaw State University Shouling Ji, and Raheem Beyah School of Electrical and Computer Engineering, Georgia Institute of Technology Zhipeng Cai Department of Computer Science, Georgia State University

INTRODUCTION What is a social network? The graph of relationships and interactions within a group of individuals. 2

SOCIAL NETWORK AND SPREAD OF INFLUENCE Social network plays a fundamental role as a medium for the spread of INFLUENCE among its members Opinions, ideas, information, innovation Direct Marketing takes the word-of-mouth effects to significantly increase profits (facebook, twitter, myspace, ) 3

MOTIVATION 900 million users, Apr. 2012 the 3rd largest Country in the world More visitors than Google Action: Update statues, create event More than 4 billion images Action: Add tags, Add favorites Social networks already become a bridge to connect our really daily life and the virtual web space 2009, 2 billion tweets per quarter 2010, 4 billion tweets per quarter Action: Post tweets, Retweet 4

MOTIVATION (CONT.) Modeling and tracking users actions in social networks is a very important issue and can benefit many real applications Advertising Social recommendation Expert finding Marketing 5

APPLICATION George Who are the opinion leaders in a community? 2 2 Ada 1 Bob Marketer Alice Frank 4 1 Carol 2 Eve David 3 3 Find minimum-sized node (user) set in a social network that could influence on every node in the network 6

OUTLINE Network Model Model of influence Minimum-sized Influence Node Set selection problem Problem definition Greedy Algorithm Proof of performance bound Experiments Data and setting Results 7

OUTLINE Network Model Models of influence Minimum-sized Influence Node Set selection problem Problem definition Greedy Algorithm Proof of performance bound Experiments Data and setting Results 8

NETWORK MODEL A social network is represented as a undirected graph Nodes start either active or inactive An active node may trigger activation of neighboring nodes based on a pre-defined threshold τ Monotonicity assumption: active nodes never deactivate 9

OUTLINE Network Model Model of influence Minimum-sized Influence Node Set selection problem Problem definition Greedy Algorithm Proof of performance bound Experiments Data and setting Results 10

MODEL OF INFLUENCE If u 1 is active, then the active node set I = {u 1 } P 1 (I) = 1 P 2 (I) = 0.5 P 3 (I) = 0.7 P 4 (I) = 0.6 11

MODEL OF INFLUENCE P ii = 1, if u i ϵ I P ii = 0, otherwise P i (I) = 1 1 Pij τ u j I If u 1 and u 4 are active, then the active node set I = {u 1, u 4 } P 1 (I) = 1 (1 P 11 )(1 P 14 ) = 1 P 2 (I) = 1 (1 P 21 )(1 P 24 ) = 0.9 P 3 (I) = 1 (1 P 31 )(1 P 34 ) = 0.97 P 4 (I) = 1 (1 P 41 )(1 P 44 ) = 1 12

OUTLINE Network Model Model of influence Minimum-sized Influence Node Set selection problem Problem definition Greedy Algorithm Proof of performance bound Experiments Data and setting Results 13

MINIMUM-SIZED INFLUENCE NODE SET SELECTION PROBLEM (MINS) Given a social network G = (V, E, P) a threshold τ Goal The initially selected active node set denoted by I could influence every node in the network ui V, P i (I) = 1 1 Pij τ u j I Objective Minimize the size of I 14

OUTLINE Network Model Model of influence Minimum-sized Influence Node Set selection problem Problem definition Greedy Algorithm Proof of performance bound Experiments Data and setting Results 15

CONTRIBUTION FUNCTION f(i) = min (Pi I, τ) u i V Greedy algorithm Initialize I = empty set While f(i) < V τ do Choose u to maximize f(i {u}) I = I {u} End while Return I 16

EXAMPLE First round: I = empty set Second round: I = {u 1 } f(i) = 0.8 + 0.5 + 0.7 + 0.6 = 2.6 I = {u 2 } f(i) = 0.5 + 0.8 + 0.4 + 0.8 = 2.5 I = {u 3 } f(i) = 0.7 + 0.4 + 0.8 + 0.8 = 2.7 I = {u 4 } f(i) = 0.6 + 0.8 + 0.8 + 0.8 = 3.0 τ = 0.8 f(i) = min (Pi I, τ) u i V 17 17

EXAMPLE Third round: I = {u 4, u 1 } f(i) = 0.8 + 0.8 + 0.8 + 0.8 = 3.2 I = {u 4, u 2 } f(i) = 0.8 + 0.8 + 0.8 + 0.8 = 3.2 I = {u 4, u 3 } f(i) = 0.8 + 0.8 + 0.8 + 0.8 = 3.2 Use node ID to break the tie I = {u 4, u 1 } The greedy algorithm stops, since f(i) = V τ = 4 * 0.8 = 3.2. τ = 0.8 f(i) = min (Pi I, τ) u i V 18 18

OUTLINE Network Model Model of influence Minimum-sized Influence Node Set selection problem Problem definition Greedy Algorithm Proof of performance bound Experiments Data and setting Results 19

THEORETICAL ANALYSIS Theorem 1. The MINS selection problem is NP-hard. 20

OUTLINE Network Model Model of influence Minimum-sized Influence Node Set selection problem Problem definition Greedy Algorithm Proof of performance bound Experiments Data and setting Results 21

SIMULATION SETTINGS generate random graphs based on the random graph model G(n,p) = {G G has n nodes, and an edge between any pair of nodes is generated with probability p}. 22

EXPERIMENT DATA Real-world data set: academic coauthor network, which is extracted from academic search system Arnetminer [19]. co-authorship networks arguably capture many of the key features of social networks more generally. Resulting graph: 640, 134 nodes (authors), 1, 554, 643distinct edges (coauthor relations) 23

OUTLINE Network Model Model of influence Minimum-sized Influence Node Set selection problem Problem definition Greedy Algorithm Proof of performance bound Experiments Data and setting Results 24

RESULTS: SIMULATION 25

RESULTS: SIMULATION 26

RESULTS: REAL DATA 27

CONCLUSIONS We introduce a new optimization problem, named the Minimum-sized Influential Node Set (MINS) selection problem. We prove that it is a NP-hard problem under the independent cascade model. We define a polymatroid contribution function, which suggests us a greedy approximation algorithm. Comprehensive theoretical analysis about its performance ratio is given. We conduct extensive experiments and simulations to validate our proposed greedy algorithm both on real world coauthor data sets and random graphs. 28

FUTURE WORK Study more realistic network model Directed graph Study more general influence models Deal with negative influences Study the network evolution as time changes 29

Q & A 30