Assessment of Multi-Hop Interpersonal Trust in Social Networks by 3VSL

Similar documents
Computational Trust. Perspectives in Concurrency Theory Chennai December Mogens Nielsen University of Aarhus DK

A Tunable Mechanism for Identifying Trusted Nodes in Large Scale Distributed Networks

Interact with Strangers

MULTINOMIAL AGENT S TRUST MODELING USING ENTROPY OF THE DIRICHLET DISTRIBUTION

Social Interaction Based Video Recommendation: Recommending YouTube Videos to Facebook Users

Computational Trust. SFM 11 Bertinoro June Mogens Nielsen University of Aarhus DK. Aarhus Graduate School of Science Mogens Nielsen

Formal Trust Model for Multiagent Systems

Final Exam December 12, 2017

Fission of Opinions in Subjective Logic

Exploring the Patterns of Human Mobility Using Heterogeneous Traffic Trajectory Data

Multiplication of Multinomial Subjective Opinions

Believe it Today or Tomorrow? Detecting Untrustworthy Information from Dynamic Multi-Source Data

B-trust: Bayesian Trust Framework for Pervasive Computing

Modeling the Dynamic Trust of Online Service Providers using HMM

Final Exam December 12, 2017

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

Published in: Tenth Tbilisi Symposium on Language, Logic and Computation: Gudauri, Georgia, September 2013

Automatic Differentiation Equipped Variable Elimination for Sensitivity Analysis on Probabilistic Inference Queries

On Beta Models with Trust Chains

An analysis of the exponential decay principle in probabilistic trust models

Continuous Ratings in Discrete Bayesian Reputation Systems

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

INVERTING CONDITIONAL OPINIONS IN SUBJECTIVE LOGIC

A Reputation-Based Approach for Securing Vivaldi Embedding

Distributed ML for DOSNs: giving power back to users

Logic and Artificial Intelligence Lecture 13

A Bayesian model for event-based trust

Machine Learning Summer School

Semantic Constraints for Trust Transitivity

To Trust or Not to Trust? Predicting Online Trusts using Trust Antecedent Framework

Optimal Social Trust Path Selection in Complex Social Networks

CS 4649/7649 Robot Intelligence: Planning

Generalising Bayes Theorem in Subjective Logic

Mihai Boicu, Gheorghe Tecuci, Dorin Marcu Learning Agent Center, George Mason University

CSE 473: Artificial Intelligence

Chapter 4 Dynamic Bayesian Networks Fall Jin Gu, Michael Zhang

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

Planning Under Uncertainty II

Probability. CS 3793/5233 Artificial Intelligence Probability 1

CS 188: Artificial Intelligence Fall 2011

STAR: Semiring Trust Inference for Trust-Aware Social Recommenders

Is It Harmful When Advisors Only Pretend to Be Honest?

Introduction to Artificial Intelligence (AI)

On the Complexity of the Minimum Independent Set Partition Problem

Lecture 6: Graphical Models: Learning

Density Propagation for Continuous Temporal Chains Generative and Discriminative Models

DS504/CS586: Big Data Analytics Graph Mining II

RaRE: Social Rank Regulated Large-scale Network Embedding

A Heuristic Algorithm for Trust-Oriented Service Provider Selection in Complex Social Networks

Overview and comparison of random walk based techniques for estimating network averages

FlowTrust: Trust Inference with Network Flows

Probabilistic Logic under Uncertainty

Winter 2019 Math 106 Topics in Applied Mathematics. Lecture 1: Introduction

An Interoperable Context Sensitive Model of Trust

Using first-order logic, formalize the following knowledge:

Circle-based Recommendation in Online Social Networks

Probabilistic Graphical Models Homework 2: Due February 24, 2014 at 4 pm

Cooperation Stimulation in Cooperative Communications: An Indirect Reciprocity Game

Measurement of Burst Topic in Microblog

Introduction to Probabilistic Programming Languages (PPL)

