ISBN:
9783319510491
Language:
English
Pages:
1 Online-Ressource (231 pages)
Series Statement:
Lecture Notes in Social Networks
Series Statement:
Lecture Notes in Social Networks Ser.
Parallel Title:
Print version Kawash, Jalal Prediction and Inference from Social Networks and Social Media
DDC:
004
Keywords:
Computer science
;
Computer science
;
Electronic books
Abstract:
Preface -- Contents -- 1 Having Fun?: Personalized Activity-Based Mood Prediction in Social Media -- 1 Introduction -- 2 Related Work -- 3 Social Media Data -- 3.1 Twitter Dataset -- 3.2 Ground Truth -- 4 Features -- 5 Prediction -- 5.1 Prediction Framework -- 5.2 General Prediction Results -- 5.3 Personalized Prediction Results -- 6 Conclusion and Future Work -- References -- 2 Automatic Medical Image Multilingual Indexation Through a Medical Social Network -- 1 Introduction -- 2 Related Work -- 2.1 Medical Social Networks -- 2.2 Multilingual Indexation Approaches -- 2.2.1 An Overview -- 2.2.2 Indexation Approaches via Social Networks -- 3 Social Network Architecture Description and Implementation -- 4 The Proposed Methodology -- 4.1 Comments' Pre-processing -- 4.2 Cleaning, Correcting, and Lemmatization -- 4.2.1 Cleaning -- 4.2.2 Correcting Words -- 4.2.3 Lemmatization Words -- 4.3 Terms' Extraction -- 4.3.1 Simple Terms' Extraction -- 4.3.2 Compound Terms' Extraction -- 4.3.3 Concepts' Extraction -- 5 Experimental Results -- 5.1 Data Test and Evaluation Criteria -- 5.2 Evaluation and Results of Our Approach -- 6 Conclusion and Future Work -- References -- 3 The Significant Effect of Overlapping Community Structures in Signed Social Networks -- 1 Introduction -- 1.1 Contribution of the Paper -- 2 Related Work -- 3 Use of Terms, Variables and Definitions -- 4 Signed Disassortative Degree Mixing and Information Diffusion Approach -- 4.1 Identifying Leaders -- 4.2 Signed Cascading Process -- 4.3 Overlapping Community-Based Ranking Algorithms -- 4.3.1 Overlapping Community-Based HITS -- 4.3.2 Overlapping Community-Based PageRank -- 4.4 Baseline OCD Methods -- 4.4.1 Signed Probabilistic Mixture Model -- 4.4.2 Multi-Objective Evolutionary Algorithm in Signed Networks -- 5 Sign Prediction -- 5.1 Classifiers -- 5.1.1 Logistic Regression
Abstract:
5.1.2 Bagging -- 5.1.3 J48 -- 5.1.4 Decision Table -- 5.1.5 Bayesian Network and Naive Bayesian -- 5.2 Sign Prediction Features -- 5.2.1 Simple Degree Sign Prediction Features -- 5.2.2 OC-HITS Sign Prediction -- 5.2.3 OC-PageRank Sign Prediction -- 6 Dataset and Metrics -- 6.1 Real World Networks -- 6.2 Synthetic Networks -- 6.3 Evaluation Metrics -- 6.3.1 Normalized Mutual Information -- 6.3.2 Modularity -- 6.3.3 Frustration -- 7 Results -- 7.1 Results of OCD -- 7.1.1 Network Size n -- 7.1.2 Average Node Degree k -- 7.1.3 Maximum Node Degree maxk -- 7.1.4 Fraction of Edges Sharing with Other Communities μ -- 7.1.5 Maximum Community Size maxc -- 7.1.6 Number of Nodes in Overlapping Communities on -- 7.1.7 Number of Communities Which Nodes in Overlapping Communities Belong to om -- 7.1.8 Fractions of Positive Connections Between Communities P+ -- 7.1.9 Experiments on Real World Network -- 7.2 Simple Degree Sign Prediction Results -- 7.2.1 OC-HITS Sign Prediction -- 7.2.2 OC-PageRank Sign Prediction -- 8 Conclusion and Future Work -- References -- 4 Extracting Relations Between Symptoms by Age-Frame Based Link Prediction -- 1 Introduction -- 2 Evolving Symptom Networks -- 3 Proposed Method -- 3.1 The Evolving Structure of Symptom Network -- 3.2 The Evolving