Social Sensing : Building Reliable Systems on Unreliable Data.

Front Cover -- Social Sensing: Building Reliable Systems on Unreliable Data -- Copyright -- Dedication -- Contents -- Acknowledgments -- Authors -- Dong Wang -- Tarek Abdelzaher -- Lance M. Kaplan -- Foreword -- Preface -- Chapter 1: A new information age -- 1.1 Overview -- 1.2 Challenges -- 1.3 State of the Art -- 1.3.1 Efforts on Discount Fusion -- 1.3.2 Efforts on Trust and Reputation Systems -- 1.3.3 Efforts on Fact-Finding -- 1.4 Organization -- Chapter 2: Social Sensing Trends and Applications -- 2.1 Information Sharing: The Paradigm Shift -- 2.2 An Application Taxonomy -- 2.3 Early Research -- 2.4 The Present Time -- 2.5 ANote on Privacy -- Chapter 3: Mathematical foundations of social sensing: An introductory tutorial -- 3.1 AMultidisciplinary Background -- 3.2 Basics of Generic Networks -- 3.3 Basics of Bayesian Analysis -- 3.4 Basics of Maximum Likelihood Estimation -- 3.5 Basics of Expectation Maximization -- 3.6 Basics of Confidence Intervals -- 3.7 Putting It All Together -- Chapter 4: Fact-finding in information networks -- 4.1 Facts, Fact-Finders, and the Existence of Ground Truth -- 4.2 Overview of Fact-Finders in Information Networks -- 4.3 A Bayesian Interpretation of Basic Fact-Finding -- 4.3.1 Claim Credibility -- 4.3.2 Source Credibility -- 4.4 The Iterative Algorithm -- 4.5 Examples and Results -- 4.6 Discussion -- Appendix -- Chapter 5: Social Sensing: A maximum likelihood estimation approach -- 5.1 The Social Sensing Problem -- 5.2 Expectation Maximization -- 5.2.1 Background -- 5.2.2 Mathematical Formulation -- 5.2.3 Deriving the E-Step and M-Step -- 5.3 The EM Fact-Finding Algorithm -- 5.4 Examples and Results -- 5.4.1 A Simulation Study -- 5.4.2 A Geotagging Case Study -- 5.4.3 A Real World Application -- 5.5 Discussion -- Chapter 6: Confidence bounds in social sensing -- 6.1 The Reliability Assurance Problem.

6.2 Actual Cramer-Rao Lower Bound -- 6.3 Asymptotic Cramer-Rao Lower Bound -- 6.4 Confidence Interval Derivation -- 6.5 Examples and Results -- 6.5.1 Evaluation of Confidence Interval -- 6.5.2 Evaluation of CRLB -- Scalability study -- Trustworthiness and assertiveness study -- Robustness study -- 6.5.3 Evaluation of Estimated False Positives/Negatives on Claim Classification -- Scalability study -- Trustworthiness and assertiveness study -- Robustness study -- 6.5.4 AReal World Case Study -- 6.6 Discussion -- Appendix -- Chapter 7: Resolving conflicting observations and non-binary claims -- 7.1 Handling Conflicting Binary Observations -- 7.1.1 Extended Model -- 7.1.2 Re-Derive the E-Step and M-Step -- 7.1.3 The Binary Conflict EM Algorithm -- 7.2 Handling Non-Binary Claims -- 7.2.1 Generalized E and M Steps for Non-Binary Measured Variables -- 7.2.2 The Generalized EM Algorithm for Non-Binary Measured Variables -- 7.3 Performance Evaluation -- 7.3.1 AReal World Application -- 7.3.2 ASimulation Study for Conflicting Observations -- 7.3.3 ASimulation Study for Non-Binary Claims -- 7.4 Discussion -- Appendix -- Chapter 8: Understanding the social network -- 8.1 Information Propagation Cascades -- 8.2 ABinary Model of Human Sensing -- 8.2.1 ABinary Sensor Model -- 8.2.2 Uncertain Provenance -- 8.2.3 AWord on Simplicity -- 8.3 Inferring the Social Network -- 8.3.1 Data Collection -- 8.3.2 Computing the Source-Claim Graph -- 8.3.3 Inferring the Social Network -- 8.3.4 Solving the Estimation Problem -- 8.4 ASocial-Aware Algorithm -- 8.4.1 Deriving the Likelihood -- 8.4.2 Deriving the E-Step and M-Step -- 8.4.3 The Social-Aware EM Algorithm -- 8.5 Evaluation -- 8.6 Discussion and Limitations -- Chapter 9: Understanding physical dependencies -- 9.1 Correlations in the Physical World -- 9.2 Accounting for the Opportunity to Observe.

9.2.1 Deriving the Likelihood -- 9.2.2 The OtO EM Algorithm -- 9.3 Accounting for Physical Dependencies -- 9.3.1 Deriving the Likelihood -- 9.3.2 The OtO+DV Algorithm -- 9.4 Real-World Case Studies -- 9.4.1 Opportunity to Observe -- 9.4.2 Dependent Variables -- 9.5 Discussion -- Appendix -- Derivation of the E-Step and M-Step of OtO EM -- Derivation of E-Step and M-Step of DV and OtO+DV EM -- Chapter 10: Recursive fact-finding -- 10.1 Real Time Social Sensing -- 10.2 A Streaming Truth Estimation Model -- 10.3 Dynamics and the Recursive Algorithm -- 10.3.1 The Derivation -- 10.3.2 The Recursive EM Algorithm -- 10.4 Performance Evaluation -- 10.4.1 Simulation Study -- 10.4.2 AReal World Case Study -- 10.5 Discussion -- Chapter 11: Further readings -- 11.1 Estimation Theory -- 11.2 Data Quality and Trust Analysis -- 11.3 Outlier Analysis and Attack Detection -- 11.4 Recommender Systems -- 11.5 Surveys and Opinion Polling -- Chapter 12: Conclusions and future challenges -- 12.1 Summary and Conclusions -- 12.2 Remaining Challenges and Future Work -- References -- Index -- Front Cover.

Description based on publisher supplied metadata and other sources.

Electronic reproduction. Ann Arbor, Michigan : ProQuest Ebook Central, 2024. Available via World Wide Web. Access may be limited to ProQuest Ebook Central affiliated libraries.