DS504/CS586: Big Data Analytics Graph Mining II

TPS DR: A Universal Dimension Reducing Algorithm for Optimal Trust Path Selection in Complex Sensor Network

Truth Discovery for Spatio-Temporal Events from Crowdsourced Data

Introduction to Bayesian Learning

Outline. CSE 573: Artificial Intelligence Autumn Agent. Partial Observability. Markov Decision Process (MDP) 10/31/2012

Automating variational inference for statistics and data mining

Interpretation and Fusion of Hyper Opinions in Subjective Logic

ABSTRACT. WANG, YONGHONG. Evidence-Based Trust in Distributed Agent Systems. (Under the direction of Professor Munindar P. Singh).

Propositions. c D. Poole and A. Mackworth 2010 Artificial Intelligence, Lecture 5.1, Page 1

: Approximation and Online Algorithms with Applications Lecture Note 2: NP-Hardness

Uncertainty and knowledge. Uncertainty and knowledge. Reasoning with uncertainty. Notes

The Lady Tasting Tea. How to deal with multiple testing. Need to explore many models. More Predictive Modeling

Friends or Foes: Detecting Dishonest Recommenders in Online Social Networks

15-381: Artificial Intelligence. Hidden Markov Models (HMMs)

Digital Chart Cartography: Error and Quality control

Display Advertising Optimization by Quantum Annealing Processor

Mining Triadic Closure Patterns in Social Networks

Logic and Artificial Intelligence Lecture 7

Bayesian Networks BY: MOHAMAD ALSABBAGH

Available online at ScienceDirect. Procedia Engineering 119 (2015 ) 13 18

Logics of Rational Agency Lecture 3

Activity Recognition with a Smart Walker

Preference Elicitation for Sequential Decision Problems

Une approche multi-agent basée sur la confiance pour évaluer la performance des plateformes de crowdsourcing d idées

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

CPS 173 Mechanism design. Vincent Conitzer

Opinion Dynamics on Triad Scale Free Network

Gibbs Sampling Methods for Multiple Sequence Alignment

System Reliability Analysis. CS6323 Networks and Systems

Combining Recommender and Reputation Systems to Produce Better Online Advice

Effective Design of Trust Ontologies for Improvement in the Structure of Socio-Semantic Trust Networks

UNDER UNCERTAINTY TRUST ESTIMATION IN MULTI-VALUED SETTINGS

Data Quality in Hybrid Neuro-Fuzzy based Soft-Sensor Models: An Experimental Study

Random Surfing on Multipartite Graphs

International Journal "Information Theories & Applications" Vol.14 /

Kazuyuki Tanaka s work on AND-OR trees and subsequent development

A Generalized Decision Logic in Interval-set-valued Information Tables

Using Belief Propagation to Counter Correlated Reports in Cooperative Spectrum Sensing

Basic Probabilistic Reasoning SEG

Modeling face-to-face social interaction networks

Transcription:

Assessment of Multi-Hop Interpersonal Trust in Social Networks by 3VSL Guangchi Liu, Qing Yang, Honggang Wang, Xiaodong Lin and Mike P. Wittie Presented by Guangchi Liu Department of Computer Science Montana State University Bozeman, MT, USA E-mail: guangchi.liu@msu.montana.edu IEEE INFOCOM 2014, Toronto, CA

Motivation 2

Motivation Ask Yourself: How to trust someone you have never known before? Trustor? Trustee Alice David 3

Motivation Probably, If Trustor Trustee Alice Bob David Transitivity of Trust [1] [1] Falcone, Rino, and Cristiano Castelfranchi. "Transitivity in trust a discussed property." (2010). 4

Motivation Then, How Should Alice Trust David? Trustor Trustee Alice Bob? David 5

Motivation Complex Topology: Harder Cases?? Parallel Topology? Bridge Topology Arbitrary Topology 6

Motivation Finally, we could expect. Small World Network Six degrees of separation 7