Cases -- 3.2.1 Consistent Case -- 3.2.2 Strengthening Case -- 3.2.3 Weakening Case -- 3.3 The Proximity Score in Evolving Symptom Networks -- 3.4 The Algorithm -- 4 Experimental Results -- 5 Conclusions -- References -- 5 Link Prediction by Network Analysis -- 1 Introduction -- 2 Related Work -- 3 The Methodology -- 3.1 The Algorithm -- 3.2 Graph Database -- 4 Datasets -- 5 Experiments and Results -- 6 Conclusions -- References -- 6 Structure-Based Features for Predicting the Quality of Articles in Wikipedia -- 1 Introduction -- 2 Related Work -- 3 Problem Formulation -- 4 Quality Model
Abstract:
4.1 Notations -- 4.2 Definitions -- 4.3 Model -- 4.4 Approvement Functions -- 4.5 Calculation -- 5 Experiments -- 5.1 Wikipedia Dataset -- 5.2 Articles Features -- 5.3 Evaluation -- 5.4 Quantitative Experiments -- 5.4.1 Unsupervised Models -- 5.4.2 Supervised Scenario -- 5.5 Qualitative Interpretation -- 6 Conclusion -- References -- 7 Predicting Collective Action from Micro-Blog Data -- 1 Introduction -- 2 Related Work -- 3 Proposed Predictive Model -- 4 Evaluation Methodology -- 4.1 The 2011 London Riots -- 4.2 Data Collection -- 4.3 Data Analysis -- 5 Results -- 5.1 The Tottenham Riots -- 5.1.1 The Shooting Incident -- 5.1.2 The Beginning of Protest -- 5.1.3 The Protest Turns Violent -- 5.1.4 Riots Spread Across Tottenham -- 5.1.5 Riots Spread to Woodgreen -- 5.2 The London Riots -- 5.2.1 Riots Spread to Enfield -- 5.2.2 Riots Spread to Walthamstow, Westfield and Edmonton -- 6 Machine Learning Model -- 6.1 Data Labelling -- 6.2 Feature Extraction -- 6.3 Data Preparation -- 6.4 Feature Selection -- 6.5 Experiments and Results -- 6.5.1 Naive Bayes -- 6.5.2 SVM -- 6.5.3 J48 -- 6.5.4 RandomForest -- 6.6 Model Selection -- 7 Discussion -- 8 Conclusion and Future Work -- References -- 8 Discovery of Structural and Temporal Patterns in MOOC Discussion Forums -- 1 Introduction -- 2 Background -- 2.1 Analysis of Discussion Forums -- 2.2 Identification of Roles in Communication Networks -- 2.2.1 Blockmodelling -- 2.2.2 Tensor Decomposition for Role Modelling -- 2.2.3 Estimating the Number of Clusters -- 3 Network Extraction from Forum Posts -- 3.1 Forum Post Classification -- 3.2 Network Extraction -- 4 Individual Development: Behavioural Roles over Time -- 4.1 Definition of Inreach and Outreach -- 4.2 In- and Outreach over Time: Identification of Characteristic Actor Trajectories -- 5 Macro-Structure of Evolving Knowledge Exchange Networks
Abstract:
5.1 Dynamic Blockmodelling -- 5.2 Role Modelling Based on Tensor Decomposition -- 5.3 Formal Evaluation -- 5.3.1 Fitting an Ideal Regular Block Structure -- 5.3.2 Density Patterns -- 5.3.3 Assessment of Methods -- 6 Applications -- 6.1 Trajectories of Behavioural Roles -- 6.2 Macro-Structures of Knowledge Exchange -- 7 Conclusion and Further Work -- References -- 9 Diffusion Process in a Multi-Dimension Networks: Generating, Modelling, and Simulation -- 1 Introduction -- 2 Preliminaries -- 3 Background and Related Work -- 4 Materials and Methods -- 4.1 Human Terrain -- 4.2 The Three Primary Dimensions: Family, Friends, and Neighbors -- 4.3 The Partisan Association Dimensions -- 4.4 The War Time Dimensions -- 4.5 Formalization of Human Behavior -- 4.5.1 DEVS Formalism -- 4.5.2 Specification of Message Processing by the Receiver -- 5 Experiment and Results -- 5.1 Social Network Measures -- 5.2 Using MSN in Simulation -- 5.2.1 Experiment -- 6 Conclusion and Perspectives -- References
URL:
Volltext
(lizenzpflichtig)
Permalink