Applications 8

Applications Consumer-to-Consumer Recommendation Systems Untrustworthy User Trustworthy User Seller Buyer 9

Applications Active Friending in OSNs [2] Weak Relation Strong Relation Trustworthy User Target User Finder [2] Yang, De-Nian, et al. "Maximizing acceptance probability for active friending in online social networks." Proceedings of the 19th ACM SIGKDD international conference on Knowledge discovery and data mining. ACM, 2013. 10

Applications Sybil Identification Untrustworthy User Trustworthy User Unknown user User? 11

Contributions 12

Contributions Methodology Three-Valued Subjective Logic (3VSL) Multi-hop trustworthiness assessment Advantage Applicable to Complex Topology Be able to handle any large scale social network Validation Real World Experiment 100+ participants in online survey 13

Related Works 14

Related Works 3VSL Bayesian Network Hidden Markov Model Subjective Logic Artificial Intelligence Methods Random Walk Community Analysis Graph Partitioning Graph Theory Methods 15

Three-Valued Subjective Logic (3VSL) 16

3VSL: Opinion Vector Interpersonal Trust: Is David telling the truth? Positive Uncertain Negative Impression Trustor Opinion Trustee Alice David 17

3VSL: Opinion Vector Mom Alice s Evidences: Posteriori Evidence? Interaction History Positive Uncertain Negative Impression 18

3VSL: Opinion Vector Stranger Alice s Evidences: Priori Evidence Cognition in Mind Positive Uncertain Negative Impression 19

3VSL: Opinion Vector Priori Evidence + + + Posteriori Evidence Positive Uncertain Negative Impression 20

3VSL: Opinion Vector Dirichlet Distribution: + f ( p, p,, ) 1 2 1 (1 p p ) p p B(,, ) 1 1 1 1 2 1 2 Positive Uncertain Negative p 1 p 2 p 3 # # # : P 1 P P 3 1 2 21

3VSL: Opinion Vector 1 f ( p, p,, ) (1 p p ) p p B(,, ) 1 1 1 1 2 1 2 1 2 Positive Uncertain Negative EP ( ) 1 EP ( 2) = EP ( ) 3 # # + # + # # # + # + # # # + # + # 22

3VSL: Opinion Vector Positive EP ( ) 1 Uncertain 2 Negative EP ( ) EP ( 3) Impression # # + # + # # # + # + # # # + # + # # + # + # Positive # + # + # # + # + # # + # + # Pos-uncertain Negative + + 23

3VSL: Opinion Vector Opinion Vector A b X Positive n A X Pos-uncertain A d X Negative A e X A a X Impression 1 A A A A A A X X X X X X [ b, d, n, e, a ] A Opinion 24 X

3VSL: Opinion Vector Expected Belief Positive Pos-uncertain Negative Impression 1 E( ) b a n e 3 A A A A A X X X X X A X 25

3VSL: Opinion Vector Is David telling the truth? Opinion of David s Mom E( ) 0.33 0.442 0.45 0.5 0.05 Opinion of a Stranger E( ) 0.33 0.1 0.5 26

Multi-hop Trust Computation in OSNs: w v i e v j G(V, E, W) v? u 27

3VSL: Discounting Operation Operation for Trust Propagation 28

3VSL: Discounting Operation A D Alice A B Bob B D David Positive Uncertain Negative Positive Uncertain Negative 29

3VSL: Discounting Operation A B B D? A D Positive? Uncertain Negative 30

3VSL: Discounting Operation B D A D Distortion Alice A B Bob B D David 31

3VSL: Discounting Operation Absorbing Process Hop # Alice A D David 2 Hops Bob B D David 1 Hop 32

3VSL: Discounting Operation A B Discounting Operation: (, ) B D A A B (, ) D B D A B B D Alice Bob David A [ b A, d A, n A, e A, a A ] B B B B B B A A B bd bb bd A A B dd bb dd (, ) n 1 b d e A B ed ed A A ad ad A B A A A B B D D D D D B [ b B, d B, n B, e B, a B ] D D D D D D 33

3VSL: Combining Operation Operation for Trust Fusion 34

3VSL: Combining Operation A 1 D A D Alice David A 2 D A 1 D A 2 D Positive Positive Uncertain Uncertain Negative Negative 35

3VSL: Combining Operation Posteriori Evidences Priori Evidences A 1 D A 1 D Positive + Positive Pos-uncertain Negative + + Pos-uncertain Negative Independent Evidences Dependent Evidences Positive Pos-uncertain Negative A D Positive Pos-uncertain Normalize Negative 36

3VSL: Combining Operation A1 A2 Combining Operation: (, ) Alice A1 A2 (, ) A D D D A 1 D A 2 D David [ b, d, n, e, a ] A 1 1 1 1 1 1 D D D D D D A A A A A A D D D D D D [ b, d, n, e, a ] 2 2 2 2 2 2 D D b d 1 2 (, ) n A1 A2 A2 A1 A D D D D D A A A A ed ed ed ed 1 2 2 1 A1 A2 A2 A1 A D D D D D A A A A ed ed ed ed 1 2 2 1 A1 A2 A2 A1 A A A D D D D D D D A A A A ed ed ed ed e a 1 2 2 1 A2 A1 A D D D A A A A ed ed ed ed A D a b e A D b e d e d e n e e e n e 1 2 2 1 37

3VSL: Property Method 1: A B1 A B2 ( (, ), (, )) B C B C Method 2: A B1 B2 (, (, ) B C C Step.1 Step.1 Step.2 Equivalent Step.2 Step.3 38

3VSL: Topologies Bridge Topology Arbitrary Topology Large Social Networks 39

Evaluation 40

Evaluation: Real World Validation Goal Accuracy of 3VSL in Real Social Network Topology 1. 2. Method : Online Questionnaire [3,4] (100+ samples) 1. Error = ( ) ( ) 2. Error = ( ) ( ) [3]. Xiaoming Li and Guangchi Liu. Online Trust Survey. http://www.cs.montana.edu/yang/trust-us, 2013. [4]. C. Johnson-George and W. C. Swap, Measurement of specific interpersonal trust: Construction and validation of a scale to assess trust in a specific other. Journal of Personality and Social Psychology, vol. 43, no. 6, pp. 1306 1317, 1982. 41

Evaluation: Real World Validation 42

Evaluation: Numerical Analysis Goal 3VSL Vs. Common Sense Topology 1. 2. 3. Method E( )? 43

Evaluation: Numerical Analysis E( )? 44

Evaluation: Numerical Analysis E( )? 2 1 45

Evaluation: Numerical Analysis E( )? 46

Conclusion 47

Conclusion Three-Valued Subjective Logic 48

Thank You! Q & A Guangchi Liu (Luke) 刘光迟 ( 劉光遲 ) Networking Lab, Dept. of Computer Science Montana State University, MT, United States LinkedIn: www.linkedin.com/pub/guangchi-liu/60/a96/961/ E-mail: guangchi.liu@msu.montana.edu IEEE INFOCOM 2014, Toronto, CA

Backup I Trust Trust (Cognitive) Trust (Probabilistic) 50

Backup II Single hop trust assessment Wang, Yao, and Julita Vassileva. "Trust and reputation model in peer-to-peer networks." Multi Hop Trust Computation Jøsang, Audun. "A logic for uncertain probabilities." Bayesian Network Trust Assessment (AI Methods) Subjective Logic (SL) HMM Model Dynamic of Trust ElSalamouny, Ehab, Vladimiro Sassone, and Mogens Nielsen. "HMM-based trust model." 51

Backup III Subjective Logic 3VSL (, ) 1 2 (, ) 1 2 Collapse Constant Large Social Networks Large Social Networks 52